DIREC TIONS IN DE VELOPMENT Human Development Minds and Behaviors at Work Boosting Socioemotional Skills for Latin America’s Workforce Wendy Cunningham, Pablo Acosta, and Noël Muller Minds and Behaviors at Work Direc tions in De velopment Human Development Minds and Behaviors at Work Boosting Socioemotional Skills for Latin America’s Workforce Wendy Cunningham, Pablo Acosta, and Noël Muller © 2016 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 19 18 17 16 This work is a product of the staff of The World Bank with external contributions. The findings, interpreta- tions, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Cunningham, Wendy, Pablo Acosta, and Noël Muller. 2016. Minds and Behaviors at Work: Boosting Socioemotional Skills for Latin America’s Workforce. Directions in Development. Washington, DC: World Bank. doi: 10.1596/978-1-4648-0884-5. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third- ­ party–owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights​ @­worldbank.org. ISBN (paper): 978-1-4648-0884-5 ISBN (electronic): 978-1-4648-0885-2 DOI: 10.1596/978-1-4648-0884-5 Cover photo: © Getty Images/DrAfter123. Used with permission. Further permission required for reuse. Cover design: Debra Naylor, Naylor Design, Inc. Library of Congress Cataloging-in-Publication Data has been requested. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Contents Abstract xi Acknowledgments xiii About the Authors xv Executive Summary xvii Abbreviations xxvii Chapter 1 Introduction 1 Objectives and Value Added of the Study 5 Definitions of Skills and Data for Skills Measurement 7 Notes 10 References 11 Chapter 2 Cognitive and Socioemotional Skills Profile of the Latin American Workforce 15 Mapping the Distribution of Cognitive Skills 15 Mapping the Distribution of Socioemotional Skills 22 Note 25 References 25 Chapter 3 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 27 Cognitive and Socioemotional Skills Are Correlated with Labor Earnings 28 Socioemotional Skills Are Correlated with Employment and Productive Activity 30 Both Types of Skills Are Correlated with Job Type 32 Both Skills Types Positively Correlate with Tertiary School Attendance 37 Interpretation of Cross-Country Variations 38 Notes 40 References 41 Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   v   vi Contents Chapter 4 Policy and Programming for Socioemotional Skill Development 43 Translating Research Findings into Policy-Relevant Concepts 44 Socioemotional Learning through the Life Cycle 46 Cognitive and Socioemotional Learning Processes 55 Notes 60 References 60 Chapter 5 Conclusions 65 Both Cognitive and Socioemotional Skills Are Associated with Labor Market and Tertiary Education Outcomes 65 Instruction in Both Cognitive and Socioemotional Skills Is Both Possible and Necessary to Better Prepare Latin American Workers for the Labor Market 66 Research Is Needed to Guide Policy Design 67 References 68 Appendix A Abstracts of Background Papers 69 Appendix B Methodologies Used in This Study 73 Appendix C Summary of Associations between Measures of Skills and Labor Market and Tertiary Education Outcomes in Bolivia, Colombia, El Salvador, and Peru 77 Appendix D Regression Results from Country Studies 85 Appendix E Cross-Country Variations in Associations between Skills Dimensions and Labor Market and Tertiary Education Outcomes 97 Appendix F Inventory of Promising Interventions to Foster Socioemotional Skills 99 Boxes 1.1 Changes in the Skill Content of Occupations in Latin America 2 1.2 Measures of Socioemotional Skills Using the “Big Five” Classification 8 1.3 How Data Were Collected for This Study 10 2.1 How Do Cognitive Skills of Future Labor-Market Entrants in Latin America Compare with Their Peers from Outside the Region? Results from the PISA 20 Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Contents vii 4.1 Promoting Self-Regulated Learning in Young Children through the Tools of the Mind Program 48 4.2 Developing Teachers’ Socioemotional Skills through Peru’s Escuela Amiga Program 49 4.3 Improving the School Climate through the School-Wide Positive Behavior Support Model 51 4.4 Incorporating Socioemotional Skills into the Teaching of Other Subjects: Facing History and Ourselves 52 4.5 Using a School Curriculum: The Incredible Years Program 53 4.6 Reaching At-Risk Adolescents through Colombia’s Fútbol con Corazón Out-of-School Program 54 4.7 Fostering Active Learning in Preschool 57 4.8 Active Learning through Colombia’s Escuela Nueva Approach 58 Figures ES.1 Adult Reading Proficiency Levels in Selected Countries, 2012–13 xviii ES.2 Skills Most Valued by World Employers, 2010s xix ES.3 Framework for Cognitive and Socioemotional Skills xix ES.4 Combined Effect of Cognitive and Socioemotional Skills on Labor Outcomes in Colombia and Peru xxi B1.1.1 Intensity of Use of Manual, Routine Cognitive, Analytical, and Interpersonal Skills in Brazil, Chile, Costa Rica, and the United States, 1980–2009 3 1.1 Skills Most Valued by World Employers, 2010s 4 1.2 Framework for Cognitive and Socioemotional Skills 7 2.1 Distribution of Test Scores of Cognitive Skills in Bolivia, Colombia, and Peru 16 2.2 Distribution of Reading Proficiency Scores in Bolivia and Colombia, by Educational Level 17 2.3 Correlation between Adult Reading Proficiency Scores and Per Capita Income in Selected Countries, 2012 18 2.4 Adult Reading Proficiency Levels in Bolivia, Colombia, and Selected Other Countries, 2012 20 B2.1.1 PISA Math Scores in Latin American Countries and OECD, 2000–12 21 2.5 Distribution of Resilience among Men and Women in Bolivia, Colombia, El Salvador, and Peru 23 2.6 Distribution of Hostile Attribution Bias in Bolivia, Colombia, and El Salvador and Cooperation in Peru, by Age Group 24 2.7 Distribution of Openness to Experience in Bolivia, Colombia, El Salvador, and Peru, by Educational Level 24 3.1 Correlation between Labor Earnings and Cognitive and Socioemotional Skills in Bolivia, Colombia, and Peru 31 Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 viii Contents 3.2 Correlation between Employment and Cognitive and Socioemotional Skills in Peru 33 3.3 Correlation between Any Productive Activity (Working, Looking for Job, or Studying) and Cognitive and Socioemotional Skills in Colombia 34 3.4 Correlation between Formal Employment and Cognitive and Socioemotional Skills in Bolivia, Colombia, and Peru 35 3.5 Correlation between Tertiary Education Attendance and Cognitive and Socioemotional Skills in Bolivia and Colombia 38 Tables ES.1 Strength of Correlations between Adults’ Skills and Labor Market and Education Outcomes in Selected Countries in Latin America and the OECD, circa 2012 xx ES.2 Optimal Stages of Development of Socioemotional Skills xxiii ES.3 Examples of Interventions Fostering Socioemotional Skills at School xxiv 1.1 Definitions of Skill Measures 8 B1.2.1 Interpretation of Low and High Scores of Big Five Personality Traits 9 2.1 Description of Reading Proficiency Levels in the PIAAC and STEP Surveys 19 3.1 Skills Correlated with Labor Earnings in Bolivia, Colombia, El Salvador, or Peru 28 3.2 Skills Correlated with Employment in Bolivia, Colombia, El Salvador, or Peru 32 3.3 Skills Correlated with Formal Employment in Bolivia, Colombia, El Salvador, or Peru 34 3.4 Skills Correlated with Wage Employment in Bolivia, Colombia, El Salvador, or Peru 36 3.5 Skills Correlated with Tertiary Education Attendance in Bolivia, Colombia, or El Salvador 37 4.1 Definitions of PRACTICE Skills 44 4.2 Optimal Stages of Development of PRACTICE Skills 45 C.1 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in Bolivia, 2012 78 C.2 Summary of Associations between Aggregated Measures of Skills and Labor and Tertiary Education Outcomes in Bolivia, 2012 79 C.3 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in Colombia, 2012 80 Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Contents ix C.4 Summary of Associations between Aggregated Measures of Skills and Labor and Tertiary Education Outcomes in Colombia, 2012 81 C.5 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in El Salvador, 2013 82 C.6 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in Peru, 2010 83 C.7 Summary of Associations between Aggregated Measures of Skills and Labor and Tertiary Education Outcomes in Peru, 2010 84 D.1 Structural Estimates of Conditional Correlations between Labor Market and Tertiary Education Outcomes and Latent Skills in Bolivia, 2012 85 D.2 Conditional Correlations between Labor Earnings, Formality, and Occupational Status and Measures of Skills and Schooling in Bolivia, 2012 86 D.3 Conditional Correlations between Employment, Activity, and Educational Trajectory and Measures of Skills and Schooling in Bolivia, 2012 87 D.4 Structural Estimates of Conditional Correlations between Labor Market and Tertiary Education Outcomes and Latent Skills in Colombia, 2012 88 D.5 Conditional Correlations between Labor Earnings, Formality, and Occupational Status and Measures of Skills and Schooling in Colombia, 2012 88 D.6 Conditional Correlations between Employment, Activity, and Educational Trajectory and Measures of Skills and Schooling in Colombia, 2012 89 D.7 Conditional Correlations between Labor Earnings, Formality, and Occupational Status and Measures of Skills and Schooling in El Salvador, 2013 91 D.8 Conditional Correlations between Employment, Activity, and Educational Trajectory with Measures of Skills and Schooling in El Salvador, 2013 92 D.9 Structural Estimates of Conditional Correlations between Labor Market Outcomes and Latent Skills Factors in Peru, 2010 93 D.10 Conditional Correlations between Labor Outcomes and Measures of Skills in Peru, 2010 94 E.1 Cross-Country Variations in Associations between Skills Dimensions and Labor Market and Tertiary Education Outcomes in Bolivia, Colombia, El Salvador, and Peru 98 F.1 Promising Programs That Foster Socioemotional Skills, by Target Age Group 99 Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 x Contents F.2 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) 101 F.3 Description of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) 102 F.4 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) 104 F.5 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) 106 F.6 Description of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) 108 F.7 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) 110 F.8 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) 112 F.9 Description of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) 113 F.10 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) 115 F.11 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) 117 F.12 Description of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) 118 F.13 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) 120 Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Abstract Latin America has shown impressive growth in educational attainment over the past two decades—but that education has failed to yield the expected benefits. A mounting body of research and policy debates suggests that the quantity of education is not an adequate metric of human capital acquisition. Rather, indi- viduals’ skills—what people actually know and can do—should stand as policy targets and be fostered across the life cycle. Evidence from around the world suggests that employers require both cognitive and socioemotional skills and that both types of skills are associated with a range of positive employment and edu- cational attainment outcomes. Minds and Behaviors at Work: Boosting Socioemotional Skills for Latin America’s Workforce synthetizes original empirical research on the role of cognitive and socioemotional skills in shaping adults’ labor market outcomes in Bolivia, Colombia, El Salvador, and Peru. This work is put in perspective with insights from similar studies in other Latin American countries and high-income countries. The findings show that cognitive skills matter for reaping labor market ­ gains in terms of higher wages and job formality in Latin America but so do socioemotional skills. Moreover, socioemotional skills seem to have a particularly strong effect on labor force participation and tertiary education attendance as a platform to build knowledge. Minds and Behaviors at Work also presents a policy framework for developing skills by providing insights from developmental psychology about when people are neurobiologically, socioemotionally, and ­ situationally ready to develop socioemotional skills and provides examples of ­ interventions that combine socioemotional learning and cognitive development. This book will be of importance to policy makers, researchers, and anyone else interested in human development, in Latin America and beyond. In particular, this book will be most valuable for the curious minds wondering how our mental abilities and behaviors shape our education and employment trajectories, and how to foster these abilities along our lives. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   xi   Acknowledgments “No one can whistle a symphony,” said Halford E. Luccock, an American Methodist minister. “It takes a whole orchestra to play it.” This volume is no exception and would not have been possible without the contributions of a num- ber of colleagues. This volume builds on a series of background papers prepared by a core research team and other contributors. Wendy Cunningham, Pablo Acosta, and Noël Muller led the World Bank regional study and prepared this volume (unless stated otherwise, contributors were World Bank staff at the time of publication). In addition to the main authors, the core research team included Juan D. Barón, Ana María Oviedo, Mónica Parra Torrado (National Department of Planning of Colombia), and Miguel Sarzosa (Purdue University, USA). Several other people also made valuable contributions, which are gratefully acknowledged. Nancy Guerra (University of Delaware, USA) and Kathryn Modecki (Murdoch University, Australia) cowrote the background paper on the PRACTICE taxonomy of socioemotional skills. Paula Villaseñor (Undersecretariat of Upper-Secondary Education of Mexico) drew policy directions in the background paper on employers’ demand for skills. Sergio Urzúa (University of Maryland, USA) provided suggestions on and meticu- lously reviewed the early development stage of the study methodology. Alexandría Valerio and María Laura Sanchez Puerta coordinated the World Bank’s STEP survey collection and provided technical assistance with the data. Natalia Millán contributed to the data collection of Colombia’s STEP Household survey and preliminary data analysis. José Mola (Universidad Tecnológica de Bolívar, Colombia) provided research assistance for the back- ground paper on skills and labor outcomes in Bolivia. At the outset and along the way, the team greatly benefited from interactions with and advice from colleagues. The team is especially grateful to Margaret Grosh for her continuous support, guidance, and careful review of all aspects of the study. Omar Arias, Christian Bodewig, Barbara Bruns, María Marta Ferreyra, Daniel Lederman, Julián Messina Granovsky (Inter-American Development Bank), Cem Mete, and Alexandría Valerio provided invaluable feedback and sug- gestions at various stages. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   xiii   xiv Acknowledgments Many other colleagues contributed ideas and feedback. They include Helen Abadzi (University of Texas at Arlington, USA), Rita Almeida, Stephen Close, Mariana Escalante, Rafael De Hoyos, David Evans, Katia Herrera, Inès Kudo, Koji Miyamoto (OECD), Sophie Naudeau, Reema Nayar, Mansoora Rashid, Jamele Rigolini, Alberto Rodríguez, Jan Rutkowski, Venkatesh Sundararaman, and Renos Vakis. The team is also grateful to Arup Banerji and Jorge Familiar Calderón for their oversight of the study and championing of this work. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 About the Authors Wendy Cunningham, a U.S. national, is a lead economist in the World Bank’s Social Protection and Labor Practice, where she works with client countries to develop and implement policies and programs to improve labor market access and outcomes, particularly for more vulnerable populations. Her research focuses on measuring and programming for skills for the labor market, with a focus on socioemotional skills. She has published on labor markets, informal employment, gender, and youth development. Before joining the Social Protection and Labor Practice, she was the World Bank’s program leader for human development and poverty in Mexico and Colombia and the coordinator of the World Bank’s program on child and youth development. She holds a Ph.D. in economics from ­ the University of Illinois Urbana-Champaign. Pablo Acosta, an Argentinean national, is a senior economist in the World Bank’s Social Protection and Labor Practice and a research fellow at the Institute for the Study of Labor (IZA), where he leads research and works with client countries to implement social protection and employment ­ programs. His main areas of research are labor economics, migration, skills, and international trade. Before joining the World Bank, he worked for the Development Bank of Latin America (CAF) in Venezuela, the Ministry of Economy in Argentina, and the Foundation for Latin American Economic Research (FIEL) in Argentina. He holds a Ph.D. in economics from the University of Illinois at Urbana-Champaign. Noël Muller, a French national, is a consultant economist to the World Bank’s Social Protection and Labor Practice, where he conducts empirical research and supports teams in Latin America and Ukraine in advising client countries on the development of social and employment programs. His research focuses on skills development, the role of skills in the labor market, employment policies, and the constraints faced by the jobless and vulnerable workers. Previously, he has worked on informal employment in Latin America and on social cohesion poli- cies in Vietnam at the Organisation for Economic Co-operation and Development (OECD) Development Centre. He holds a master’s degree in international and development economics from the University Paris-Dauphine and a bachelor in economics from the University Paris I Panthéon-Sorbonne. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   xv   Executive Summary Parents, teachers, and economic theory tell us that if we study hard, we will get good jobs, earn high salaries, and achieve professional success. Like the rest of the world, Latin American countries subscribe to this assumption. Countries of the region witnessed tremendous progress in education attainment over the past two decades. Between 1990 and 2010, the proportion of people entering the labor force (age 20–24) who had completed secondary education increased from 35 percent to 55 percent, and the average years of schooling of the labor force increased from 8.2 in 1990 to 10.2 in 2010 (Barro and Lee 2013; Bruns and Luque 2014). But recent evidence suggests that more schooling may not deliver the benefits promised. The rewards to acquiring higher levels of education actually declined in most Latin American countries over the past two decades, and workers in the region did not substantially improve their productivity (Pagés 2010; Aedo and Walker 2012; Gasparini and others 2011). Employers around the world, includ- ing in Latin America, lament the lack of adequate skills of current and prospec- tive employees. Are Skills Deficits the Problem? A tweak to the traditional advice may be necessary: Perhaps it is more and better skills, rather than more education, that matter (Hanushek and Woessmann 2008; Hanushek 2015). New data from Bolivia and Colombia suggest that years of education only partially reflect what people can actually do. For example, half of Bolivian tertiary school graduates have the same level of reading proficiency as do half of ­ those who only graduated from secondary school. International assessments of adults’ skills show that, despite the surge in schooling, Latin America’s labor force is lagging. The ability of adults in Bolivia and Colombia to understand and reflect on written texts remains lower than that of their peers in countries at similar levels of economic development: A third of Colombian adults display only a basic level of proficiency (that is, they can per- form reading tasks only from short pieces with no or little competing informa- tion) (figure ES.1), compared with only 15 percent of adults in Ukraine (where per capita GDP is 30 percent lower than that of Colombia) and member coun- tries of the Organisation for Economic Co-operation and Development (OECD). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   xvii   xviii Executive Summary Figure ES.1  Adult Reading Proficiency Levels in Selected Countries, 2012–13 100 90 80 70 Percent of adults 60 50 40 30 20 10 0 Ghana Kenya Bolivia Colombia Vietnam Georgia Armenia Ukraine OECD-23 Level 1 and below Level 2 Level 3 Levels 4 and 5 Sources: World Bank’s STEP Household Surveys (2012–13) and OECD’s Program of International Assessment of Adult Skills (PIAAC) data (2013). Note: Each score level corresponds to a set of reading and analytical abilities measured by the test. Level 1 and below is the lowest level; level 5 is the highest level. Higher levels of proficiency indicate the ability to perform more complex tasks on longer and harder written materials. See ETS (2014) and table 2.1 for details of the reading proficiency levels. The concept of skills itself may need to be reexamined in the context of the rapidly changing nature of work. Jobs are changing, the tasks required by those jobs are evolving, and workers are frequently changing jobs. Skills therefore no longer refer only to job-specific knowledge but rather to a set of attributes needed to navigate across life situations and jobs that are increasingly complex in nature. In Brazil, Chile, Costa Rica, and the United States, the number of jobs that require predominantly routine manual skills has decreased since the 1980s, while the number of jobs using nonroutine analytical skills has grown (Aedo and others 2013). Around the world, employers report greatly valuing the skills needed for these emerging jobs—not only basic academic knowledge and techni- cal skills but also advanced cognitive skills (including critical thinking, efficiency, and leadership) and a set of behaviors, attitudes, personality traits, and values (referred to collectively as socioemotional skills), including honesty, teamwork, punctuality, and responsibility, among others (figure ES.2). Policy makers need to think about skills more broadly and recognize their multidimensionality. They commonly equate skills with cognitive skills (intelli- gence or the ability to perform mental tasks) (figure ES.3).1 But at least as impor- tant for one’s success are socioemotional skills.2 People use socioemotional skills, associated with achieving goals, managing emotions, and working with others, to transform cognitive skills into outputs (figure ES.3). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Executive Summary xix Figure ES.2 Skills Most Valued by World Employers, 2010s 51% 29% Advanced cognitive Communication Socioemotional Problem solving Critical thinking Work ethic Teamwork Honesty Punctuality Responsibility 4% Basic cognitive Creativity 16% Technical Technical knowledge Computing Source: Cunningham and Villaseñor 2016. Figure ES.3  Framework for Cognitive and Socioemotional Skills Achieving Basic goals cognitive Facets linked to Basic academic organization and knowledge, such perseverance as literacy or numeracy Socioemotional Cognitive Managing Behaviors, personality traits, and attitudes that enable emotions Mental abilities to engage in comprehension and reasoning individuals to navigate personal Facets related to Advanced and acquire knowledge and social situations effectively confidence and cognitive dealing with stress and emotions Complex thinking, such as critical Working thinking or with others problem-solving Facets reflecting how people interact with others Sources: Almlund and others 2011; SEMS 2014; World Bank 2014; and OECD 2015. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 xx Executive Summary Cognitive and Socioemotional Skills Influence Labor Market Outcomes Both cognitive and socioemotional skills play important roles in shaping employ- ment and tertiary education outcomes in Latin America and high-income coun- tries. A myriad of studies from the United States and other high-income countries shows that for a given level of education and other characteristics, children with higher levels of either type of skills become more successful students and work- ers (see for example Heckman, Stixrud, and Urzúa 2006; Nyhus and Pons 2005; Mueller and Plug 2006; Carneiro, Crawford, and Goodman 2007; Lindqvist and Vestman 2011; OECD 2015). Similar findings have been found for young adults (25–30) in Argentina and Chile (Bassi and others 2012). New evidence produced for this study for Bolivia, Colombia, El Salvador, and Peru confirms that adults with higher levels of cognitive or socioemo- tional skills are more likely to enjoy better labor market outcomes and pursue tertiary education, as compared to those with lower levels of skills:3 Those with higher levels of cognitive or socioemotional skills scores earn higher wages and are more likely to attend a tertiary education institution than simi- lar people with lower scores (table ES.1). Socioemotional skills are highly correlated with being employed in Latin America, while cognitive skills are particularly important for employment in the OECD.4 Cognitive skills are strongly correlated with better jobs—formal job and being in a high-skilled occupation—in Latin America while socioemotional skills play a smaller, yet still significant, role. Specific subfacets of socioemotional skills are correlated with different labor market and educational outcomes. Across the four Latin American countries studied Table ES.1 Strength of Correlations between Adults’ Skills and Labor Market and Education Outcomes in Selected Countries in Latin America and the OECD, circa 2012 Bolivia, Colombia, El Salvador, and Peru (age 15–64) OECD countries (age 25–30) Outcome Cognitive skills Socioemotional skills Cognitive skills Socioemotional skills Wages High Medium High Low-medium Formal job High Low-medium n.a. n.a. High-skilled High Low-medium n.a. n.a. occupation Wage workers Medium Medium n.a. n.a. Employed Low Medium High Low-medium Active in the labor Medium High n.a. n.a. market or studying Tertiary education High High High Medium attendance Sources: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010); OECD data are from OECD 2015. Note: n.a. = Not applicable (outcome not studied). Thresholds are based on regression tables of skills on outcomes (see appendixes C and D). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Executive Summary xxi (Bolivia, Colombia, El Salvador, and Peru), all three dimensions of socioemotional skills—achieving goals, managing emotions, and working with others—are correlated with higher wages. Specific underlying skills include openness to experience (appre- ciation for learning and a variety of experiences); conscientiousness (being orga- nized, responsible, perseverant, and hardworking); agreeableness (being cooperative and unselfish); and resilience (being calm, containing emotional reactions, and mak- ing decisions carefully). The Combined Effect of Cognitive and Socioemotional Skills Is Stronger than the Individual Effects The two types of skills interact in a positive way; their combined effect is greater than the sum of their individual effects. In Colombia, for example, almost all people with the highest levels of both cognitive and socioemotional skills are engaged in a productive activity while people with highest levels of only one type of skills are less so (panel a of figure ES.4). In Peru the highest wage earners are people with the highest cognitive skills and a range of strong socioemotional skills (panel b of figure ES.4). These highly skilled workers earn three times the hourly wage as those with the highest level of cognitive skills and the lowest level of socioemotional skills. Socioemotional Skills Differ Only Slightly across Sex and Age Groups Given the importance of a range of skills in labor market success, it is comforting to observe that the skills levels are similar across demographic characteristics. Only slight differences between men and women are observed in socioemotional skills related to achieving goals (conscientiousness and grit, defined as Figure ES.4 Combined Effect of Cognitive and Socioemotional Skills on Labor Outcomes in Colombia and Peru a. Association between being active (employed or b. Association between labor earnings and studying) and skills in Colombia skills in Peru Probability of working, looking for job, or studying (percent) Expected hourly income 96 10 from main job (PEN) 94 9 92 8 90 7 88 6 86 5 84 4 82 3 80 2 78 10 10 De 9 10 cile 8 De 9 8 9 10 (co of s 7 8 9 cile 7 8 nsi 6 7 (ac o 6 7 ste tabili 5 4 5 6 skil l s hie f soc 5 6 ills and cy in ty pe n ive age) v 4 5 e sk ls) 3 4 gnit oth ing ioe 4 nitiv il soc mo rson 3 f co ry, lang u ers m , m goals otio 3 2 2 3 o f cog uage sk ial tiva alit 2 2 i l e o iles ng inte tio 1 o ana , w n 1 Dec g and la y tr rac n, m aits Dec , mem gin orki al sk tion oo t h ge ng i mo wit lls (rea din s) d, (ma tion h s) Sources: Colombia: STEP Household Survey (2012); Peru: ENHAB (2010). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 xxii Executive Summary perseverance and passion for long-term goals) and managing emotions (decision making and hostility toward others). The only notable difference is that in all four countries, men tend to be more resilient than women. In all four countries, youth are less likely than adults to display behaviors associated with working with others (extroversion and agreeableness) and achieving goals (persevering). In Bolivia, Colombia, and El Salvador, they display some greater skills in managing emotions: specifically they are less likely than adults to perceive hostility in others. Cross-Country Differences Could Be Explained by Cultural Factors, Labor Market Structure, or Survey Discrepancies Specific skills associated with a given outcome vary across countries, especially for socioemotional skills. Although the determinants of these variations cannot be unambiguously untangled, some factors may explain cross-country patterns. Distinct cultural contexts may affect both the ways certain behaviors are rewarded in the work- place and the manner in which participants respond to survey questions on socioemotional skills. The country-specific structure of employment and differ- ential rewards of some skills by occupations may lead to heterogeneity in returns to skills. Given the slight differences in the number of questions about socioemo- tional skills and the different cognitive skills measured across surveys, differential returns may reflect imperfect comparability across country surveys. Measurement error could also be at play. Public Interventions Can Foster Socioemotional Skills in a Variety of Settings and Must Target Optimal Development Periods The formation of skills is a cumulative process. Because it is affected by the envi- ronment and investments, programs for developing socioemotional skills are best implemented at particular times in the life cycle. Three factors need to be taken in to consideration in designing socioemotional development programs. First, the developmental age of the child, in terms of psychological, neurobiological, and social readiness to learn and practice con- cepts, is key. Just as very young children are not ready to read, they are not ready to develop social problem-solving skills until they have the psychosocial wiring needed for empathy. Second, different actors can help develop skills at different ages and in various contexts. For very young children, parents and caregivers play the main role. Among adolescents, peers and school play the dominant role. Families, higher education institutions, and the workplace shape the skill formation of adults. Third, skills should be identifiable and malleable to be successfully taught. The personality traits analyzed in this study are a combination of a set of underly- ing skills and thus are a challenge to isolate, measure, and change. However, the socioemotional skills that employers identify as important are measurable and Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Executive Summary xxiii malleable to policy interventions. Thus, we can map the socioemotional skills analyzed in this study to the behaviors and attitudes identified by employers, leading us to a policy framework. The PRACTICE taxonomy groups more than 140 skills that employers in 28 countries identify as important for labor market success.5 Table ES.2 maps the three sets of socioemotional skills analyzed in this study to the PRACTICE taxonomy and indicates the appropriate stage of development for acquiring them. Much work has been done on teaching socioemotional skills to very young (0–5) children; less is known about teaching such skills in middle childhood, adolescence, and early adulthood. However, there is growing evidence that these skills can be taught in the classroom environment. They may emerge from teach- ers modeling behaviors that they themselves have learned, a positive and safe school climate that fosters and rewards positive socioemotional behaviors, teach- ing methods that use socioemotional skills in the learning process, or specific curricula to teach these skills. Table ES.3 provides some examples. For both in- and out-of-school adolescents, after-school activities have shown to be beneficial (Tierney and Baldwin Grossman 2000; Boys & Girls Clubs of America 2004), as are programs that blend job training and socioemotional skills training (Ibarrarán and others 2014; Vezza and others 2014).6 The international evidence shows that these skills can be taught via existing institutions. The main challenge is to organize the actors and the pedagogical pathways to do so, thereby setting up Latin American workers for greater pro- ductivity and success. Table ES.2 Optimal Stages of Development of Socioemotional Skills Stage of development and key actors 0–5 6–11 12–18 19–29 Dimension of PRACTICE (school, family, socioemotional skills taxonomy (parents) (parents, school) (school, peers) workplace) Achieving goals Achievement Optimal Reinforcement motivation Ethics Foundational Optimal Optimal Initiative Optimal Optimal Optimal Optimal Problem solving Foundational Optimal Optimal Reinforcement Working with others Teamwork Optimal Optimal Reinforcement Managing emotions Confidence Foundational Optimal Optimal Reinforcement Control Optimal Optimal Optimal Reinforcement Resilience Optimal Optimal Reinforcement Source: Guerra, Modecki, and Cunningham 2014. Note: PRACTICE is a taxonomy of socioemotional skills that summarizes a long list of socioemotional skills that employers recognize as very important in workers. The acronym stands for Problem solving, Resilience, (Achievement) Motivation, Control, Teamwork, Initiative, Confidence, and Ethics. “Foundational” refers to the initial skill-building process that will predominately occur in a following period. “Optimal” refers to periods of maximum sensitivity when it is easiest for individuals to acquire specific skills. “Reinforcement” means that intense practice is necessary to master the skill. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 xxiv Executive Summary Table ES.3 Examples of Interventions Fostering Socioemotional Skills at School Intervention Example Country Developing teachers’ socioemotional Escuela Amiga (Friendly School) program Peru skills so that they can model them (Paredes 2014) in the classroom Strengthening the school climate, to School-Wide Positive Behavior Support Australia, Canada, provide a safe place for practicing model (Bradshaw, Mitchell, and Leaf Mexico, Norway, positive social behaviors 2012) United States Creating and implementing a Incredible Years program(Webster- United States socioemotional curriculum to Stratton, Reid, and Stoolmiller 2008) explicitly teach and reinforce behaviors Incorporating socioemotional skills, Facing History, Facing Ourselves model 110 countries such as teamwork and problem- (Barr 2010) solving, into teaching and presentational methods Sources: CASEL 2013, 2015. Notes 1. Cognitive skills can be grouped into two categories: basic cognitive skills (basic aca- demic learning, including memory, numeracy, literacy, and evaluation of written infor- mation) and advanced cognitive skills (more complex mental tasks, such as critical thinking, advanced problem solving, and time management). Technical skills—the specific knowledge needed to carry out a task—can be thought of as a subset of cogni- tive skills (Almlund and others 2011). They are not studied specifically in this study. 2. Although these skills involve some level of cognition, economists refer to them as noncognitive skills, in order to differentiate them from academic or learning skills. Traits are characteristics or patterns of thought and action that are relatively stable across the life cycle. Behaviors are performance in response to stimulation. Attitudes comprise beliefs and values that guide skill formation and behavior. 3. The data used for this study do not permit unambiguous causal links to be established between skills and labor outcomes. Most studies based on longitudinal data in high- income countries robustly establish causal links between cognitive and socioemotional skills on the one hand and labor and education outcomes on the other (see for example Heckman, Stixrud, and Urzúa 2006; Nyhus and Pons 2005; Mueller and Plug 2006; Carneiro, Crawford, and Goodman 2007; Lindqvist and Vestman 2011; OECD 2015). Although high-income countries are different contexts, the studies based on data from these countries suggests that skills could lead to better labor market out- comes in Latin America, as well. 4. In particular, more conscientious adults (goal-oriented and self-disciplined) and those showing higher grit (perseverance and passion for long-term goals) are more likely to be employed in the four Latin American countries studied. 5. PRACTICE is a taxonomy of socioemotional skills that summarizes a long list of socioemotional skills that employers recognize as very important in workers. The acro- ­ nym stands for Problem solving, Resilience, (Achievement) Motivation, Control, Teamwork, Initiative, Confidence, and Ethics (Guerra, Modecki, and Cunningham 2014). 6. Guerra, Modecki, and Cunningham (2014) discuss evidence-based interventions for specific ages. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Executive Summary xxv References Aedo, C., J. Hentschel, J. Luque, and M. Moreno. 2013. “From Occupations to Embedded Skills: A Cross-Country Comparison.” Policy Research Working Paper 6560, World Bank, Washington, DC. Aedo, C., and I. Walker. 2012. Skills for the 21st Century in Latin America and the Caribbean. World Bank, Washington, DC. Almlund, M., A. L. Duckworth, J. J. Heckman, and T. Kautz. 2011. “Personality Psychology and Economics.” In Handbook of the Economics of Education, Vol. 4, edited by E. A. Hanushek, 1–181. Amsterdam: North Holland. Barr, D. J. 2010. Continuing a Tradition of Research on the Foundations of Democratic Education: The National Professional Development and Evaluation Project. Facing History and Ourselves National Foundation, Inc. Barro, R. J., and J. W. Lee. 2013. “A New Data Set of Educational Attainment in the World, 1950–2010.” Journal of Development Economics 104: 184–98. Bassi, M., M. Busso, S. Urzúa, and J. Vargas. 2012. Disconnected: Skills, Education and Employment in Latin America. Inter-American Development Bank, Washington, DC. Boys & Girls Clubs of America. 2004. Proven Results: A Compendium of Program Evaluations from Boys & Girls Clubs of America 1985–Present. Atlanta: Boys & Girls Clubs of America. Bradshaw, C. P., M. M. Mitchell, and P. J. Leaf. 2012. “Examining the Effects of Schoolwide Positive Behavioral Interventions and Supports on Student Outcomes: Results from a Randomized Controlled Effectiveness Trial in Elementary Schools.” Journal of Positive Behavior Interventions 12 (3): 133–48. Bruns, B., and J. Luque. 2014. Great Teachers: How to Raise Student Learning in Latin America and the Caribbean. World Bank, Washington, DC. Carneiro, P., C. Crawford, and A. Goodman. 2007. “The Impact of Early Cognitive and Noncognitive Skills on Later Outcomes.” CEE DP 92, Centre for the Economics of Education, London School of Economics, London. CASEL (Collaborative for Academic, Social, and Emotional Learning). 2013. Effective Social and Emotional Learning Programs: Preschool and Elementary School Edition. Chicago. ———. 2015. Effective Social and Emotional Learning Programs: Middle and High School Edition. Chicago. Cunningham, W., and P. Villaseñor. 2016. “Employer Voices, Employer Demands, and Implications for Public Skills Development Policy Connecting the Labor and Education Sectors.” World Bank Research Observer 31 (1): 102–34. ETS (Educational Testing Services). 2014. A Guide to Understanding the Literacy Assessment of the STEP Skills Measurement Survey. IEA-ETS Research Institute, Princeton, NJ. Gasparini, L., S. Galiani, G. Cruces, and P. Acosta. 2011. “Educational Upgrading and Returns to Skills in Latin America: Evidence from a Supply-Demand Framework, 1990–2010.” Policy Research Working Paper 5921, World Bank, Washington, DC. Guerra, N., K. Modecki, and W. Cunningham. 2014. “Social-Emotional Skills Development across the Life Span: PRACTICE.” Policy Research Working Paper 7123, World Bank, Washington, DC. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 xxvi Executive Summary Hanushek, E. A. 2015. “Why Standard Measures of Human Capital are Misleading.” KDI Journal of Economic Policy 37 (2): 22–39. Hanushek, E. A., and L. Woessmann. 2008. “The Role of Cognitive Skills in Economic Development.” Journal of Economic Literature 46 (3): 607–68. Heckman, J. J., J. Stixrud, and S. Urzúa. 2006. “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior.” Journal of Labor Economics 24 (3): 411–82. Ibarrarán, P., L. Ripani, B. Taboada, J. M. Villa, and B. Garcia. 2014. “Life Skills, Employability and Training for Disadvantaged Youth: Evidence from a Randomized Evaluation Design.” IZA Journal of Labor and Development 3: 1–24. Lindqvist, E., and R. Vestman. 2011. “The Labor Market Returns to Cognitive and Noncognitive Ability: Evidence from the Swedish Enlistment.” American Economic Journal: Applied Economics 3 (1): 101–28. Mueller, G., and E. J. S. Plug. 2006. “Estimating the Effect of Personality on Male and Female Earnings.” Industrial and Labor Relations Review 60 (1): 3–22. Nyhus, E. K., and E. Pons. 2005. “The Effects of Personality on Earnings.” Journal of Economic Psychology 26: 363–84. OECD (Organisation for Economic Co-operation and Development). 2015. Skills for Social Progress: The Power of Social and Emotional Skills. OECD Skills Studies. Paris: OECD Publishing. Pages, C. ed. 2010. The Age of Productivity: Transforming Economies from the Bottom Up. New York: Inter-American Development Bank. Paredes, G. 2014. “Diplomado en educación socioemocional para la convivencia escolar: propuesta y resultados.” [Graduate in Socioemotional Education for School ­ Conviviality: Proposal and Results]. Academic Board of Social Responsibility of Peru. http://vimeo.com/108231477. SEMS (Subsecretaría de Educación Media Superior de México [Undersecretariat of Upper Secondary Education of Mexico]). 2014. Programa Construye T 2014–2018: Fortalecer las capacidades de la escuela para promover el desarrollo integral de los jóvenes [Program “Build Yourself” 2014–2018: Strengthening School Capacities to Promote the Comprehensive Development of Young People]. Mexico: SEMS. www.construye-t.org​ .mx/resources/DocumentoConstruyeT.pdf. Tierney, J., and J. Baldwin Grossman. 2000. Making a Difference: An Impact Study of Big Brothers Big Sisters. Public/Private Ventures, Philadelphia. Vezza, E., B. García, G. Cruces, and J. Amendolaggine. 2014. Programa Juventud y Empleo: Informe de evaluación de impacto cohortes 2008–2009 [Youth and Employment Program: Report of the Impact Evaluation on 2008–09 cohorts], World Bank and the Ministry of Labor of the Dominican Republic, Washington, DC. Webster-Stratton, C., M. J. Reid, and M. Stoolmiller. 2008. “Preventing Conduct Problems and Improving School Readiness: Evaluation of the Incredible Years Teacher and Child Training Programs in High-Risk Schools.” Journal of Child Psychology and Psychiatry 49 (5): 471–88. World Bank. 2014. Toolkit for Socio-Emotional Learning “Paso a Paso” (Step by Step) for the Program “Escuela Amiga” (Friendly School) for Primary and Secondary Public Schools of Peru. Internal report, World Bank, Washington, DC. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Abbreviations Bs Bolivian Boliviano CAF Development Bank of Latin America CBT Cognitive Behavioral Therapy CIDAC  Centro de Investigación para el Desarrollo (Research Center for Development) Col$ Colombian Peso ENHAB Encuesta Nacional de Habilidades (National Skills Survey) ETS Educational Testing Services FIEL Foundation for Latin American Economic Research GDP Gross Domestic Product GED General Education Degree GNP Gross National Product ISCO International Standard Classification of Occupations IQ Intelligence Quotient IZA Institute for the Study of Labor KIPP Knowledge Is Power Program OECD Organisation for Economic Co-operation and Development OLS Ordinary Least Square O*NET Occupational Information Network PEN Peru Nuevo Sol PIAAC Program for the International Assessment of Adult Skills PISA Programme for International Student Assessment PRACTICE Taxonomy of socioemotional skills including (social) Problem solving, Resilience, Achievement Motivation, Control, Teamwork, Initiative, Confidence, and Ethics Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   xxvii   xxviii Abbreviations SEMS  Subsecretaría de Educación Media Superior de Mexicó (Undersecretariat of Upper Secondary Education of Mexico) STEP Skills Toward Employment and Productivity SWPBS School-Wide Positive Behavior Support All dollar amounts are U.S. dollars. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Chapter 1 Introduction Parents, teachers, and economic theory tell us that people who study hard get good jobs, earn high salaries, and achieve professional success. But recent evidence sug- gests that more schooling may not deliver the benefits promised. Latin American countries witnessed tremendous progress in education attainment over the past two decades. Between 1990 and 2010, the proportion of people entering the labor force (age 20–24) who had completed secondary education increased from 35 percent to 55 percent, and the average years of schooling of the labor force increased from 8.2 years in 1990 to 10.2 in 2010 (Barro and Lee 2013; Bruns and Luque 2014). Over the same period, returns to tertiary education actually declined in most Latin American countries (Gasparini and others 2011; Aedo and Walker 2012), and labor productivity did not improve significantly (Pagés 2010). Meanwhile, employers around the world, including in Latin America, lament the shortage of appropriate skills: A 13-country study reports that only 42 percent of employers believe that youth are prepared for the labor market (curiously, 72 percent of educators think they are [Mourshed, Farrell, and Barton 2012]). The increase in education attainment and the dissatisfaction of employers sug- gests that a tweak to the traditional advice may be necessary: Perhaps greater skills, rather than more education, improve labor market outcomes (Hanushek and Woessman 2008). This definitional change breaks with the practice of equat- ing years of schooling with skills acquired, but data support the revision. In 2012 only 33 percent of Brazilian 15-year-olds who completed ninth grade had acquired sufficient math skills to be able to solve basic problems; in Colombia and Peru only a quarter of students could do so (OECD 2013; Bruns and Luque 2014). The evidence in Latin America thus points to a serious mismatch between years of education and skills acquired. Many observers blame the low level of skills in Latin America on weak learn- ing outcomes (as evidenced in headlines of low scores on international bench- marks, such as the Programme for International Student Assessment [PISA]) as well as on documented deficiencies in teacher qualifications (Bruns and Luque 2014). The expansion of enrollment without sufficient investment in infrastruc- ture or human resource preparedness and weak regulation may partially explain Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   1   2 Introduction the regional mismatch between additional years of education and actual c ­ ognitive skills acquired (Levy and Schady 2013). The types of skills valued by the labor market are also changing. A multicoun- try study that included Brazil, Chile, and Costa Rica finds that the use of routine manual skills decreased over the past three decades while the use of nonroutine analytical skills increased (Aedo and others 2013). A similar trend has emerged in the United States (box 1.1). Box 1.1 Changes in the Skill Content of Occupations in Latin America Technology has changed dramatically over the past decades, and so have jobs. Pioneering research documents the shift in the United States away from occupations that require workers to execute a range of predicable tasks toward jobs requiring nonroutine, nonmanual tasks (Autor, Levy, and Murnane 2003; Acemoglu and Autor 2011). Skills required for nonroutine and nonmanual tasks include a mix of what this study categorizes as advanced cognitive and socioemotional skills. The U.S. government’s Occupational Information Network (O*NET) database provides detailed descriptions of task requirements by occupations. Each occupation is assigned a score that reflects the typical intensity of skill use for each category of skills. Information on job skill requirements can be coupled with survey data on the occupational structure of a country to yield national scores for skill categories. Following this methodology, Aedo and others (2013) examine the skill content of occupa- tions in 30 middle-income countries and the United States circa 2010, providing time series for a subset of countries. Six Latin American countries are included in the cross-sectional analysis (Brazil, Chile, Costa Rica, El Salvador, Mexico, Nicaragua, and Peru), and time series are provided for Brazil (1981–2009), Chile (1992–2009), and Costa Rica (2001–08). The cross-country analysis shows that the intensity of use of manual skills (both routine and nonroutine) is lower and the intensity of routine and nonroutine (analytical and interpersonal) cognitive skills is higher in countries with higher gross national product (GNP) per capita. Given their level of GNP per capita, all Latin American countries in the sample except Nicaragua have lower than average intensities of nonroutine cognitive analytical skills. Time series data for Brazil (30 years), Chile (20 years), and Costa Rica (10 years) show pat- terns similar to but less pronounced than those observed in the United States. In all three countries, the intensity of use of nonroutine skills (interpersonal and analytical) and routine cognitive skills increased monotonically over time while the intensity of nonroutine manual physical skills decreased, except in Costa Rica, where it remained stagnant (figure B1.1.1). This analysis has limitations, particularly because it applies the degree of skills use in the United States to middle-income countries, implicitly assuming that they have the same task requirements for each occupation. Despite technical adjustments in the computation, this assumption is very likely to cause an upward bias toward skills that are more prevalent in the United States—namely, advanced cognitive and socioemotional skills. box continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Introduction 3 Box 1.1  Changes in the Skill Content of Occupations in Latin America (continued) Figure B1.1.1 Intensity of Use of Manual, Routine Cognitive, Analytical, and Interpersonal Skills in Brazil, Chile, Costa Rica, and the United States, 1980–2009 a. Routine manual skills b. Routine cognitive skills 3.1 3.3 Skills intensity index relative to Skills intensity index relative to 3.0 3.2 3.1 2.9 3.0 2.8 2.9 1980–81 1980–81 2.7 2.8 2.6 2.7 2.6 2.5 2.5 2.4 2.4 2.3 2.3 1980–81 1990–92 2000–01 2008–09 1980–81 1990–92 2000–01 2008–09 c. Nonroutine analytical skills d. Nonroutine interpersonal skills 3.1 3.1 Skills intensity index relative to Skills intensity index relative to 3.0 3.0 2.9 2.9 2.8 2.8 1980–81 1980–81 2.7 2.7 2.6 2.6 2.5 2.5 2.4 2.4 2.3 2.3 1980–81 1990–92 2000–01 2008–09 1980–81 1990–92 2000–01 2008–09 e. Nonroutine manual physical skills 3.1 Skills intensity index relative to 3.0 2.9 2.8 1980–81 2.7 2.6 2.5 2.4 2.3 1980–81 1990–92 2000–01 2008–09 Brazil Chile Costa Rica United States Source: Aedo and others 2013. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 4 Introduction Employers seek a broader range of skills than reading, writing, and basic tech- nical skills; they look for workers who have also mastered values, behavioral, and thinking skills. Evidence from 27 studies reveals remarkable consistency in the skills demanded around the world (Cunningham and Villaseñor 2016).1 Although employers value all skill sets, demand is greater for socioemotional and advanced cognitive (complex thinking) skills than for basic cognitive (general knowledge) or technical skills (figure 1.1). More than three-quarters of the 27 studies cite a socioemotional skill as the most valued, and half identify a socioemotional skill among the top five preferred. Among the preferred socioemotional skills are work ethic, interpersonal skills, honesty, teamwork, attitude, integrity, punctuality, and responsibility. In addition, in nearly 30 percent of the studies, an advanced cognitive skill—primarily critical thinking, communication, or problem solving— is among the top five. These results are robust across different economy sizes, levels of development, sectors, export orientation, and occupations. Latin American employers reflect the same preferences as global employers. In a 2012 survey, employers from Argentina, Brazil, and Chile uniformly ranked socioemotional skills as most desirable, followed by cognitive skills and technical skills (Bassi and others 2012). A survey of Latin American executives ranked critical thinking, problem solving, and life skills as the top three skills they seek in new employees (Ogier 2009). Mexican employers specify teamwork, com- munications, and leadership as preferred skills for both managers and workers (CIDAC 2014). Peruvian employers specify teamwork and interpersonal skills (World Bank 2011). Employers in St. Kitts and Nevis cite honesty, work ethic, Figure 1.1 Skills Most Valued by World Employers, 2010s 51% 29% Advanced cognitive Communication Socioemotional Problem solving Critical thinking Work ethic Teamwork Honesty Punctuality Responsibility 4% Basic cognitive Creativity 16% Technical Technical knowledge Computing Source: Cunningham and Villaseñor 2016. Note: Results are based on meta-analysis of 27 studies. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Introduction 5 and problem solving (Blom and Hobbs 2008). All of these surveys included tech- nical and basic cognitive skills on the list of potential preferences for employers to choose from but employers clustered to the socioemotional and advanced cognitive skills.2 The results from Latin America and global studies are reflected in more exten- sive research from developed countries, where the market rewards a range of skills, not just years of education. A 13-country sample estimates that a one standard deviation increase in an individual’s literacy score (a cognitive skill) increases earnings by an average of 9.3 percent and that the effect of years of schooling on earnings falls by 30 percent after controlling for literacy scores (Hanushek and Zhang 2009). Numerous other studies also find that socioemo- tional skills strongly influence earnings and labor supply (Muller 2014), and some research finds that the impact of these skills on earnings exceeds the impact of cognitive skills (Heckman, Stixrud, and Urzúa 2006). The value placed on socioemotional skills may be related to a decline in jobs that require routine manual skills and the rise of nonroutine analytical jobs that require more com- plex skills (Aedo and others 2013). A range of socioemotional and basic cognitive skills are required to develop advanced skills. Schools are privileged places to foster cognitive skills, but socio- emotional skills also influence schooling decisions and educational outcomes (Almlund and others 2011). One study using longitudinal data for the United States estimates that socioemotional measures explain 12 percent of the varia- tion in educational attainment and cognitive ability measures explain 16 percent (Cunha, Heckman, and Schennach 2010). Just as accumulated levels of basic cognitive skills (such as math, reading, and attention skills) at a young age define children’s school readiness and predict their capacity to remain in school through the tertiary level (Duncan and others 2007), socioemotional skills also accumu- late and affect future education success. In a region where economic growth prospects promise to be far less robust in the near future than in past decades, improvements in education and employment outcomes may well be at the top of Latin American policy mak- ers’ priority lists, whether for productivity or equity reasons. One way to advance these goals is to equip children, adolescents, and adults with the attri- butes that will help them persevere and learn at school and look for, find, and hold good jobs. Objectives and Value Added of the Study This study has two main objectives. The first is to provide a deep-dive analysis of the role of socioemotional skills in labor market performance in Latin America. In addition to measuring the correlations, the study aims to assess the role of socioemotional skills relative to basic cognitive skills, to compare the Latin America trends to trends in member countries of the Organisation for Economic Co-operation and Development (OECD), and to generate more robust evidence than exists by taking advantage of richer data from a multicountry analysis and Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 6 Introduction using various methodologies to triangulate results. The second objective is to provide a framework that links the supply of and demand for socioemotional skills and to draw insights from the emerging evidence of policies/interventions on developing socioemotional skills. This study tests two hypotheses derived from the international literature. The first is that a broader range of skills than those currently taught are valuable for labor market success in Latin America. The second is that socioemotional skills can be taught through existing institutions in Latin American countries. This study summarizes the findings of all previous studies that directly mea- sure the role of socioemotional skills in labor market success in Latin America. It relies on four background studies that use newly collected household data for urban areas of Bolivia, Colombia, El Salvador, and Peru (see appendix A). It complements earlier work on the link between a set of cognitive and socio- emotional skills and labor market outcomes for youth in the urban areas of Argentina and Chile (Bassi and others 2012). Given the high levels of urban- ization of Latin American countries, the findings are likely to apply to most of the working-age population in the region.3 The literature documenting the role of skills in U.S. and Western European labor markets is also reported, to put the results for Latin American countries in a global perspective. This study makes five additional contributions. First, it explores the skills of the entire working-age population, not just youth, as in the only other comprehensive study of socioemotional skills and Latin America’s labor mar- kets (Bassi and others 2012). Second, it considers how combinations of skills affect labor market outcomes (Heckman, Stixrud, and Urzúa 2006). Third, it unpacks the effects of cognitive and socioemotional skills on labor market outcomes, in contrast to evaluations that consider only the package of skills (such as youth employment program evaluations). Fourth, it complements the conventional methodology for calculating these relationships with a sec- ond method that more precisely measures the relationship by reducing mea- surement bias. It employs traditional and advanced econometric techniques to measure correlations between labor market success—higher wages, increased chances of employment, formal sector employment, employed (or self-employed) status, and white-collar employment—and a range of skills. In this way, the analysis presents the cleanest possible correlation between skills (as opposed to education) and a range of labor outcomes. The results are consistent with those estimated with U.S. and European data, which present the cleanest evidence to date on the role of cognitive and socioemotional skills following individuals over time. Fifth, the study pro- poses a policy framework for facilitating the acquisition of a broader range of skills. It considers the biology, psychology, and sociology of the learning process to provide specific policy advice for the development of skills that are not usually in the curriculum but that matter for today’s labor market: socioemotional skills. It also presents pedagogical methods that are employed primarily for teaching cognitive skills but that also provide guidance for the development of socioemotional skills. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Introduction 7 Definitions of Skills and Data for Skills Measurement We define skills as the ability to perform a specific task. We classify them into two broad, overlapping subgroups: cognitive skills and socioemotional skills (figure 1.2). Cognitive skills can be defined as the “ability to understand complex ideas, to adapt effectively to the environment, to learn from experience, to engage in vari- ous forms of reasoning, and to overcome obstacles by taking thought” (Neisser and others 1996, 77). These skills can be thought of as intelligence or mental tasks, such as reasoning and information processing. Cognitive facets can be categorized into two dimensions: basic cognitive skills (basic academic knowledge, such as lit- eracy or numeracy) and advanced cognitive skills (more complex mental tasks, such as critical thinking, advanced problem solving, and time management). This study focuses on basic cognitive skills, which capture basic academic learning, including memory, numeracy, literacy, and evaluation of written information. The policy section includes a discussion of advanced cognitive skill development.4 Socioemotional skills are behaviors, attitudes, and personality traits that deter- mine how people do things.5 They transform cognitive skills into output, com- plementing knowledge with grit, teamwork, organization, commitment, creativity, and honesty, among other attributes. The data collected for the study measure a wide range of socioemotional skills (table 1.1 and box 1.2).6 For ease of presenta- tion we group them into three categories: achieving goals, working with others, and managing emotions.7 We refer to the specific underlying skills when they are useful to elucidate a point. Figure 1.2  Framework for Cognitive and Socioemotional Skills Achieving Basic goals cognitive Facets linked to Basic academic organization and knowledge, such perseverance as literacy or numeracy Socioemotional Cognitive Managing Behaviors, personality traits, and attitudes that enable emotions Mental abilities to engage in comprehension and reasoning individuals to navigate personal Facets related to Advanced and acquire knowledge and social situations effectively confidence and cognitive dealing with stress and emotions Complex thinking, such as critical Working thinking or with others problem solving Facets reflecting how people interact with others Sources: Almlund and others 2011; SEMS 2014; World Bank 2014b; OECD 2015; and authors’ elaboration. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 8 Introduction Table 1.1  Definitions of Skill Measures Type of skill Dimension Specific skill Definition Survey measure Basic cognitive Basic academic Memory Short-term memory, representative of Direct assessment knowledge working memory (test) and reasoning Math ability Ability to perform basic arithmetic operations, such as addition, subtraction, multiplication, division, and percentage Reading proficiency Ability to understand, evaluate, use, and engage with written texts Verbal ability Receptive vocabulary and verbal ability of adults Verbal fluency Speed and ease with which words are accessed from memory Socioemotional Achieving goals Conscientiousness Tendency to be organized, responsible, Aggregation of and hardworking self-reported Openness to Appreciation for art, learning, unusual items experience ideas, and variety of experience Grit Perseverance and passion for long-term goals Working with Agreeableness Tendency to act in cooperative, unselfish others manner Extroversion Sociability, tendency to seek stimulation in company of others, talkativeness Managing Emotional stability Predictability and consistency in emotions emotional reactions; absence of rapid mood changes Hostile attribution Tendency to perceive hostile intents in bias others Decision making Manner in which decision situations are approached Sources: John and Srivastava 1999; Cueto, Muñoz, and Baertl 2010; Almlund and others 2011; ETS 2014; SEMS 2014; World Bank 2014a, 2014b; and OECD 2015. Box 1.2 Measures of Socioemotional Skills Using the “Big Five” Classification This study uses the Big Five Model (Goldberg 1993), a widely accepted taxonomy, to measure personality traits. The five traits are openness to experience (also called intellect or culture), con- scientiousness, extroversion, agreeableness, and emotional stability (referred to as “resilience” in this study). Each trait comprises many specific personality characteristics, behaviors, and beliefs. Unlike cognitive skills, socioemotional skills are not necessarily monotonically increasing in desirability. Conscientiousness, for example, is a desirable trait: People who score high on this facet may be self-disciplined, plan, and deliver on time. However, highly conscientious indi- viduals may score low on spontaneity, which might be a crucial characteristic for successful performance in certain jobs. Agreeable people may be considerate and generous, but their box continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Introduction 9 Box 1.2  Measures of Socioemotional Skills Using the “Big Five” Classification (continued) personality might not lend itself to making controversial decisions. Other traits, such as extro- version, are simply different ways of processing the world: Extroverts draw energy from being with others, whereas introverts draw energy from internal stimulation. Some terms, such as emotional stability, have connotations that may lead observers to read more into them than they represent. People who score high on “neuroticism” (the opposite of emotional stability) are not (necessarily) clinically neurotic. They are simply more anxious and reactive than people at the other end of the scale. It is thus useful to understand the Big Five traits as measures of where a person is located along the continuum of a personality facet as opposed to a measure of trait desir- ability. The continuum (low to high score) for each facet can be understood as shown in box table B1.2.1. In addition to the Big Five personality traits, the study also measures behaviors and beliefs such as grit (the perseverance and motivation to achieve long-term goals [Duckworth and others 2007]); hostile attribution bias (the tendency to perceive others’ actions as hostile and react aggressively in consequence [a mix of resilience and agreeableness] [Dodge 2003]); and decision making (the manner in which individuals cope with the stress of decision situa- tions, a subfacet of resilience) (Mann and others 1997). Table B1.2.1 Interpretation of Low and High Scores of Big Five Personality Traits Big Five personality trait Low score High score Openness to experience Routine, straightforward Complex, experimental Conscientiousness Spontaneous Self-disciplined, planned Extroversion Energized by internal stimulation Energized by external stimulation Agreeableness Self-interested, suspicious Kind, willing to compromise Emotional stability Anxious, reactive Resilient, calm The data underpinning this report were collected through four surveys (box 1.3). Household and labor force surveys usually include educational attain- ment and labor force behaviors. They do not include information on skills, partly because appreciation of the role skills play in a range of behaviors is only recent and partly because of the complex (and costly) methods needed to collect and report skills data. The rest of this book is organized as follows. Chapter 2 presents the cognitive and socioemotional skills profile of the labor force in the four countries under study and explores the level of skills of various groups, including by comparing cognitive skills in Latin American countries and countries outside the region. Chapter 3 identifies skills that are correlated with labor market outcomes in Latin America and compares the results with the results for high-income coun- tries. Chapter 4 presents a framework for offering policy recommendations. Chapter 5 summarizes the results of the study and compares them with findings from high-income countries. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 10 Introduction Box 1.3 How Data Were Collected for This Study Four data sets were collected for this study, derived from two survey instruments designed by the World Bank. Data on Peru come from the 2010 Peruvian Encuesta Nacional de Habilidades (National Skills Survey). which supplements the annual Peruvian household survey by adding modules that measure cognitive skills (verbal fluency, verbal literacy, math, and memory) and personality traits. The sample includes 1,394 adults age 18–50 in four urban areas of Peru. Data for Bolivia, Colombia, and El Salvador come from the STEP Household Surveys, a multicountry measurement initiative asking a range of background questions typical to ­ labor  force surveys, including modules on personality traits and skills usage (World Bank 2014b). For Bolivia and Colombia, data on reading proficiency (a construct capturing the ability to read, process, evaluate, and use written information [ETS 2014]) are also included. The El Salvador Skills Survey does not include direct assessment of cognitive skills. These surveys cover the working-age population (15–64) living in urban areas, with sample sizes of 2,439 in Bolivia, 2,617 in Colombia, and 2,335 in El Salvador. Each survey measures the skills variables slightly differently. The Peruvian survey assigns the total number of correct answers on each cognitive test as a value for each observation. Personality traits are created by conducting a principal components analysis of the 58 ques- tions surveyed to construct six factors that mirror Goldberg’s Big Five model plus two factors to measure grit (Cueto, Muñoz, and Baertl 2010). The reading proficiency variable for the STEP Household Survey for Bolivia and Colombia is based on a range of plausible scores that are a function of the test results and a set of back- ground characteristics (see ETS 2004 and Von Davier, Gonzalez, and Mislevy 2009 for method- ological discussion). Socioemotional skills constructs in the STEP Household Survey are derived by computing an inter-item average of the values across 24 questions to represent each of the Goldberg Big Five traits, the grit trait, and behavioral gauges capturing hostile attribution bias and decision-making style (World Bank 2014b). The El Salvador Skills Survey includes measures of the intensity of use of cognitive skills and a socioemotional skills module similar to the one in the STEP survey; it includes 26 questions covering the same facets. The limited number of questions on socioemotional skills in the Bolivia, Colombia, and El Salvador data sets may raise concerns about a lack of precision in measurement that could influence the relationships with our outcomes of interest. The data do not permit assessment of whether some of the results reflect true associations or measurement issues. Other large- scale surveys, like the German Socio-Economic Panel survey, have implemented similar mod- ules with reliable estimates (Lang and others 2011; World Bank 2014b). Notes 1. The meta-evaluation classifies more than 140 skills that emerged from 27 studies into four groups: socioemotional, basic cognitive (basic knowledge and problem solving), advanced cognitive (more complex thinking), and technical skills. 2. To test whether the remarkable consistency in employer demand for socioemotional and advanced cognitive skills reflects management fashion rather than true Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Introduction 11 preferences, CIDAC (2014) compares the skills prioritized in an employer survey with the skills used by headhunters to identify job candidates or assess workers. The significant overlap between the skills employers claim to value and those they pay headhunters to measure and identify suggests that the stated preferences reflect true preferences. 3. In 2013 the proportion of the population living in urban areas in the four countries was 65 percent in El Salvador, 68 percent in Bolivia, 76 percent in Colombia, and 78 percent in Peru (World Bank 2015). 4. Technical skills can be thought of as a subset of cognitive skills (Almlund and others 2011). They can be defined as those abilities that are associated with the specific knowledge to carry out one’s occupation—the ability to repair a car muffler, the knowledge to identify specific bacteria under a microscope, the know-how to assem- ble dozens of shirts per hour. Analysis of the formation and influence of specific technical skills on outcomes is beyond the scope of this study. 5. Although these skills clearly involve some level of cognition, economists have desig- nated them as “noncognitive skills” to differentiate them from academic or learning skills. “Traits” are characteristics or patterns of thought and action that are relatively stable across the life cycle. “Behaviors” are performances in response to stimulation. “Attitudes” encompass beliefs and values that guide skill formation and behavior. 6. Table 1.1 does not include advanced cognitive skills (which may be understood as critical thinking, such as application of knowledge, analysis, synthesis, and evaluation [Bloom and others 1956]), because they were not measured in the data used in this study or in the other Latin American studies that measures skills. They are briefly discussed in the policy chapter, because employers place great importance on them. 7. Many taxonomies for summarizing socioemotional skills exist. This categorization is used for two reasons. First, it mirrors the taxonomy in OECD (2015), which facili- tates the comparison of the results from the two studies. It represents a small step toward finding a common language in the analysis and policy discussion of these skills. Second, it reflects a taxonomy being developed and implemented in Latin America. Peru’s Ministry of Education uses this taxonomy as the organizing frame- work for a range of programs, including Escuela Amiga (Friendly School). The Undersecretariat of Upper Secondary Education of Mexico uses this framework as the basis of its nationwide program Construye-T (Build Yourself) (SEMS 2014; World Bank 2014a). References Acemoglu, D., and D. H. Autor. 2011. “Skills, Tasks and Technologies: Implications for Employment and Earnings.” In Handbook of Labor Economics, Volume 4. Eds. O. Ashenfelter and D. E. Card, 1043–1171. Amsterdam: Elsevier. Aedo, C., J. Hentschel, J. Luque, and M. Moreno. 2013. “From Occupations to Embedded Skills: A Cross-Country Comparison.” Policy Research Working Paper 6560, World Bank, Washington, DC. Aedo, C., and I. Walker. 2012. Skills for the 21st Century in Latin America and the Caribbean. Washington, DC: World Bank. Almlund, M., A. L. Duckworth, J. J. Heckman, and T. Kautz. 2011. “Personality Psychology and Economics.” In Handbook of the Economics of Education, Volume 4. Ed. E. A. Hanushek, 1–181. Amsterdam: North Holland. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 12 Introduction Autor, D. H., H. Levy, and F. R. Murnane. 2003. “The Skill Content of Recent Technological Change: An Empirical Exploration.” Quarterly Journal of Economics. 118 (4): 1279–1334. Barro, R. J., and J. W. Lee. 2013. “A New Data Set of Educational Attainment in the World, 1950–2010.” Journal of Development Economics 104: 184–98. Bassi, M., M. Busso, S. Urzúa, and J. Vargas. 2012. Disconnected: Skills, Education and Employment in Latin America. Inter-American Development Bank, Washington, DC. Blom, A., and C. Hobbs. 2008. School and Work in the Eastern Caribbean: Does the Education System Adequately Prepare Youth for the Global Economy? World Bank, Washington, DC. Bloom, B., M. Engelhart, J. Furst, W. Hill, and D. Krathwohl. 1956. Taxonomy of Educational Objectives: The Classification of Educational Goals. Handbook I: Cognitive Domain. New York: David McKay Company. Bruns, B., and J. Luque. 2014. Great Teachers: How to Raise Student Learning in Latin America and the Caribbean. World Bank, Washington, DC. CIDAC (Centro de Investigación para el Desarrollo). 2014. “Encuesta de competencias profesionales: ¿Qué buscan—y no encuentran—las empresas en los profesionistas jóvenes?” Mexico City. Cueto, S., I. Muñoz, and A. Baertl. 2010. “Scholastic Achievement, Cognitive Skills and Personality Traits of Youths and Adults in Peru: A Cross-Sectional and Intergenerational Analysis.” Group for the Analysis of Development (GRADE), Lima. Cunha, F., J. J. Heckman, and S. M. Schennach. 2010. “Estimating the Technology of Cognitive and Noncognitive Skill Formation.” Econometrica 78 (4): 883–931. Cunningham, W., and P. Villaseñor. 2016. “Employer Voices, Employer Demands, and Implications for Public Skills Development Policy Connecting the Labor and Education Sectors.” World Bank Research Observer 31 (1): 102–34. Dodge, K. A. 2003. “Do Social Information Processing Patterns Mediate Aggressive Behavior? In Causes of Conduct Disorder and Juvenile Delinquency, eds. B. B. Lahey, T. E. Moffitt, and A. Caspi (pp. 254–274). New York: Guilford Press. Duckworth, A., C. Peterson, M. Matthews, and D. Kelly. 2007. “Grit: Perseverance and Passion for Long-Term Goals.” Journal of Personality and Social Psychology 92 (6): 1087–101. Duncan, G. J., C. J. Dowsett, A. Claessens, K. Magnuson, A. C. Huston, P. Klebanov, L. S. Pagani, L. Feinstein, M. Engel, J. Brooks-Gunn, H. Sexton, K. Duckworth, and C. Japel. 2007. “School Readiness and Later Achievement.” Developmental Psychology 43 (6): 1428–46. ETS (Educational Testing Services). 2014. A Guide to Understanding the Literacy Assessment of the STEP Skills Measurement Survey. Princeton, NJ: IEA-ETS Research Institute. Gasparini, L., G. Cruces, S. Galiani, and P. Acosta. 2011. “Educational Upgrading and Returns to Skills in Latin America: Evidence from a Supply-Demand Framework for the Decades of 1990 and 2000.” Background paper prepared for C. Aedo and I. Walker. 2012. Skills for the 21st Century in Latin America and the Caribbean. Washington, DC: World Bank. Goldberg, L. R. 1993. “The Structure of Phenotypic Personality Traits.” American Psychologist 48 (1): 26–34. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Introduction 13 Hanushek, E. A., and L. Woessmann. 2008. “The Role of Cognitive Skills in Economic Development.” Journal of Economic Literature 46 (3): 607–68. Hanushek, E. A., and L. Zhang. 2009. “Quality-Consistent Estimates of International Schooling and Skill Gradients.” Journal of Human Capital 3 (2): 107–43. Heckman, J. J., J. Stixrud, and S. Urzúa. 2006. “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior.” Journal of Labor Economics 24 (3): 411–82. John, O. P., and S. Srivastava. 1999. “The Big Five Trait Taxonomy: History, Measurement and Theoretical Perspectives.” In Handbook of Personality: Theory and Research, ed. L. A. Pervin and O. P. John (2nd ed., pp. 102–138). New York: Guilford Press. Lang, F. R., D. John, O. Lüdtke, J. Schupp, and G. G. Wagner. 2011. “Short Assessment of the Big Five: Robust Across Survey Methods Except Telephone Interviewing.” Behavior Research Methods 43 (2): 548–67. Levy, S., and N. Schady. 2013. “Latin America’s Social Policy Challenge: Education, Social Insurance, Redistribution.” Journal of Economic Perspectives 27 (2): 193–218. Mann, L., P. Burnett, M. Radford, and S. Ford. 1997. “The Melbourne Decision Making Questionnaire: An Instrument for Measuring Patterns for Coping with Decisional Conflict.” Journal of Behavioral Decision Making 10 (1): 1–19. Mourshed, M., D. Farrell, and D. Barton. 2012. Education to Employment: Designing a System that Works. McKinsey Center for Government, Washington, DC. Muller, N. 2014. “Literature Review on the Role of Skills in the Labor Market.” Background paper for the regional study Minds and Behaviors at Work: Socioemotional Skills for Latin America’s Workforce.” World Bank, Washington, DC. Neisser, U., G. Boodoo, T. J. Bouchard, A. W. Boykin, N. Brody, S. J. Ceci, D. F. Halpern, J. C. Loehlin, R. Perloff, R. J. Sternberg, and S. Urbina. 1996. “Intelligence: Knowns and Unknowns.” American Psychologist 51 (2): 77–101. OECD (Organisation for Economic Co-operation and Development). 2013. PISA 2012 Results: What Students Know and Can Do—Student Performance in Mathematics, Reading and Science, vol. 1. Paris: OECD Publishing. ———. 2015. Skills for Social Progress: The Power of Social and Emotional Skills. OECD Skills Studies. Paris: OECD Publishing. Ogier, T. 2009. Skills to Compete: Post-Secondary Education and Business Sustainability in Latin America. Economist Intelligence Unit, London. Pages, C., ed. 2010. The Age of Productivity: Transforming Economies from the Bottom Up. Washington, DC: Inter-American Development Bank. SEMS (Subsecretaría de Educación Media Superior de México; Undersecretariat of Upper Secondary Education of Mexico). 2014. Programa Construye T 2014–2018: Fortalecer las capacidades de la escuela para promover el desarrollo integral de los jóvenes [Program Build Yourself 2014–2018: Strengthening the Capacities of the School to Promote the Comprehensive Development of Young People]. http://www.construye-t.org.mx​ /­resources/DocumentoConstruyeT.pdf. Von Davier, M., E. Gonzalez, and R. Mislevy. 2009. “What Are Plausible Values and Why Are They Useful?” In M. von Davier and D. Hastedt (eds.). IERI Monograph Series: Issues and Methodologies in Large Scale Assessments. Vol. 2. IEA-ETS Research Institute, Princeton, NJ. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 14 Introduction World Bank. 2011. Strengthening Skills and Employability in Peru. Report 61699-PE, Washington, DC. ———. 2014a. “STEP Skills Measurement Surveys: Innovative Tools For Assessing Skills.” Social Protection and Labor Discussion Paper 1421, World Bank, Washington, DC. ———. 2014b. “Toolkit for Socio-Emotional Learning ‘Paso a Paso’ (Step by Step) for the Program ‘Escuela Amiga’ (Friendly School) for Primary and Secondary Public Schools of Peru.” In Internal Report of Non-Lending Technical Assistance on Education, Skills, and Employment. Washington, DC: World Bank. ———. 2015. World Development Indicators. http://data.worldbank.org/data-­catalog/world​ -development-indicators (accessed June 3, 2015). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Chapter 2 Cognitive and Socioemotional Skills Profile of the Latin American Workforce This chapter profiles the cognitive and socioemotional skills mix in the urban working-age populations of four Latin American countries: Bolivia, Colombia, El Salvador, and Peru.1 The profiles reveal the extent to which skills differ by age, gender, and education level. They can be helpful in developing intuition and basic interpretations about skills levels and how they are distributed across the population. The descriptive statistics do not take into account other factors that could influence the skills distribution or untangle their respective effects. The chapter yields four main findings: • Skills are not synonymous with years of schooling. • Skills distributions are roughly similar across different groups of the working- age population within countries (defined by age, gender, or education level). Within-country similarities may help even the playing field for some groups— such as women and youth—but they point to failures for others, such as better- educated people whose cognitive scores are not higher than those of people with less education. • Socioemotional skills distributions vary widely across countries, with cognitive skills showing fewer disparities. • The cognitive skills of the labor force in Latin America severely lag those of other regions, even after accounting for the level of economic development (proxied by GDP per capita). The disparity threatens to grow even wider in the future. Mapping the Distribution of Cognitive Skills The distribution of each measured cognitive skill is similar across the four countries, except when the distribution is disaggregated by age groups. Figure 2.1 shows the distribution of three types of cognitive skills: math ability, Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   15   16 Cognitive and Socioemotional Skills Profile of the Latin American Workforce verbal fluency, and reading proficiency (see table 1.1 for definitions). The dis- tribution of these skills is normal (with a similar number of adults above and below the average score), with some slight deviations. Verbal fluency in Peru has a longer right tail, indicating lower than average verbal fluency for the population and only a smaller number of Peruvians demonstrating high verbal fluency (figure 2.1, panel c). Reading proficiency reveals a slightly bimodal pattern in Colombia. Cognitive skills are similar across genders: The differences between men’s and women’s cognitive skills in Bolivia, Colombia, and Peru are either small or not statistically significant. The relationship between age and cognitive skills is less consistent across countries. Bolivian and Colombian youth (15–24) are more proficient at reading, evaluating, and analyzing written texts than their elders, with the largest gap with the oldest age group (50–64). By contrast, Peruvian youth (18–24) have lower Figure 2.1  Distribution of Test Scores of Cognitive Skills in Bolivia, Colombia, and Peru a. Reading proficiency in Bolivia b. Reading proficiency in Colombia 0.5 0.5 0.4 0.4 Frequency Frequency 0.3 0.3 0.2 0.2 0.1 0.1 0 0 –4 –2 0 2 4 –4 –2 0 2 4 Standardized reading proficiency score Standardized reading proficiency score c. Verbal fluency in Peru d. Math ability in Peru 0.5 0.5 0.4 0.4 Frequency Frequency 0.3 0.3 0.2 0.2 0.1 0.1 0 0 –2 0 2 4 6 –4 –2 0 2 4 Standardized verbal uency score Standardized math ability score Sources: Bolivia and Colombia: STEP Household Surveys (2012). Peru: ENHAB (2010). Note: Frequencies are computed as Kernel densities. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Cognitive and Socioemotional Skills Profile of the Latin American Workforce 17 verbal skills than older adults; their levels of math ability are about the same and their working memory is slightly better. In El Salvador, where the survey measured the application of cognitive skills, significant differences are evident between men and women and people in dif- ferent age groups in the use of reading, math ability, and writing. For example, 47 percent of urban men but only 32 percent of urban women use intermediate or advanced math ability. The distribution of cognitive skills for any educational level heavily overlaps that of other education levels, suggesting that years of education are not a good proxy for skills acquired. The increment in skills is not uniformly increasing with years of education, especially at the secondary level and beyond; the range of basic cognitive skill levels among adults who completed a particular level of education is wide. For instance, Bolivian adults with some secondary school who attained above-average reading proficiency scores do about as well as low-performing Bolivian adults with some university education (figure 2.2). In Peru the distribution of verbal, memory, and math abilities overlaps across all educational levels, although the overlap is smaller between people who completed primary school and people who are less and more educated (Cunningham, Parra-Torrado, and Sarzosa (2016)). These patterns are also evident in Colombia, although the distribution of skills increases with educa- tion level more there than it does elsewhere. The wide heterogeneity in read- ing proficiency scores across educational levels may also reflect variation across age cohorts, which experience very different education systems; the quality of services offered in different educational institutions; or both. Figure 2.2  Distribution of Reading Proficiency Scores in Bolivia and Colombia, by Educational Level a. Bolivia b. Colombia Below primary Below primary Primary Level of education Level of education Basic General secondary Upper-secondary Vocational secondary Vocational tertiary Vocational tertiary General tertiary General tertiary 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 Reading proficiency score Reading proficiency score Source: Bolivia and Colombia STEP Household Surveys (2012). Note: Vertical line indicates sample average. Boxes and whiskers show values of key points of the distribution in reading proficiency scores. The left and right sides of the box are respectively the value of the 25th percentile of skills and the 75th percentile; the inside bar of the box is the median value (half of the respective education group is above the value and the other half is below). The extreme left and right adjacent lines represent respectively the lowest datum still within 1.5 times the length of the box from the 25th percentile and the highest datum still within 1.5 times the length of the box from the 75th percentile. Figures exclude outliers, which are values above or below adjacent lines. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 18 Cognitive and Socioemotional Skills Profile of the Latin American Workforce Reading proficiency scores from the STEP Household Surveys can be compared across dozens of low- and middle-income countries and with 23 Organisation for Economic Co-operation and Development (OECD) countries that participated in the 2012 Program of International Assessment of Adult Skills (PIAAC) survey. Doing so is not without bias, however, as the STEP surveys were conducted only in urban areas, whereas the surveys in the other countries are nationally representative. Differences across countries were not tested for statistical significance. Of the countries shown in figure 2.3, Bolivia and Colombia, the two Latin American countries covered by the STEP surveys, have lower reading proficiency scores than all countries except Kenya and Ghana. Bolivia’s score is 200 and Colombia’s is 235 (table 2.1 describes the abilities associated with each score level). More worrisome is the fact that Bolivia underperforms relative to its per capita GDP by about 25 points. Colombia does as well as expected for its level of development, although its scores are lower than the scores of some countries with lower incomes, including Ukraine, Armenia, Georgia and Vietnam. The country averages mask the dire lack of reading proficiency in Latin America. When reading proficiency scores are grouped into six achievement Figure 2.3 Correlation between Adult Reading Proficiency Scores and Per Capita Income in Selected Countries, 2012 350 Japan 300 Reading pro ciency score Estonia Norway Ukraine Armenia Poland France United States 250 Georgia Spain Italy Vietnam Colombia 200 Bolivia Kenya 150 Ghana 100 0 10 20 30 40 50 60 70 2012 per capita GDP (thousands of 2011 purchasing power parity dollars) High-income OECD Low- and middle-income Bolivia and Colombia countries countries Sources: OECD countries: OECD (2013a), based on PIAAC data (2012–13); Bolivia and Colombia: STEP Household Surveys (2012); other countries: STEP Household Surveys (2012–13). Per capita GDP data are from World Bank (2015), based on World Bank’s International Comparison Program database. Note: STEP data for middle-income countries are representative only of urban areas (adults 15–64). PIAAC data cover the national population (adults 15–65). Reading proficiency scores range from 0 (lowest) to 500 (highest). Average for the 23 OECD countries is 273. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Cognitive and Socioemotional Skills Profile of the Latin American Workforce 19 Table 2.1  Description of Reading Proficiency Levels in the PIAAC and STEP Surveys Level (score range) Description Below 1 (0–175) Respondent reads brief texts on familiar topics to locate single piece of specific information. Only basic vocabulary knowledge is required. 1 (176–225) Respondent reads relatively short texts to locate single piece of information that is similar to information given in question. Knowledge and skill in recognizing basic vocabulary, evaluating the meaning of sentences, and reading of paragraph text are expected. 2 (226–275) Complexity of text increases. Respondents match text and information, paraphrase, and make low-level inferences. Competing pieces of information may be present. 3 (276–325) Texts are dense or lengthy, including multiple pages. Understanding text and rhetorical structures becomes more central to successfully completing tasks. Many tasks require respondent to construct meaning across larger chunks of text or perform multistep operations in order to identify and formulate responses. 4 (326–375) Respondents perform multistep operations to integrate, interpret, or synthesize information from complex or lengthy texts. Many tasks require identification and understanding of one or more specific, noncentral ideas in the text in order to interpret or evaluate subtle evidence claim or persuasive discourse relationships. 5 (376–500) Respondents search for and integrate information across multiple, dense texts. Application of logical and conceptual models of ideas is required to accomplish tasks. Tasks often require respondents to be aware of subtle rhetorical cues and to make high-level inferences or use specialized background knowledge. Sources: OECD 2013c; ETS 2014; World Bank 2014. categories (see table 2.1), about 60 percent of urban Bolivian adults display only a basic level of proficiency (at or below level 1). Readers at this level can perform reading tasks only from short pieces with no or little competing information (figure 2.4). In contrast, just 32 percent of adults in Vietnam, an economy with lower per capita GDP, perform at this level. In wealthier Colombia, 36 percent of workers read at the same basic level, compared with 14 percent of Ukrainian and 15 percent of adults in high-income OECD countries. At the other end of the scale, just 11 percent of Bolivians and 23 percent of Colombians can understand complex texts (levels 3–5), compared with 33 percent of Vietnamese, 47 percent of Ukrainians, and 52 percent of adults in high-income OECD countries (figure 2.4). The weak cognitive skills of Latin America’s labor market are not a passing phenomenon, as shown by the poor performance of youth, who are soon to be the newest wave of workers. International comparisons of student achievements do not presage brighter prospects in coming years (box 2.1). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 20 Cognitive and Socioemotional Skills Profile of the Latin American Workforce Figure 2.4  Adult Reading Proficiency Levels in Bolivia, Colombia, and Selected Other Countries, 2012 100 90 80 70 Percent of adults 60 50 40 30 20 10 0 Ghana Kenya Bolivia Colombia Vietnam Georgia Armenia Ukraine OECD-23 Level 1 and below Level 2 Level 3 Levels 4 and 5 Sources: OECD countries: OECD (2013a) based on PIAAC data (2012–13); Bolivia and Colombia: STEP Household Surveys (2012); other countries: STEP Household Surveys (2012–13). Note: STEP data for middle-income countries are representative only of urban areas (adults 15–64). PIAAC data cover national population (adults 15–65). Reading proficiency scores range from 0 (lowest) to 500 (highest). Average for the 23 OECD countries is 273. For description of levels, see table 2.1. Box 2.1 How Do Cognitive Skills of Future Labor-Market Entrants in Latin America Compare with Their Peers from Outside the Region? Results from the PISA Every three years since 2000, the OECD’s Programme for International Student Assessment (PISA) tests 15-year-olds on basic cognitive skills such as math, reading, and science to deter- mine whether students can process and reason from written and numerical information in a multitude of test-based situations (OECD 2013b). Eight Latin American countries—Argentina, Brazil, Chile, Colombia, Costa Rica, Mexico, Peru, and Uruguay—are among the 65 high- and middle-income countries surveyed. Latin American countries perform poorly in all three disciplines, especially given their level of national wealth. In 2012 all eight Latin American countries were in the bottom third of countries surveyed (Bos, Ganimian, and Vegas 2013; OECD 2013a). Average learning per- formance on all international tests over the past 40 years is lower in Latin America than in every region except Sub-Saharan Africa and below the mean for their level of per capita GDP (Hanushek and Woessmann 2012). Except in Chile, whose performance stands out in box continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Cognitive and Socioemotional Skills Profile of the Latin American Workforce 21 Box 2.1  How Do Cognitive Skills of Future Labor-Market Entrants in Latin America Compare with Their Peers from Outside the Region? Results from the PISA (continued) the region, typical Latin American students achieve only the minimum level of proficiency. They cannot interpret or recognize questions that require more than direct inference; they cannot use basic algorithms, formulas, or procedures to solve problems using whole num- bers and interpret results literally. The current gap in math performance between the aver- age for OECD and Latin American countries represents a disparity equivalent to more than two full years of math education (Bruns and Luque 2014). Differences in performance are considerable both across and within countries in the region (Bos, Ganimian, and Vegas 2013; Bruns and Luque 2014). Some Latin American countries are making progress in closing disparities with OECD coun- tries, but stunning gaps remain. Between 2000 and 2012, Chile, Brazil, and Peru registered some of the biggest improvements in the entire contingent of PISA-surveyed countries (Hanushek, Peterson, and Woessmann 2012; Bruns and Luque 2014) (figure B2.1.1). In contrast, Argentina and Costa Rica reported no significant changes, and the performance of Uruguayan students declined (Bos, Ganimian, and Vegas 2014). The absence of other Latin American countries and school dropouts in the PISA sam- ple suggests that the actual gap in cognitive skills between the region and high-income countries is even larger. Regional student assessments indicate that the average learning Figure B2.1.1 PISA Math Scores in Latin American Countries and OECD, 2000–12 500 470 Average national math score 440 410 380 350 320 290 2000 2003 2006 2009 2012 OECD average Mexico Colombia Uruguay Chile Brazil Argentina Costa Rica Peru Source: Bruns and Luque 2014, based on OECD 2013b. Note: The OECD average excludes Chile and Mexico, which are OECD member countries. box continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 22 Cognitive and Socioemotional Skills Profile of the Latin American Workforce Box 2.1  How Do Cognitive Skills of Future Labor-Market Entrants in Latin America Compare with Their Peers from Outside the Region? Results from the PISA (continued) performance in countries that did not participate in the PISA (including Bolivia, Honduras, and República Bolivariana de Venezuela) is substantially lower than performance in the eight countries that did participate (Bruns and Luque 2014). In addition, given that a larger share of all 15-year-olds have already dropped out of school in Latin America than in the OECD, the actual gap in skills is even larger than the PISA results suggest (De Hoyos, Rogers, and Székely 2015). Weak performance is not solely the result of school characteristics or teaching quality at the time students are tested; it originates from deficits created in early life. Students’ family background and home environment (parents’ education and socioeconomic status, access to books at home) and deficits in children’s development during their earliest years greatly affect cognitive development (Knudsen and others 2006; Schady and others 2015). Because the development of cognitive skills is a continuous and cumulative process, early cognitive deficits have adverse impacts on children’s readiness and capacity to learn (Almond and Currie 2011). Pedagogical factors also play a role in student performance. In addition to socioeconomic background (gender, age, student and school socioeconomic status) and school structure (class size, teachers’ education, whether a school is public or private), factors such as classroom time and teacher expectations of students affect performance both in Latin America and OECD countries (OECD/CAF/ECLAC 2014; Avendaño and others 2016). In sum, mean scores on reading, math, and science tests improved in some but not all Latin American countries, significant gaps remain with other high- and middle-income countries, and divergences within countries are considerable. The gaps in cognitive skills of future labor market entrants in Latin America is even bigger when considering countries that do not par- ticipate in PISA and the out-of-school youth. In addition, PISA captures only a subset of skills (basic cognitive skills); it does not address socioemotional skills that help students remain in school and learn better. Mapping the Distribution of Socioemotional Skills There is little difference in the distribution of socioemotional skills by gender or age group. There are only slight differences in the distribution of skills associ- ated with achievement of goals (conscientiousness, grit) and managing emotions (decision making and hostile attribution bias) between men and women. The only notable difference is that in all four countries, resilience, a trait associ- ated with managing emotions, tends to be higher among men than women (figure 2.5). Across countries youth (defined here as people 15–24) and adults display dif- ferent levels in some dimensions of socioemotional skills. In all four countries youth are less extroverted and persevering than young adults (defined as people age 25–49); they also appear less agreeable (less generous, polite, or forgiving). However, youth in Bolivia, Colombia, and El Salvador are less likely to perceive Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Cognitive and Socioemotional Skills Profile of the Latin American Workforce 23 Figure 2.5  Distribution of Resilience among Men and Women in Bolivia, Colombia, El Salvador, and Peru a. Bolivia b. Colombia 35 35 30 30 Percent of population Percent of population 25 25 20 20 15 15 10 10 5 5 0 0 –4 –3 –2 –1 0 1 2 3 4 –4 –3 –2 –1 0 1 2 3 4 Standardized resilience score Standardized resilience score c. El Salvador d. Peru 35 0.5 30 Percent of population 0.4 25 Kernel density 20 0.3 15 0.2 10 0.1 5 0 0 –4 –3 –2 –1 0 1 2 3 4 –4 –2 0 2 Standardized resilience score Standardized resilience score Men Women Source: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010). Note: Data cover urban adults (15–64). Individuals with high levels of resilience have predictable and consistent emotional reactions and do not rapidly change moods. hostility in others than are young adults. In Peru youth are as likely as young adults to be cooperative (figure 2.6). Less-educated individuals and their more educated peers have different socio- emotional skills across all three categories (achieving goals, managing emotions, and working with others). In Bolivia, Colombia, and Peru, people with little or no formal education often have fewer skills for managing emotions (fewer decision-making skills and less resilience) and working with others (they are less ­ open to experience and more introverted) (figure 2.7). Differences across educa- tional levels are much less marked in El Salvador. The extent to which in-country similarities in skills levels and between-­ country gaps translate into labor market success depends on the extent to which each of these skills affect labor market outcomes. The next chapter identifies the skills that seem to be associated with higher wages and more and better-quality employment. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Figure 2.6  Distribution of Hostile Attribution Bias in Bolivia, Colombia, and El Salvador and Cooperation in Peru, by Age Group a. Hostile attribution bias in Bolivia b. Hostile attribution bias in Colombia 40 40 Percent of population Percent of population 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 –4 –3 –2 –1 0 1 2 3 4 –4 –3 –2 –1 0 1 2 3 4 Standardized hostile attribution bias score Standardized hostile attribution bias score Youth (15–24) Adults (25–49) Older (50–64) c. Hostile attribution bias in El Salvador d. Cooperation in Peru 40 0.6 Percent of population 35 Kernel density 30 25 0.4 20 15 0.2 10 5 0 0 –4 –3 –2 –1 0 1 2 3 4 –6 –4 –2 0 2 4 Standardized hostile attribution bias score Standardized cooperation score Youth (15–24) Youth (18–24) Adults (25–49) Young adults I (25–30) Older (50–64) Young adults II (31–40) Middle-aged adults (41–50) Source: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010). Note: People with higher levels of hostile attribution bias tend to perceive hostile intents in others and to react hostilely as a result. People with higher levels of cooperation tend to collaborate better with others and are defined as more agreeable. Figure 2.7  Distribution of Openness to Experience in Bolivia, Colombia, El Salvador, and Peru, by Educational Level a. Bolivia b. Colombia 45 45 40 40 Percent of population Percent of population 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 –4 –3 –2 –1 0 1 2 3 4 –4 –3 –2 –1 0 1 2 3 4 Standardized openness to experience score Standardized openness to experience score Below primary Vocational secondary Below primary Vocational tertiary Primary Vocational tertiary Primary General tertiary General secondary General tertiary Secondary figure continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Cognitive and Socioemotional Skills Profile of the Latin American Workforce 25 Figure 2.7  Distribution of Openness to Experience in Bolivia, Colombia, El Salvador, and Peru, by Educational Level (continued) c. El Salvador d. Peru 45 0.5 Percent of population 40 35 0.4 Kernel density 30 25 0.3 20 0.2 15 10 0.1 5 0 0 –4 –3 –2 –1 0 1 2 3 4 –6 –4 –2 0 2 4 Standardized openness to experience score Standardized open to experience score Below primary Vocational tertiary Primary Primary General tertiary Incomplete secondary Secondary Complete secondary Incomplete higher education Complete higher education Source: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010). Note 1. The results in this chapter are drawn from four background papers prepared for this study: Barón, Sarzosa, and Mola (2015) for Bolivia (2015); Acosta, Muller, and Sarzosa (2015) for Colombia; Oviedo and Muller (2015) for El Salvador; and Cunningham, Parra Torrado, and Sarzosa (2016) for Peru. References Acosta, P. A., N. Muller, and M. Sarzosa. 2015. “Beyond Qualifications: Returns to Cognitive and Socio-Emotional Skills in Colombia.” Policy Research Working Paper 7430, World Bank, Washington, DC. Almond, D., and J. Currie. 2011. “Human Capital Development before Age Five.” In Handbook of Labor Economics, vol. 4B, ed. O. Ashenfelter and D. Card, 1315–486. Amsterdam: Elsevier. Avendaño, R., F. Barrera, S. Nieto-Parra, and F. Vever. 2016. “Understanding Student Performance beyond Traditional Factors: Evidence from PISA 2012.” OECD Development Centre Working Paper 331. Organisation for Economic Co-operation and Development, Paris. Barón, J. D., M. Sarzosa, and J. Mola. 2015. Ethnicity and Labor Market Returns to Cognitive and Socio-Emotional Skills in Urban Bolivia. World Bank, Washington, DC. Bos, M. S, A. J. Ganimian, and E. Vegas. 2013. Brief #1: ¿Cómo le fue a la región? Serie de briefs—América Latina en PISA 2012 (Brief # 1: How Well Did the Region Do? Briefs Series—Latin America in PISA 2012), Inter-American Development Bank, Washington, DC, and Organisation for Economic Co-operation and Development, Paris. ———. 2014. Brief #2: ¿Cuánto mejoró la región? Serie de briefs—América Latina en PISA 2012 (Brief # 1: How Much Did the Region Improve? Briefs Series—Latin America Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 26 Cognitive and Socioemotional Skills Profile of the Latin American Workforce in PISA 2012), Inter-American Development Bank, Washington, DC, and Organisation for Economic Co-operation and Development, Paris. Bruns, B., and J. Luque. 2014. Great Teachers: How to Raise Student Learning in Latin America and the Caribbean. Washington, DC: World Bank. Cunningham, W. M. Parra Torrado, and M. Sarzosa. 2016. “Cognitive and Non-Cognitive Skills for the Peruvian Labor Market: Addressing Measurement Error through Latent Skills Estimations.” Policy Research Working Paper 7550, World Bank, Washington, DC. De Hoyos, R., H. Rogers, and M. Székely. 2015. Out of School and out of Work: Risk and Opportunities for Latin America’s Ninis. Washington, DC: World Bank. ETS (Educational Testing Services). 2014. A Guide to Understanding the Literacy Assessment of the STEP Skills Measurement Survey. Princeton, NJ: IEA-ETS Research Institute. Hanushek, E., P. Peterson, and L. Woessmann. 2012. “Achievement Growth: International and U.S. State Trends in Student Performance.” Harvard’s Program on Education Policy and Governance/Education Next, Harvard Kennedy School, Cambridge, MA. Hanushek, E. A., and L. Woessmann. 2012. “Schooling, Educational Achievement, and the Latin American Growth Puzzle.” Journal of Development Economics 99 (2): 497–512. Knudsen, E. I., J. J. Heckman, J. L. Cameron, and J. P. Shonkoff. 2006. “Economic, Neurobiological and Behavioral Perspectives on Building America’s Future Workforce.” Proceedings of the National Academy of Sciences of the United States of America (PNAS) 103 (27): 10155–62. OECD (Organisation for Economic Co-operation and Development). 2013a. OECD Skills Outlook 2013: First Results from the Survey of Adult Skills. Paris: OECD Publishing. ______. 2013b. PISA 2012 Results: What Students Know and Can Do: Student Performance in Mathematics, Reading and Science, vol. 1. Paris: OECD Publishing. ______. 2013c. Technical Report of the Survey of Adult Skills (PIAAC). Paris. OECD/CAF/ECLAC (Organisation for Economic Co-operation and Development​ Development Bank of Latin America/Economic Commission for Latin America and /­ the Caribbean). 2014. Latin American Economic Outlook 2015: Education, Skills and Innovation for Development. Paris: OECD Publishing. Oviedo, A. M., and N. Muller. 2015. “Skills and Labor Market Outcomes in El Salvador.” World Bank, Washington, DC. Schady, N., J. Behrman, M. C. Araujo, R. Azuero, R. Bernal, D. Bravo, F. López Boo, K. Macours, D. Marshall, C. Paxson, and R. Vakis. 2015. “Wealth Gradients in Early Childhood Cognitive Development in Five Latin American Countries.” Journal of Human Resources 50 (2): 446–63. World Bank. 2014. “STEP Skills Measurement Surveys: Innovative Tools for Assessing Skills.” Social Protection and Labor Discussion Paper 1421, Washington, DC. ______. 2015. World Development Indicators (accessed June 3, 2015). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Chapter 3 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? This chapter offers insights into the cognitive and socioemotional skills that may affect employment and labor earnings. It also considers the skills ­ development process, identifying fundamental skills that are valuable for ­ building advanced skills, particularly through tertiary education, and explores the joint effect of multiple skill sets on labor force and tertiary education outcomes. The chapter draws on cross-sectional analyses of Bolivia, Colombia, El Salvador, and Peru, contextualizing the results by referring to studies that use data from high-income countries and other countries in Latin America.1 The range of labor market outcomes discussed is broader than usually considered in similar exercises: Not only labor earnings but also employment and type of employment (salaried versus self-employment, formal versus informal) are assessed. The goal is to take a comprehensive look at the role played by cognitive and socioemotional skills in the labor market. The chapter offers three main findings: • Both cognitive and socioemotional skills are correlated with a range of labor market outcomes. • Different socioemotional skills correlate with different labor market and edu- cation outcomes in the four countries studied (Bolivia, Colombia, El Salvador, and Peru), although cross-country trends are similar. • The skills that emerge as statistically significant, the sign of the correlation, and the order of magnitude are similar, though not identical, in Latin America and countries in the Organisation for Economic Co-operation and Development (OECD). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   27   28 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? Cognitive and Socioemotional Skills Are Correlated with Labor Earnings Greater cognitive abilities are associated with higher labor earnings in Latin America (table 3.1). In Peru various cognitive skills—verbal fluency, verbal ability, math ability, and memory (see table 1.1 for definitions)—are very strongly ­ correlated with labor earnings, even after controlling for demographic and indus- try characteristics. Math abilities are particularly important; verbal ability plays a weaker role. In Bolivia and Colombia, reading proficiency (the ability to read, process, and use written information) is strongly correlated with earnings (math ability and memory data were not collected for these countries). The size and statistical significance of these correlations are stronger than those estimated in a study of young workers (age 25–30) in Argentina and Chile (Bassi and others 2012). That study, which measured analogical reasoning and task planning, did not find significant payoffs for those skills. The new results align with results on high-income countries. Cognitive abili- ties, generally proxied by test scores, significantly contribute to labor earnings in Canada, the United Kingdom, and the United States (Finnie and Meng 2001; McIntosh and Vignoles 2001; Heckman, Stixrud, and Urzúa 2006; Carneiro, Crawford, and Goodman 2007; OECD 2015). For example, a longitudinal study in the United States shows that the net impact of cognitive abilities on earnings ranges from 11 to 30 percentage points, depending on the level of schooling (Heckman, Stixrud, and Urzúa 2006). The magnitude of the returns to basic cognitive skills varies considerably across countries, however: Countries with the highest returns to cognitive skills (the United States, Ireland, and Germany) have Table 3.1 Skills Correlated with Labor Earnings in Bolivia, Colombia, El Salvador, or Peru Type of skill Dimension Disaggregated measures Aggregated measures Basic cognitive Basic academic • Reading proficiency (ability to understand,evaluate, Cognitive skills (reading, knowledge use, and engage with written text) language, and math and • Math ability (basic math operations) ability) reasoning • Verbal ability (vocabulary) Socioemotional Achieving goals • Conscientiousness (tendency to be organized, Stability personality traits responsible, and hardworking) (consistency in • Openness to experience (appreciation for art, motivation, mood, and learning, and unusual ideas) social interactions; • Grit (perseverance and passion for long-term goals) include resilience, Working with • Agreeableness (−) (prosocial behaviors, cooperative agreeableness, and others orientation to others) conscientiousness) Managing • Hostile attribution bias (−) (tendency to perceive emotions hostile intents in others) • Resilience (ability to manage negative emotions) Sources: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010). Note: All skills listed in the table show correlations with labor earnings that were statistically significant (at the 10 percent, 5 percent, or 1 percent levels) for at least one of the four countries studied; see appendix D for details by country. Disaggregated measures were generated using ordinary least squares (OLS) or logit regressions; aggregated measures are based on a structural estimation of latent skills factors and OLS. Associations are positive unless marked (−). Calculations control for a range of characteristics (see appendix D and background papers for details). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 29 returns roughly twice as large as countries with the lowest returns (Sweden, Norway, the Czech Republic) (Hanushek and others 2015). Differences in labor market contexts, such as labor regulations and wage-setting mechanisms, may explain this heterogeneity. All three dimensions of socioemotional skills—achieving goals, working with others, and managing emotions—play significant roles in explaining labor earn- ings in Latin America, but their effect is more nuanced than cognitive skills. Three of the skills associated with achieving goals are positively correlated with labor earnings. Conscientiousness (self-discipline, planning) is positively corre- lated with labor earnings in El Salvador, and grit (effort, persistent interest) is positively correlated with labor earnings in Bolivia and Peru. Both are also the personality traits most associated with higher labor earnings in OECD countries (Nyhus and Pons 2005; Almlund and others 2011). These findings are consistent with the results of a cross-sectional study of young people (25–30) in Argentina and Chile. That study finds that self-efficacy (belief in one’s own abilities), a behavior related to “achieving goals,” has a stronger correlation with higher labor earnings than cognitive skills (Bassi and others 2012). Being more open to experience—being more comfortable with complex, unfamiliar, and ambiguous work situations—is strongly positively correlated with labor earnings in Bolivia, Colombia, and El Salvador, but not in Peru. These mixed effects are in line with evidence from high-income countries, in which the association between this trait and labor earnings varies across contexts and ­ studies. For example, labor earnings increased 4 percent for men (3 percent for women) for every standard deviation increase in openness to experience in the United Kingdom (Heineck 2011). However, a German study finds a positive correlation between openness to experience and labor earnings for women but the opposite effect for men (Heineck and Anger 2010). Some of the measured skills associated with working well with others affect labor earnings in the Latin American sample. Agreeableness is strongly negatively correlated with labor earnings in El Salvador, Peru, and some OECD countries. Peruvian workers who display less kindness and cooperation than average earn higher labor earnings than do kinder and more agreeable workers.2 A similar result is found among German women (Heineck and Anger 2010). In contrast, in the United States more agreeable men and women earn higher labor earnings (Mueller and Plug 2006). Extroversion is the only socioemotional skill that does not correlate with labor earnings for any of the countries in this study; it plays only a minor role in high- income countries, as well. A plausible explanation for this finding is that the reward for extroversion likely depends on occupation: Extroversion may be ben- eficial for sales representatives or teachers but not accountants or scientists (Nyhus and Pons 2005). Two skills underlying the “managing emotions” category are correlated with higher labor earnings. Resilience correlates positively with labor earnings in many Latin American and OECD countries. Workers in El Salvador and Peru who are more resilient earn significantly more than workers who are anxious, reactive, Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 30 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? and pessimistic. Controlling for cognitive skills, several empirical studies in high- income countries find that greater resilience is statistically significant in deter- mining job performance and labor earnings (Bowles, Gintis, and Osborne 2001; Judge and Hurst 2007; Drago 2011). Greater hostility attribution bias (low resilience) is negatively correlated with labor earnings in Bolivia. The labor mar- kets in Colombia do not appear to place the same value on this trait. Given that some socioemotional skills positively correlate with labor earnings and others are negatively correlated, it is perhaps not surprising that the full set of skills only weakly correlates with higher earnings. Of the three countries for which the joint role of socioemotional skills was tested (Bolivia, Colombia, and Peru), only Peru shows a weakly positively correlation with labor earnings. The combined effect of cognitive and socioemotional skills on labor earnings is greater than the effect of either skill set separately (figure 3.1). On average, Bolivian workers who are highly proficient in reading and language (score in the 10th decile) but have weak socioemotional skills (score in the 1st decile) earn 2.4–2.5 Bolivian Sol (Bs) an hour, whereas workers in the 10th decile for both types of skills earn more than Bs 2.5. The joint effect of the two skills types is stronger in Peru. Although the increase of one standard deviation of socioemotional skills only weakly correlates with extra labor earnings, when taking into account its interaction with cognitive skills, it is highly correlated with higher labor earnings. The highest wage earners are in the top decile for both cognitive skills and a set of socioemotional skills. The wage gradient for socioemotional skills is steeper than for cognitive skills, suggesting that workers whose cognitive skills are high but socioemotional skills are low earn less than workers who have lower levels of cognitive skills but exhibit greater stability personality traits (consistency in motivation, mood, and social interactions). The interaction does not emerge in all cases, though. For example, the full set of socioemotional skills does not jointly affect earnings in Colombia (panel b, figure 3.1). Only higher cognitive skills are associated with higher earnings. Socioemotional Skills Are Correlated with Employment and Productive Activity By and large, socioemotional rather than cognitive skills determine labor force participation in Latin America. Cognitive skills do not correlate with employ- ment in the four countries studied, although verbal fluency skills are weakly correlated with increased employment in Peru. Greater cognitive skills are cor- related with higher labor force participation of young adults in Argentina (Bassi and others 2012). Socioemotional skills associated with achieving goals are positively correlated with having a job in Latin America: Workers are more conscientious and have higher degrees of grit than nonworkers in Bolivia, Colombia, El Salvador, and Peru. Conscientiousness also seems to be the main skill associated with young adults’ labor force participation in Argentina and Chile (Bassi and others 2012). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 31 Figure 3.1 Correlation between Labor Earnings and Cognitive and Socioemotional Skills in Bolivia, Colombia, and Peru a. Bolivia b. Colombia Expected hourly income Expected hourly income 2.6 8,000 from main job (in Col$) from main job (in Bs) 7,600 2.5 7,200 2.4 6,800 6,400 2.3 6,000 5,600 2.2 5,200 2.1 4,800 4,400 2.0 4,000 10 10 9 10 9 10 De 8 De 8 c 7 8 9 c 7 8 9 (ac ile of 6 7 (ac ile of 6 7 hie soc 5 6 i l l s hie soc 5 6 i l l s v ing ioe 5 k v ing ioe 5 k oth 4 ve s ls) oth 4 ve s ls) ers m , m goals otio 3 3 4 niti ge skil ers m , m goals otio 3 3 4 niti ge skil 2 cog a 2 cog a ana , wo nal gin rkin skil 1 2 c i l e of d langu ana , wo nal gin rkin skil 1 2 c i l e of d langu ge l De g an ge l De g an mo wit s g din mo wit s g din tion h (rea tion h (rea s) s) c. Peru Expected hourly income 10 from main job (PEN) 9 8 7 6 5 4 3 2 10 De 9 10 cile 8 (co o 7 8 9 nsi f stab 6 5 6 7 ls ste ili 5 k i l and ncy in ty pe 4 ve s ) rso 3 3 4 niti guage soc m ial otiva nalit 2 cog n inte tio y 1 2 c i l e of ory, la rac n, m traits De , mem tion oo th s) d, (ma Source: Bolivia and Colombia: STEP Household Surveys (2012); Peru: ENHAB (2010). Note: Simulations are based on structural estimations of latent skills factors, using Sarzosa and Urzúa (2016). The socioemotional-cognitive skill combinations that share a wage range are denoted in the same color in each graph. This finding mirrors those from high-income countries, where conscientiousness appears to have a large and positive effect on labor participation in the United States and Germany.3 Other socioemotional skills are barely correlated with labor force participa- tion in Latin America. None of the skills associated with “working with others” correlates with the likelihood of holding a job (table 3.2). In the “managing emo- tions” category, only Bolivians who score low in agreeableness and Salvadorans who have difficulty making decisions are more likely to be in the labor force. A combination of a set of socioemotional skills (extroversion and openness to experience) is associated with higher employment in Peru, but the correlation is not significant in Bolivia or Colombia. Similarly to the findings for Peru, extroversion (a component of working with others) and resilience (a component of managing emotions) are correlated with Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 32 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? Table 3.2 Skills Correlated with Employment in Bolivia, Colombia, El Salvador, or Peru Type of skill Dimension Disaggregated measures Aggregated measures Cognitive skills Basic academic • Verbal fluency (how quickly and easily None knowledge and individuals access words from memory) reasoning Socioemotional Achieving goals • Conscientiousness (tendency to be Plasticity personality traits skills organized, responsible, and hardworking) (striving toward personal • Grit (perseverance and passion for growth; include long-term goals) extroversion and openness Working with others None to experience) Managing emotions • Hostile attribution bias (−) (tendency to perceive hostile intents in others) • Decision making (−) (How individuals approach decision situations) Sources: Bolivia and Colombia: STEP Household Surveys (2012): El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010). Note: All skills listed in the table show correlations with labor earnings that were statistically significant (at the 10 percent, 5 percent, or 1 percent levels) for at least one of the four countries studied; see appendix D for details by country. Disaggregated measures were generated using ordinary least squares (OLS) or logit regressions; aggregated measures are based on a structural estimation of latent skills factors and OLS. Associations are positive unless marked (−). Calculations control for a range of characteristics (see appendix D and background papers for details). employment status in high-income countries, though significant results emerge in only a few studies. Both are strongly correlated with male labor participation in Germany and the United States (Barrick and Mount 1991; Heckman, Stixrud, and Urzúa 2006; Wichert and Pohlmeier 2010). For example, a 30-year-old man in the United States who is in the 75th percentile for resilience is 15 percent more likely than a man in the 25th percentile to be employed (Heckman, Stixrud, and Urzúa 2006). Even when considering the joint effect of cognitive and socioemotional skills, cognitive skills have little effect on the likelihood of employment. Among people in Peru with low socioemotional skills, for example, a person with low cognitive skills is as likely to be employed as a person with high cognitive skills (figure 3.2). Both socioemotional and cognitive skills play roles in engaging in productive activities, which includes working, looking for a job, and studying. In Colombia activity rates are higher among people with higher cognitive or (especially) socio- emotional skills (figure 3.3). Given the lack of correlation between cognitive skills and employment and the small number of unemployed people in the sample, students are probably driving this correlation. This finding suggests that students with strong cognitive skills are more likely to stay active if they also have strong socioemotional skills. Both Types of Skills Are Correlated with Job Type Not only may certain skills be correlated with being employed or otherwise active, they also may determine the type of job that a person may hold. The skills required to be a formal sector worker may differ from those needed for informal Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 33 Figure 3.2 Correlation between Employment and Cognitive and Socioemotional Skills in Peru 81 Probability of employment (percent) 77 73 69 65 61 57 53 49 45 10 9 8 10 De cile 7 9 (str of 6 8 ivin plasti 7 g to city 5 wa per 6 rd p s ills ers onalit 4 5 e sk ona y tr 3 g nitiv e) l gr aits 4 fc o uag ow ile o lang th) 2 2 3 Dec e m ory, th, m 1 (ma Source: Peru: ENHAB (2010). Note: Simulations are based on structural estimations of latent skills factors, using Sarzosa and Urzúa (2016). The socioemotional-cognitive skill combinations that share a wage range are denoted in the same color in each graph. sector jobs. Similarly, those who select self-employment may have different strengths than those who are wage employees. Cognitive and Socioemotional Skills Increase the Likelihood of Working in the Formal Sector In Bolivia, Colombia, and Peru, greater cognitive ability is associated with a greater likelihood of holding a job with benefits. Verbal ability and reading proficiency are the driving factors in all three countries (table 3.3). No OECD ­ comparisons can be presented, because the OECD studies do not explore the role of skills in determining formal employment. Socioemotional skills are weakly correlated with formal sector employment. In Bolivia, formal workers have higher levels of all the socioemotional skills included in the analysis, as compared to informal workers. However, no single skill can be identified as the driving factor for informality. Formal sector workers in El Salvador display greater goal achievement (conscientiousness and openness to experience) and ability to manage emotions (resilience) than their informal sector peers, although the correlations are weak at best. The interactive effects of the two skills types may or may not correlate with the probability of being a formal sector worker. Both socioemotional and cogni- tive skills are correlated with a higher probability of formal sector employment in Bolivia, with socioemotional skills playing a particularly important role Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 34 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? Figure 3.3 Correlation between Any Productive Activity (Working, Looking for Job, or Studying) and Cognitive and Socioemotional Skills in Colombia 96 Probability of working, looking for job, 94 92 or studying (percent) 90 88 86 84 82 80 78 10 9 10 8 9 7 8 (ac Decile hie 6 7 vin of so ills gg cio 5 6 e sk )) ma oals, emot 5 g nitiv e skills nag wo io 4 of co u a g ing rkin nal s 4 cile la n g em g wi kills 3 De g and otio th o 3 din ns) the 2 2 (rea rs, 1 Source: Colombia STEP Household Survey (2012). Note: Simulations based on structural estimations of latent skills factors, using Sarzosa and Urzúa (2016). The socioemotional- cognitive skill combinations that share a wage range are denoted in the same color in each graph. Table 3.3 Skills Correlated with Formal Employment in Bolivia, Colombia, El Salvador, or Peru Type of skill Dimension Disaggregated measures Aggregated measures Basic cognitive Basic academic • Reading proficiency (ability to understand, Cognitive skills (reading, knowledge and evaluate, use, and engage with written text) language, and Math reasoning • Verbal ability (receptive vocabulary) ability) Socioemotional Achieving goals • Conscientiousness (tendency to be organized, Socioemotional skills responsible, and hardworking) (achieving goals, working • Openness to experience (Appreciation for art, with others, managing learning, and unusual ideas) emotions) Working with • Agreeableness (−) (prosocial behaviors, others cooperative orientation to others) Managing • Hostile attribution bias (−) (tendency to emotions perceive hostile intents in others) • Resilience (ability to manage negative emotions) Sources: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); and Peru ENHAB (2010). Note: All skills listed in the table show correlations with labor earnings that were statistically significant (at the 10 percent, 5 percent, or 1 percent levels) for at least one of the four countries studied; see appendix D for details by country. Disaggregated measures were generated using ordinary least squares (OLS) or logit regressions; aggregated measures are based on a structural estimation of latent skills factors and OLS. Associations are positive unless marked (−). Calculations control for a range of characteristics (see appendix D and background papers for details). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 35 (figure  3.4, panel a). Among people in the top cognitive decile, the probability ­ of being a formal sector worker ranges from 10 percent for workers with very low socioemotional skills to 50 percent for people with very high socioemo- tional skills. For people in the lowest decile of socioemotional abilities, higher cognitive skills have barely any impact on their ability to get a formal job. The interactive effects reveal no nuanced patterns in Colombia. Cognitive skills are the only skill set correlated with the probability of being a formal sector worker (figure 3.4, panel b). Socioemotional skills are not correlated with the likelihood of formal sector employment in Peru, but lower levels of socioemotional skills (agreeableness, conscientiousness, and extroversion) are associated with formal employment for each cognitive skill level (figure 3.4, panel c). The highest likelihood of formal employment occurs among workers with the highest cognitive and lowest socio- emotional skills. Figure 3.4 Correlation between Formal Employment and Cognitive and Socioemotional Skills in Bolivia, Colombia, and Peru a. Bolivia b. Colombia 52 Probability of holding a formal job (percent) 50 48 Probability of holding a formal job (percent) 40 44 40 30 36 20 32 10 28 24 0 0 10 10 9 9 De c 8 7 9 10 De cile 8 7 9 10 (ac ile of 6 7 8 (ac o 6 7 8 hie s o 5 6 hie f so 5 6 oth ving cioe 4 5 kills oth ving cioe 4 5 kills ers g m 4 ve s skills) ers g m 4 ve s skills) , m oals otio 3 3 niti , m oals otio 3 3 niti ana , wo nal 2 2 o f cog guage ana , wo nal 2 2 o f cog guage gin rkin skil ile lan gin rkin skil ile lan ge mo g wit s l 1 Dec g and ge mo g wit s l 1 Dec g and tion h din tion h din s) (rea s) (rea c. Peru 70 Probability of holding a formal job (percent) 62 54 46 38 30 10 De 9 cile 8 7 9 10 (str of p 6 7 8 ivin las 5 6 g to ticit 4 5 kills ) wa y pe 4 ve s rd p r 3 niti guage ers sonal 2 2 3 o g f c ry, lan ona ity ile o o l gr trai 1 ow t Dec , mem th) s ( m a th Sources: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010). Note: Simulations are based on structural estimations of latent skills factors, using Sarzosa and Urzúa (2016). The socioemotional-cognitive skill combinations that share a wage range are denoted in the same color in each graph. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 36 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? Skills Play a Limited Role in Sorting Workers into Self-Employed and Wage-Earning Categories in Latin America There is not a consistent pattern of skills that correlate with being a wage earning employee rather than self-employed. People with higher cognitive skills are more likely to be wage-earning employees—rather than self-employed—in Colombia and Peru, but no correlation is evident in Bolivia or El Salvador (table 3.4). Verbal abilities are particularly important in Peru. Only socioemotional skills associated with achieving goals are correlated with being a wage earner rather than self- employed: Colombians who are more open to experience and Peruvians with greater grit are more likely to be wage earners. Decision making, a skill under the dimension of managing emotions, is only modestly important, with a weak cor- relation in Bolivia and no correlation in the others. In contrast, a small body of literature shows that cognitive abilities and personality traits may matter in becoming a successful entrepreneur. In the United States, for instance, people with higher learning aptitudes (advanced cognitive skills, which are not measured in our data), a tendency to “break the rules” (as measured by the degree to which they engaged in illicit and risky activities before the age of 22), and high self-esteem in adolescence (which, like the tendency to break the rules, may be correlated with achieving goals) are more likely to become successful incorporated entrepreneurs (Levine and Rubinstein 2013). The joint effects of both skills groups do not show any correlations with being a wage worker rather than self-employed. This result is perhaps not sur- prising given the weak role skills play in determining wage versus self-employed employment. Table 3.4 Skills Correlated with Wage Employment in Bolivia, Colombia, El Salvador, or Peru Type of skill Dimension Disaggregated measures Aggregated measures Cognitive skills Basic academic • Verbal ability (receptive vocabulary and Cognitive skills (reading knowledge and verbal ability of adult subjects) and language skills) reasoning • Verbal fluency (how quickly and easily individuals access words from memory) Socioemotional Achieving goals • Openness to experience (appreciation for art, Perseverance personality skills learning, and unusual ideas) traits (perseverance of • Grit (perseverance and passion for long-term effort and goals, and goals) consistency of interest) Working with others • None Managing emotions • Decision making (−) (how individuals approach decision situations) Sources: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey (2013); Peru: ENHAB (2010). Note: All skills listed in the table show correlations with labor earnings that were statistically significant (at the 10 percent, 5 percent, or 1 percent levels) for at least one of the four countries studied; see appendix D for details by country. Disaggregated measures were generated using ordinary least squares (OLS) or logit regressions; aggregated measures are based on a structural estimation of latent skills factors and OLS. Associations are positive unless marked (−). Calculations control for a range of characteristics (see appendix D and background papers for details). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 37 Both Skills Types Positively Correlate with Tertiary School Attendance High levels of cognitive skills are related to a high probability of pursuing ter- tiary education in Bolivia, Colombia, and El Salvador (this outcome was not studied for Peru), consistent with findings from high-income countries. This is consistent with a study of 11 OECD countries that shows that in all of them but Belgium, greater cognitive skills have a strong impact on tertiary attendance and that the effect is stronger than the impact of greater socioemotional skills (OECD 2015). This finding may illustrate the fact that selection criteria for tertiary-education institutions rely on high school graduation certificates, grades, or achievements, which capture mainly cognitive skills. Once accepted, socio- emotional skills, such as persistence, may be useful for students to persevere and complete their studies. A wider range of socioemotional skills is correlated with tertiary education in Latin America than in OECD countries. An achieving-goals skill—openness to experience (linked to spontaneity and flexibility)—is correlated with tertiary education in Bolivia, Colombia, and El Salvador (this outcome was not analyzed for Peru). In Colombia and the United States managing emotions skills, such as resilience, are correlated with tertiary education attainment.4 Working with oth- ers skills show mixed effects in the three Latin American countries studied: The correlation between agreeableness and tertiary education enrollment is posi- tive in El Salvador but negative in Colombia. Similarly, higher levels of skills associated with achieving goals—specifically conscientiousness and grit—are the socioemotional skills most highly correlated with tertiary education completion in European samples (Almlund and others 2011).5 Table 3.5 Skills Correlated with Tertiary Education Attendance in Bolivia, Colombia, or El Salvador Type of skill Dimension Disaggregated measures Aggregated measures Basic cognitive Basic academic • Reading proficiency (ability to understand, Cognitive skills (reading knowledge and evaluate, use, and engage with written text) and language skills) reasoning • Math ability (basic math operations) Socioemotional Achieving goals • Openness to experience (appreciation for art, Socioemotional skills learning, and unusual ideas) (achieving goals, Working with others • Agreeableness (−) (prosocial behaviors, working with others, cooperative orientation to others) managing emotions) Managing emotions • Decision making (how individuals approach decision situations) • Hostile attribution bias (−) (tendency to perceive hostile intents in others) • Resilience (ability to manage negative emotions) Sources: Bolivia and Colombia: STEP Household Surveys (2012): El Salvador: El Salvador Skills Survey (2013). Note: All skills listed in the table show correlations with labor earnings that were statistically significant (at the 10 percent, 5 percent, or 1 percent levels) for at least one of the four countries studied; see appendix D for details by country. Disaggregated measures were generated using ordinary least squares (OLS) or logit regressions; aggregated measures are based on a structural estimation of latent skills factors and OLS. Associations are positive unless marked (−). Calculations control for a range of characteristics (see appendix D and background papers for details). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 38 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? When combining both cognitive and socioemotional skills, again the data confirm that cognitive skills play the primary role in tertiary school atten- dance (figure 3.5). In Colombia a person with socioemotional skills in the top decile has a 17 percent higher probability of attending or having attended college than someone in the bottom decile; the increase rises to 71 percent when cognitive skills are considered. A similar pattern is observed in Bolivia. These findings are consistent with the idea that a solid cognitive knowledge base is a prerequisite to effectively transitioning to tertiary education and that certain traits can help maximize the chance of doing so. Socioemotional skills also matter: In Colombia a person in the highest reading proficiency decile but a low socioemotional decile has a 60 percent chance of attending tertiary education; the likelihood rises to 85 percent for a person in the highest decile of the socioemotional scale. Interpretation of Cross-Country Variations Both cognitive and socioemotional skills are correlated with more favorable labor force and tertiary education outcomes in Bolivia, Colombia, El Salvador, and Peru. These findings confirm results from Argentina and Chile and are is overall consistent with findings from the United States and other OECD countries using longitudinal data and establishing causal effects. Different cognitive skills are positively associated with better jobs (jobs that pay more and require higher qualifications). Reading proficiency (in Bolivia, Colombia, and El Salvador) and verbal fluency and math ability (in Peru) are Figure 3.5 Correlation between Tertiary Education Attendance and Cognitive and Socioemotional Skills in Bolivia and Colombia a. Bolivia b. Colombia Probability of having attended tertiary education (percent) Probability of having attented 90 tertiary education (percent) 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 10 10 9 10 9 De 8 9 10 c 7 8 De 8 7 8 9 (ac ile of 6 7 cile hie s 5 6 ills (ac o 6 7 oth vin ocio 5 e sk hie f so 5 6 kills ers g go emo 4 4 nitiv skills) oth v ing cioe 4 5 ive s ) 3 3 f cog guage ers 3 4 gnit ge skills ,m a l ana s, w ona t i 2 2 ile s o n , m goals motio 2 2 3 o f c o ngu a gin ork l sk 1 Dec g and la ana , w n iles gin ork al sk 1 Dec g and la g e ing ills mo wi (rea din g e ing ills din tio t mo wi (rea ns) h tio th ns) Sources: Bolivia and Colombia STEP Household Surveys (2012). Note: Simulations based on structural estimations of latent skills, using Sarzosa and Urzúa (2016). The socioemotional-cognitive skill combinations that share a wage range are denoted in the same color in each graph. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 39 positively correlated with higher labor earnings and formal employment. Cognitive skills do not affect employment in any of the four countries studied, however. Socioemotional skills are associated with employment and a broader range of labor market outcomes. In all four countries (as well as higher-income coun- tries), skills associated with achieving goals (being conscientious and having grit) are the only skills correlated with employment. This finding has also been found in higher-income countries. Nearly all the measured skills that comprise achieving goals, working with others, and managing emotions—namely, open- ness to experience, conscientiousness, agreeableness, resilience, grit, decision making, and hostile attribution bias—are correlated with higher labor earnings and, to a lesser degree, formal employment and wage employment. Only one measured skill that can be classified as working with others (extroversion) was not correlated with any labor market outcome. In Bolivia, Colombia, and El Salvador, as in higher-income countries, socio- emotional skills are correlated with tertiary education. All the measured socio- emotional skills predict whether students seek tertiary education; achieving goals, measured by perseverance, plays a particularly large role. The skills associated with a given outcome vary greatly across countries; the range is particularly wide for socioemotional skills. For example, conscien- tiousness is correlated with labor earnings in El Salvador but not in other countries; it is correlated with employment in all countries but Peru. Although the determinants of these variations cannot be unambiguously untangled, some factors could explain cross-country patterns. First, different cultural contexts may influence both the ways certain behaviors are rewarded in the workplace and the manner in which survey respondents answer questions related to socioemotional skills. Second, the structure of employment differs across countries; given that some skills may not be rewarded equally across occupations (extroversion, for example, probably matters in some fields but not others), the distribution of occupations may result in some skills being valued over others. Third, differences in the methodology to generate the socioemotional skills measures and in the questions used to measure cognitive skills may account for some of the observed differences between the four countries in our sample and with results from OECD and the other Latin America analysis (Bassi and others 2012). Finally, measurement errors could arise from the limited number of questions on socioemotional skills surveyed for Bolivia, Colombia, and El Salvador. Because the data used for this study were observed at a single point in time, they do not allow causal links between skills and labor outcomes to be estab- lished. However, the bulk of evidence from high-income countries use longitudi- nal data and robustly establish causal links between cognitive and socioemotional skills and labor and education outcomes (see, for example, Heckman, Stixrud, and Urzúa 2006; Carneiro, Crawford, and Goodman 2007; Lindqvist and Vestman 2011; Segal 2013; and OECD 2015). Although the contexts Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 40 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? in high-income countries are different from those in low- and middle-income countries, the result in high-income countries suggests that better skills could cause better labor market outcomes. In addition, the psychology literature suggests that socioemotional skills, ­ particularly skills that are measured in the surveys used here, are relatively fixed in adult populations. As these samples are randomly distributed across the work- ing age population, the skills measured today are likely to be very similar (at least in terms of rank-order) to the skills respondents had when they established their work patterns. Although our results do not establish a causal relationship, evidence from other studies suggests that the estimated correlates can be interpreted as signal- ing that both cognitive and socioemotional skills are inputs in the labor market success process. They should be fostered, using skill development strategies described in chapter 4. Notes 1. The results in this chapter are drawn from the following background papers prepared as part of this study: Barón, Sarzosa, and Mola for Bolivia (2015); Acosta, Muller, and Sarzosa (2015) for Colombia; Oviedo and Muller (2015) for El Salvador; and Cunningham, Parra Torrado, and Sarzosa (2016) for Peru (for descriptions of these papers, see appendix A). The empirical analysis uses two different methods in order to present a fuller picture of the relationship between skills and work. Method A allows a nuanced exploration of the different types of skills, but at the cost of estimation bias. Method B corrects for measurement error but aggregates variables of ­ interest. Appendix B describes both methods (for details, see the background papers). For ease of presentation, we do not distinguish between the methodologies used to derive each result in this chapter. Appendix D reproduces the regression results reported in the background papers. 2. The data from Peru generate two variables (“kindness” and “cooperation”) that are similar to the Goldberg Big Five trait of “agreeableness.” One might argue that people who are more agreeable tend to sort themselves into the caring professions, which tend to be lower paid. The case study of Peru controls for industry of employment. 3. Counterintuitively, a higher score for openness to experience—a behavior within the achieving goals category—is negatively correlated with employment in Germany; this skill does not emerge as significant in other studies. 4. Heckman Stixrud, and Urzúa (2006) explore how self-esteem and self-control, both facets of resilience, affect college graduation. According to them, the relationship between resilience and schooling does not appear to be monotonic in the United States. 5. Psychology research shows that self-discipline and grit are crucial determinants of adolescent academic success in the United States. Several measures of self-discipline outperform IQ as a predictor of the academic performance of adolescents. Grit is correlated with a range of schooling outcomes, such as educational attainment, grades, and retention (Duckworth and others 2007). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? 41 References Acosta, P. A., N. Muller, and M. Sarzosa. 2015. “Beyond Qualifications: Returns to Cognitive and Socio-Emotional Skills in Colombia.” Policy Research Working Paper 7430, World Bank, Washington, DC. Almlund, M., A. L. Duckworth, J. J. Heckman, and T. Kautz. 2011. “Personality Psychology and Economics.” In Handbook of the Economics of Education, vol. 4, ed. E. A. Hanushek. Amsterdam: North Holland. Barón, J. D., M. Sarzosa, and J. Mola. 2015. Ethnicity and Labor Market Returns to Cognitive and Socio-Emotional Skills in Urban Bolivia. World Bank, Washington, DC. Barrick, M. R., and M. K. Mount. 1991. “The Big Five Personality Dimensions and Job Performance: A Meta-Analysis.” Personnel Psychology 44 (1): 1–26. Bassi, M., M. Busso, S. Urzúa, and J. Vargas. 2012. Disconnected: Skills, Education and Employment in Latin America. Washington, DC: Inter-American Development Bank. Bowles, S., H. Gintis, and M. Osborne. 2001.“Incentive-Enhancing Preferences: Personality, Behavior, and Earnings.” American Economic Review 91 (2): 155–58. Carneiro, P., C. Crawford, and A. Goodman. 2007. “The Impact of Early Cognitive and Noncognitive Skills on Later Outcomes.” CEE DP 92, Centre for the Economics of Education, London School of Economics, London. Cunningham, W. M. Parra Torrado, and M. Sarzosa. 2016. “Cognitive and Non-Cognitive Skills for the Peruvian Labor Market: Addressing Measurement Error through Latent Skills Estimations.” Policy Research Working Paper 7550, World Bank, Washington, DC. Drago, F. 2011. “Self-Esteem and Earnings.” Journal of Economic Psychology 32 (3): 480–88. Duckworth, A., C. Peterson, M. Matthews, and D. Kelly. 2007. “Grit: Perseverance and Passion for Long-Term Goals.” Journal of Personality and Social Psychology 92 (6): 1087–101. Finnie, R., and R. Meng. 2001. “Minorities, Cognitive Skills, and Incomes of Canadians.” Canadian Public Policy 28 (2): 257–73. Hanushek, E. A., G. Schwerdt, S. Wiederhold, and L. Woessman. 2015. “Returns to Skill around the World: Evidence from PIAAC.” European Economic Review 73: 103–30. Heckman, J. J., J. Stixrud, and S. Urzúa. 2006. “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior.” Journal of Labor Economics 24 (3): 411–82. Heineck, G. 2011. “Does It Pay to Be Nice? Personality and Earnings in the U.K.” Industrial and Labor Relations Review 64 (5): 1020–38. Heineck, G., and S. Anger. 2010. “The Returns to Cognitive Abilities and Personality Traits in Germany.” Labour Economics 17 (3): 535–46. Judge, T. A., and C. Hurst. 2007. ‘‘Capitalizing on One’s Advantages: Role of Core Self- Evaluations.” Journal of Applied Psychology 92 (5): 1212–27. Levine, R., and Y. Rubinstein. 2013. “Smart and Illicit: Who Becomes an Entrepreneur and Does It Pay?” NBER Working Paper 19276, National Bureau of Economic Research, Cambridge, MA. Lindqvist, E., and R. Vestman. 2011. “The Labor Market Returns to Cognitive and Noncognitive Ability: Evidence from the Swedish Enlistment.” American Economic Journal: Applied Economics 3 (1): 101–28. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 42 Do Skills Affect Labor Market and Tertiary Education Outcomes in Latin America? McIntosh, S., and A. Vignoles. 2001.“Measuring and Assessing the Impact of Basic Skills on Labour Market Outcomes.” Oxford Economic Papers 53 (3): 435–81. Mueller, G., and E. J. S. Plug. 2006. “Estimating the Effect of Personality on Male and Female Earnings.” Industrial and Labor Relations Review 60 (1): 3–22. Nyhus, E. K., and E. Pons. 2005. “The Effects of Personality on Earnings.” Journal of Economic Psychology 26: 363–84. OECD (Organisation for Economic Co-operation and Development). 2015. Skills for Social Progress: The Power of Social and Emotional Skills. OECD Skills Studies. Paris: OECD Publishing. Oviedo, A. M., and N. Muller. 2015. “Skills and Labor Market Outcomes in El Salvador.” World Bank, Washington, DC. Sarzosa, M., and S. Urzúa. 2016. “Implementing factor models for unobserved heterogene- ity in Stata: The heterofactor command.” Stata Journal 16 (1): 197–228. Segal, C. 2013. “Misbehavior, Education, and Labor Market Outcomes.” Journal of the European Economic Association 11 (4): 743–79. Wichert, L., and W. Pohlmeier. 2010. “Female Labor Force Participation and the Big Five.” ZEW Discussion Paper 10-003, Zentrum für Europäische Wirtschaftsforschung (Center for European Economic Research), Mannheim, Germany. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Chapter 4 Policy and Programming for Socioemotional Skill Development The previous chapter showed that both socioemotional and cognitive skills are necessary in Latin America for a job skills development strategy. While there is a large literature on methods for cognitive skill instruction, the literature is much more limited on methods for acquiring socioemotional skills. Further, some con- cepts from the socioemotional skills literature may be useful to advance and refine the methods for teaching cognition. It may be necessary to expand the job preparation process beyond existing cognitive skills development frameworks into new socioemotional and cognitive learning models, using a broader range of actors. The guiding framework for this chapter is the concept of readiness. It refers to the notion that skills can be developed only when an individual is neurobio- logically, psychosocially, and situationally ready to learn them (Guerra, Modecki, and Cunningham 2014). Just as it is not possible to teach one-year-olds to write (because they cannot manipulate a pencil or computer or understand the concept of writing), it is not possible to teach very young children social problem solving (because they do not yet have the psychosocial wiring needed for empathy). In addition to readiness, a skills strategy should consider the different actors who can help develop skills at different ages and in various contexts (Cunningham and Villaseñor 2016). Parents and family members are the primary teachers of very young children (age 0–2). Early childhood programs boost the skills-formation process for children 3–5. For school-age children, family and school are the dominant teachers, but peers also start to have an influence. Peer influence grows and parental influence fades in adoles- cence (13–16), while schools continue to play a strong role in skills develop- ment. Once an individual leaves school, skills are learned in the workplace and in new family settings.1 This chapter examines the concept of readiness for developing socioemo- tional and cognitive skills that are relevant for the labor market. It identifies methods for teaching these skills and the appropriate teachers at different stages. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   43   44 Policy and Programming for Socioemotional Skill Development It describes some successful programs to illustrate how such socioemotional skills can be developed (appendix F provides an inventory of 27 programs that aim to foster socioemotional skills, from the early years to young adulthood, from Latin America and other regions) and provides some examples of flexible teaching methods. The chapter’s primary conclusion is that socioemotional skills can, and have been, successfully taught through public sector interventions across the world. Cognitive skill development can also be enhanced by taking account of readiness and by practicing socioemotional skills in the cognitive skills learning process. Translating Research Findings into Policy-Relevant Concepts To begin, we map into a single set of definitions the skills used in the analysis, the skills that employers value, and the skills that are commonly the focus of programming. We use data from employer surveys that identify high-priority socioemotional skills to construct a taxonomy of skills that reflect values and behaviors, are assets in the labor market, and can be taught at key stages of the life cycle. This classification consists of eight categories—social problem solving, resilience, achievement motivation, self-control, teamwork, initiative, confidence, and ethics—summarized by the acronym PRACTICE (table 4.1). Table 4.1  Definitions of PRACTICE Skills Skill Definition (Social) problem solving Set of information-processing skills that determine how individuals solve social problems, such as attention to relevant cues, interpretation of cues and emotional reactions, goal setting and planning, access to behavioral responses from memory, evaluation of responses, decision making, behavioral enactment, and reflection. Resilience Ability to bounce back from adversity and thrive in the context of risk. Encompasses ability to realistically connect future goals and opportunities to one’s own abilities and to adapt as needed to situational constraints. Achievement motivation Orientation toward success, mastery, and sense of purpose. Associated with high degree of independence and capacity and drive to pursue difficult tasks and work toward desired goals. Control (self-control) Range of self-regulatory skills that allow individuals to modulate and restrain their impulses, including ability to focus attention, delay gratification, and inhibit impulsive responding. Teamwork Skills involved in getting along with others, understanding their feelings and points of view, communicating effectively, being helpful and agreeable, and not engaging in aggressive or bullying behaviors. Initiative Personal agency and internal locus of control; belief that outcomes depend on one’s own actions rather than fate, chance, or others. Linked to enterprise, taking charge, follow-through, determination, and leadership. Confidence Beliefs and feelings about one’s abilities, a realistic self-concept, and self-esteem. Ethics Strength of character, social responsibility, and principled behavior, including being honest and fair, following rules, following through on actions, and behaving responsibly. Source: Guerra, Modecki, and Cunningham 2014. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 45 Each of the PRACTICE skills can be mapped to the dimensions of socio- emotional skills analyzed in chapter 3. The first column of table 4.2 presents the skills dimensions that were linked to labor market success in chapter 3. The second column presents the corresponding PRACTICE skill defined by employers. The third column presents some specific skills associated with each category, which are the focus of policies. The table also presents the appropriate age for the development of the eight PRACTICE socioemotional skills sets, based on the concept of readiness as it is derived from the psychol- ogy literature. For example, (social) problem solving is best developed during middle childhood (age 6–11) and adolescence (age 12–18), with foundational Table 4.2 Optimal Stages of Development of PRACTICE Skills Stage of development and key actors PRACTICE Age: Dimension of taxonomy of 6–11 Age: 19–29 socioemotional socioemotional Age: 0–5 (parents, Age: 12–18 (school, family, skills skills Subskills (parents) school) (school, peers) workplace) Achieving goals Achievement Mastery orientation Optimal Reinforcement motivation Sense of purpose Motivation to learn Ethics Honesty Foundational Optimal Optimal Fairness orientation Moral reasoning Initiative Agency Optimal Optimal Optimal Optimal Internal locus of control Leadership Problem- Social-information Foundational Optimal Optimal Reinforcement solving processing skills Decision making Planning skills Working with Teamwork Empathy/prosocial Optimal Optimal Reinforcement others behavior Low aggression Communication skills Relationship skills Managing Confidence Self-efficacy Foundational Optimal Optimal Reinforcement emotions Self-esteem Positive identity Control Delay of gratification Optimal Optimal Optimal Reinforcement Impulse control Attentional focus Self-management Resilience Stress resistance Optimal Optimal Reinforcement Perseverance Optimism Adaptability Source: Guerra, Modecki, and Cunningham 2014. Note: “Foundational” refers to the basis for the core skill building that takes place in a subsequent period. “Optimal” refers to periods of maximum sensitivity when it is easiest for individuals to acquire specific skills. “Reinforcement” means that intense practice is needed for the skills to be mastered. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 46 Policy and Programming for Socioemotional Skill Development skills developed in early childhood (age 0–5) and reinforcement opportuni- ties in early adulthood (age 19–29).2 Three findings emerge from this exercise. First, middle childhood is an opti- mal period for developing all of the PRACTICE skills: Children have the neces- sary neurobiological wiring and psychosocial maturity and are in an environment that encourages practice and learning. Second, adolescence is also a period of optimal skills development, allowing for more complex development of skills initiated in middle childhood. The finding suggests that socioemotional develop- ment efforts should continue through the high school years and should not be limited to at-risk youth, as is the current common practice. Third, schools are an ideal place to teach these skills, because they are the public institutions that have greatest influence over children and youth during this period. These findings highlight the importance of keeping children in school through adolescence. Socioemotional Learning through the Life Cycle The formation of skills is a cumulative process. Because it is affected by the envi- ronment and investments, programs for developing socioemotional skills are best implemented at particular times in the life cycle. Early Years (0–5): Shaping the Foundations in a Safe, Structured Context Important foundational socioemotional skills are developed in the early years. They include social problem solving when playing with peers; impulse control; and the development of confidence, a set of ethics, and the ability to work with others. The young brain is highly malleable and is being actively wired in this stage, creating a foundation for later skill development (Kautz and others 2014). Much of the positive brain wiring before the age of three occurs through interactions with a trusted, supportive caregiver. Children who have a secure attachment to a supportive caregiver learn to trust others; by the age of two or three, they begin socializing with peers and learning how to navigate social inter- actions. Because young children’s behaviors are controlled by external guidance rather than internal motivation, they need contexts that allow them to observe, model, and be rewarded for developmentally appropriate skills that are a founda- tion for future skill building (Guerra, Modecki, and Cunningham 2014). Given the importance for very young children of connection with a caregiver and the need for a scaffolding within which to develop foundational skills, two sets of interventions are particularly effective. Parental support programs teach caregivers how to connect with their young children in a positive manner and to provide a structured context in which the infants or toddlers can begin developing foundations of socioemotional (and cognitive) skills. The support may be center-based or the program may send child development experts to the house to teach parents how to use their own context to stimulate their infants and toddlers. Programs to enhance parental learning and encouragement of early stimulation and nutrition have increased the acquisition of cognitive skills and socioemotional skills in young children (Kagitcibasi, Sunar, and Berkman 1988; Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 47 Gertler and others 2014), although effects are more limited for high-risk chil- dren (Peacock and others 2013). Children between the ages of three and five thrive in high-quality, child- centered daycare and preschools that focus on improving personality traits and managing externalizing behaviors while also teaching basic cognitive skills. Such programs have shown positive results in employment, wages, and positive behaviors for more than 30 years after program participation (Schweinhart and others 2005). Much has been written in the psychological and economics literature about the benefits of early investment in high- quality preschool programs, such as the enriched Head Start and the Perry Preschool Program, particularly when followed by later enrichment programs (Cunha and Heckman 2008). Through well-structured and responsive envi- ronments that include low student-teacher ratios, child-centered activities, and opportunities for cooperative learning and structured play, these settings can promote a range of skills, including the PRACTICE skills control, team- work, and initiative. Indeed, many effective preschool programs specifically target these skills (Schweinhart and others 2010). A number of evidence-based programs currently are available to directly build socioemotional skills in preschool-age children. They often address skills related to social interactions and teamwork, including communication and relationship skills, empathy, prosocial behavior, and prevention of aggression. Some of these programs, such as Second Step, focus on self-control and initiative. Others, such as Tools of the Mind, target only self-control skills (box 4.1). Middle Childhood (6–11): Experimenting, Practicing, and Entrenching Socioemotional Skills The peak target age for socioemotional skill development is 6–11, the age group for which the evidence for program effectiveness is greatest (Durlak and others 2011). This is a prime period for impacting all of the PRACTICE skills. Beginning around age six or seven, the brain becomes increasingly coherent. Frequently used brain connections become stronger, leading to the formation of habitual social, emotional, and behavioral responses. Children 6–11 improve their ability to manipulate concrete information (they still cannot think abstractly), and their personality traits start to crystalize (Shiner 2000). They still need supportive and structured contexts, but they also need freedom to experiment and practice newly acquired skills so that they become habitual. Although parents still play a significant role in supporting socioemotional and cognitive skills development in middle childhood, schools can—and should—play a significant role. They are particularly well-suited for socioemotional skill devel- opment for this age group, for several reasons. Primary schooling is nearly universal in Latin America, granting access to the entire middle childhood population; the structured and habitual school day allows for entrenching habits; children typi- cally spend time in a single classroom with a single teacher and the same group of peers for an entire school year; and this single point of entry renders programs easier to implement, less costly, and more likely to have consistent effects. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 48 Policy and Programming for Socioemotional Skill Development Box 4.1 Promoting Self-Regulated Learning in Young Children through the Tools of the Mind Program Tools of the Mind is an early childhood education program that uses teaching and classroom strategies to promote intentional and self-regulated learning in preschool and kindergarten children, with the aim of equipping them to solve problems and create solutions in the mod- ern world. The program is based on the work of Russian psychologist Lev Vygotsky, whose theory (known as cultural-historical theory) maintains that child development is the result of interactions between children and their social environment. Tools of the Mind aims to scaffold children so that they become “masters of their own behavior” by promoting the development of social and cognitive competencies that allow them to be deliberate, self-regulated learners interested in the process of learning, in doing well, and in other students. It considers cognitive self-regulation (paying attention, remember- ing, and moving flexibly between learning tasks) to be integral to the quality and quantity of academic learning. Academic content, such as literacy and numeracy, is a means for practice, not the sole goal of learning; educational goals are defined more broadly as child achieve- ment, engagement, and social competence. Program elements to promote self-regulation consist of preschool and kindergarten curricula, a teaching approach, and a professional development program for teachers. Teachers are also trained to use continuous dynamic assessment to scaffold learning and individualize instruction. The program operates primarily with at-risk children in urban areas in the United States, Canada, and Chile. An impact evaluation (Blair and Raver 2014) finds that children participat- ing in Tools of the Mind are better at focusing attention in the face of distraction and have better working memory (both key components of executive function) than children in control classrooms, with differences especially pronounced in schools in high-poverty areas. These gains carried into first grade, where students from Tools of the Mind classrooms scored higher in reading and vocabulary than control students. Sources: Diamond and others 2007; Bodrova and Leong 2015. While socioemotional skills development often occurs outside of the primary school curriculum, several types of interventions have been tested in school set- tings. They can be classified into four groups: developing teacher’s socioemo- tional skills, improving the school climate, incorporating the development of socioemotional skills into the teaching of other subjects, and socioemotional curriculum. All of these methodologies can develop the eight PRACTICE skills. Developing teachers’ socioemotional skills For teachers to incorporate socioemotional learning into the classroom, they must possess the skills themselves. This type of intervention focuses on develop- ing socioemotional skills in teachers and principals so that they can model them in the classroom. During middle childhood, children observe and practice adult behaviors. Teachers’ and administrators’ methods for resolving conflict, working in teams, Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 49 expressing self-confidence, and practicing other socioemotional skills are therefore important. The intervention teaches teachers to strengthen their socioemotional skills and use (and model) them in classroom and school settings. There is little empirical evidence on the effectiveness of teaching teachers to model positive socioemotional skills. The government of Peru piloted a one-year course to teach teachers and principals to do so. Box 4.2 presents the model and the initial results. Box 4.2  Developing Teachers’ Socioemotional Skills through Peru’s Escuela Amiga Program The Peruvian Ministry of Education’s Escuela Amiga (Friendly School) program takes a three- pronged approach to build socioemotional skills in children in poor neighborhoods. One prong is a year-long training session for teachers and principals. The two-semester university- level course was designed to develop participants’ socioemotional skills so that they could apply them to interactions at school. The course was taught by university staff in the psychol- ogy department. A diploma was awarded upon graduation. Nearly 15,000 classroom teachers and 81 principals participated in the program’s first year (2013), meeting for 10 hours every week, for a total of 380 hours of class time over two semes- ters. Most were mid-career professionals who were working in schools while enrolled in the course. The course was designed to teach teachers and instructors to recognize and develop their own socioemotional skills—empathy, tolerance, self-regulation, and social decision making, among others—by drawing from their personal experiences and methodologies for using these skills in the classroom. The course was delivered through lectures, paper exercises, role- playing, and group interactions to recognize and manage a range of skills. Participants also formed informal study groups to provide support to one another throughout the week to apply the course lessons to their daily professional challenges. The course incorporated real- time issues in the curricula, drawn from participants’ work lives. Once the course was com- pleted, the program’s roving support teams visited schools regularly to continue supporting teachers and principals in applying the course tools. An impact evaluation of the program is underway, but exit surveys suggest that it helped build knowledge of socioemotional skills and how to use them in a school setting: 90 percent of participants felt that they were better able to manage their classrooms, 93 percent believed that they were better equipped to manage conflict at school, and 50 percent felt that their professional relationships had improved. Informal interviews with teachers and principals con- sistently found that the primary benefit from the course was that better understanding of one- self facilitated better management of the classroom; participants also reported learning how to better manage their own children, marriage, neighbors, and self-esteem. Teachers were initially reluctant to commit so much time to the diploma, but in retrospect they found that it yielded a high return on investment. Source: Paredes 2014. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 50 Policy and Programming for Socioemotional Skill Development Improving the school climate Children and youth need safe places to practice and reinforce socioemotional skills. Physical safety is crucial, but cultural safety that allows for experimentation and different behaviors is also a basis for effective learning. A positive school climate is one in which school staff reach out to all students (not only those at risk), clear rules are developed and followed by students and staff, reform rather than punishment is used to manage behaviors that are contrary to school norms, rewards are regularly used to recognize positive behaviors, and special services are provided to students with particular challenges. The School-Wide Behavioral Support Model is an example of a process used around the world to create school climates that are conducive to learning (box 4.3). Incorporating the development of socioemotional skills into the teaching of other subjects Teachers can incorporate socioemotional learning into standard course material by altering their pedagogical methods. The global shift away from passive learn- ing (in which students quietly listen to information communicated by a teacher) toward active learning (in which students learn through discovery, facilitated by a teacher) is a step in the right direction. Having students work in groups, present their results, and conduct project-based problem solving and exploration—gives an opportunity to use the PRACTICE skills of teamwork, problem solving, con- fidence, and achievement motivation (box 4.4). Using a socioemotional curriculum Some schools have a socioemotional curriculum, materials, teaching method- ologies, and dedicated time for developing socioemotional skills. This type of intervention is the closest to treating the teaching of socioemotional skills on a par with the teaching of other subjects, such as history or language. The struc- ture of the course may be similar to that of a traditional subject—a dedicated time each week, with classroom instruction, exercises to practice the skills being learned, grades and feedback, and evaluation. More commonly, a scaled-down version of such a course is delivered. For example, the course may be taught once a week rather than daily, or it may dispense with evaluation and feedback. It may be broad-based or focus on a small set of skills, such as self-regulation or self-esteem. It may be taught by teachers or by school psychologists. Box 4.5 describes the Incredible Years model, which is widely used in elementary schools in the United States. Linked to a socioemotional curriculum is evaluation of socioemotional skills, regardless of where or how they are taught in the school. For example, the Knowledge Is Power Program (KIPP) includes a “character” report card, in which all teachers who work with a student assess the student’s success in demonstrating a range of socioemotional skills. The teacher feedback is aggregated into a score for each skill and reported in the character report card. During student perfor- mance reviews with students and parents, teachers discuss the character report card in the same way they discuss the standard report card, assessing student Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 51 Box 4.3 Improving the School Climate through the School-Wide Positive Behavior Support Model The School-Wide Positive Behavior Support (SWPBS) model is a set of procedures and organi- zational systems intended to establish a school culture of positive behavior, complemented by additional support for students who need more than a positive school climate. It is not a cur- riculum but a definition of school norms, practices for reinforcing positive behavior, total school support (not just support by teachers), and data-driven decision making. The SWPBS model consists of three tiers of intervention. The first tier is largely about improving the school climate. Certain activities are implemented across the entire school to teach all students a set of behaviors. They include the following: • Behavioral expectations. The school community adopts school rules that are to be expected. Deviation from the rules may result in a more individualized intervention. • Rewards for appropriate behavior. All school staff—principals, teachers, bus drivers, janitors—recognize the importance of adhering to school rules and offer rewards on ­ ­ occasion. Rewards may be small, such as free entry to a school event or permission to not wear a uniform to school for a day. • Continuum of consequences for problem behavior. Problem behavior is not tolerated; it is managed in a positive manner. Group-based support to assist with structure and help man- age the social and academic aspects of school is provided for people who do not follow school rules. Individualized behavioral assessments are conducted and skills development provided to students with persistent challenges. • School-wide classroom management practices. The consequences of not adhering to school rules are widely known, and nonadherence is managed in a consistent manner across class- rooms. Because classroom management processes are well-defined, students know the consequences of their actions and teachers spend less time managing problem behavior. Randomized controlled trials show that the SWPBS positively affects student behavior. A  three-year randomized trial (using waiting lists to identify control groups) found that the model decreased disciplinary problems and increased third-grade reading performance (Horner and others 2009). A five-year longitudinal study in 37 elementary schools found that problem behavior was lower in SWPBS schools than in controls (Bradshaw, Mitchell, and Leaf 2010). Similar results were found in a 63-school study in Illinois and a study in Hawaii. progress in mastering the socioemotional skills and identifying which skills are lagging and strategies for addressing shortfalls (Angrist and others 2010). A number of individual studies as well as recent comprehensive reviews sup- port the effectiveness of school-based socioemotional programs (CASEL 2013). A meta-evaluation of findings from 213 school-based universal social and emo- tional learning programs involving more than 270,000 participants in different types of programs found that participants demonstrated significant improve- ments in a range of social and emotional skills and in academic achievement (Durlak and others 2011). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 52 Policy and Programming for Socioemotional Skill Development Box 4.4 Incorporating Socioemotional Skills into the Teaching of Other Subjects: Facing History and Ourselves Facing History and Ourselves is a training program for teachers of middle and high school students (age 12–18) that incorporates socioemotional learning and moral reasoning into his- tory instruction. Teachers are provided with professional development classes, resource books, coaching, and mentoring to help them apply new approaches to the study of the historical events that led to World War II, the Holocaust, the Chinese Cultural Revolution, and apartheid in South Africa, to teach students history and learn to link the lessons of history to their every- day lives while also developing independent and analytical thinking, self-efficacy, and empa- thy. Teachers are supported to mentor students to listen to one another, consider one another’s perspectives, connect personally with the content, take intellectual risks, and learn to form judgments based on a critical analysis of evidence. The program is implemented in more than 110 countries, through 90,000 trained educators, who reach nearly 1.9 million students a year. An evaluation was undertaken in 2007/08 in 76 randomly assigned schools to assess the impact of the program on teacher satisfaction and student learning of history, higher-order cognitive skills, and tolerance. The study sample included 134 teachers and 1,371 of their 10th grade students; two-thirds of the schools were academically low-performing. Facing History had a statistically significant impact on student capacity to analyze evidence from multiple sources; prosocial behavior, tolerance, and empathy; and knowledge of history. It increased teacher confidence in creating a student-centered classroom. Facing History class- rooms were more inclusive, respectful, and tolerant of differing viewpoints. “It really changed the way I was teaching and framing my lessons. It helped me to put the students in a position in which they are the deciders, trying to figure things out for themselves,” reported a teacher in China. Source: Barr 2010. Adolescence (12–18): Nuancing and Developing Complex Socioemotional Skills By the time children reach adolescence, the foundations for socioemotional skills should have been built. The social context is appropriate to develop complex socioemotional skills while still acquiring basic cognitive skills. In this period children experience a rapid growth in cognitive abilities that permit abstract thinking, complex reasoning, and better decision making (Kuhn 2006). These gains are challenged by the slower development of the part of the brain that regulates impulse control, risk taking, and peer pressure. Adolescents have more autonomy in decision making than younger children and increasingly choose their contexts. Although there has been considerable discussion about the potential ineffi- ciency of investment in skill building during adolescence, the psychology litera- ture suggests that it is a time of increased cognitive capacity, greater independence, and changing roles that support strategic investment. There is growing awareness of the need to develop targeted socioemotional skill-building Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 53 Box 4.5  Using a School Curriculum: The Incredible Years Program The Incredible Years program is a lesson-based program designed for elementary school (and younger) children and their parents that provides information on and practice in social competence, emotional self-regulation, persistence, problem solving, anger manage- ment,  and prosocial behavior with peers and teachers. The Dina Dinosaur Social Skills and Problem-Solving Curriculum of the Incredible Year program offers 20- to 30-minutes lessons twice a week for 15 weeks. The curriculum has seven parts: learning school rules; how to be successful in school; emotional literacy, empathy, and perspective-taking; inter- personal problem solving; anger management; social skills; and communication skills. Lessons include recognizing and understanding feelings, getting along with friends, regu- lating emotions, solving problems, and behaving at school. The lessons are reinforced by practicing the skills in 20-minute sessions after completion of the structured curriculum and continuing to practice the skills throughout the school day and at home. The program is supplemented by parent training that focuses on positive discipline and engaging in their children’s school lives through family homework. Incredible Years has been implemented in the United States with culturally diverse groups, including Caucasian, Hispanic, Asian-American, African-American, and new immi- grant families. The total cost per child ranges from $1,200 to $3,000, depending on project components. A randomized trial of the program was conducted to assess the extent to which the pro- gram affected behaviors and social outcomes. Random assignment of matched preschools and elementary schools resulted in a sample of 153 teachers and 1,768 students. The treatment group received the full program; the control group received no additional programming. The study reported outcomes after six months of program implementation. It found that program participants were significantly more likely than controls to display emotional self-regulation and social competence; they also had fewer conduct problems and engaged in less off-task behavior. The effect was particularly strong among students in classrooms with the lowest ini- tial scores on these skills. Source: Webster-Stratton, Reid, and Stoolmiller 2008. interventions for all adolescents, not only those considered at-risk (Guerra, Modecki, and Cunningham 2014). Although adolescents benefit from school-based programs (CASEL 2015), school is a less optimal intervention venue for adolescents than for younger chil- dren, for several reasons. Many youth are not in school, and secondary school is structured in a manner that is not conducive to socioemotional skills develop- ment (schedules are varied, students change classrooms and teachers regularly). Moreover, adolescents are starting to shift their context away from parents and schools toward peers and the working world. Three contexts can provide socioemotional skill development for this age group: secondary schools, out-of-school programs, and work-related pro- grams. Interventions for school-based programs for adolescents take the same Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 54 Policy and Programming for Socioemotional Skill Development form as those for middle childhood, although the skills modeled by teachers and developed via pedagogical methods are more advanced than those taught to younger children. Although most programming and evaluations focus on at-risk adolescents, several program experiences with a broader range of adolescents have shown success (Guerra, Modecki, and Cunningham 2014). Out-of-school programs Out-of-school programs can reach adolescents who are still enrolled in school and adolescents who have left school. These programs take two forms. The first is the theme-based club, which is ideally structured as a fun activity (sports, theater, art, music) and includes a staff member who is a child psychologist, who helps the club leader design the activity so that it teaches and rein- forces socioemotional skills. Clubs like these have been used across the world. Although there are few evaluations of the impact of these types of programs in developing the PRACTICE skills, some models show promise (Cunningham and others 2008; Guerra, Modecki, and Cunningham 2014; box 4.6). The second type of program is mentoring programs that provide a strong adult presence to model behaviors and support positive socioemotional development in youth. Mentoring programs can take various forms, such as after-school clubs or programs that pair model adults with children. Both models have been shown to increase the cognitive and socioemotional skills of participants relative to Box 4.6 Reaching At-Risk Adolescents through Colombia’s Fútbol con Corazón Out-of-School Program Fútbol con Corazón (Soccer with Heart) uses soccer to teach values and cognitive, emotional, and social skills to children and youth living in poor neighborhoods in Colombia. It is based on the Fútbol por la Paz (Football for Peace) methodology, which uses soccer as the hook to attract participants. Children age 5–17 play in gender-mixed teams supported by mentors who are trained to  reinforce the socioemotional objectives of the program. The rules of the game are changed in order to teach a range of socioemotional skills, including social problem-solving, resilience, self-control, teamwork, initiative, confidence, and ethics. Alternative rules include the following: • Before play, the teams agree on the rules of play. • There are no referees. Players self-judge and agree on errors, penalties, and other aspects of fair play. • The first goal must be made by a girl; if it is not, it does not count. • The winner is not necessarily the team with the most points. The final score is based on a combination of the number of goals and the extent to which each team complied with the teamwork norms set down at the beginning of play. The program has not been evaluated for impact. Source: Deporte y Desarollo 2009. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 55 control groups (Tierney and Baldwin Grossman 2000; Boys & Girls Clubs of America 2004). Labor market–related programs As the school-to-work transition process begins, job training and apprenticeship programs should be augmented by interventions that develop socioemotional and advanced cognitive skills. A job training program for low-income youth in the Dominican Republic combines 150 hours of technical training with short internships and 75 hours of socioemotional skills training and practice. Impact evaluation results show that participants performed better than controls on mea- sures of many PRACTICE skills: leadership, conflict management, self-esteem, interpersonal skills (for some groups), organization skills, empathy, and hard work (Ibarrarán and others 2014). The technical training was deemed to have no valued added, suggesting that the socioemotional training and internships were responsible for these, as well as some minor labor market, outcomes (Fazio 2011; Martinez 2011; and Vezza and others 2014). Not all labor markets are able to absorb youth, and not all youth are ready for employment, pointing to the value of activities that approximate the work envi- ronment and teach the PRACTICE skills in the process. YouthBuild is an interna- tional program that “employs” youth in community improvement projects while explicitly emphasizing improvements in PRACTICE skills such as teamwork, initiative, and confidence. Evaluations from quasi-experimental studies show improved life outcomes (educational progress, having a job) for participating youth; acquisition of specific skills has not been assessed (International Youth Foundation, YouthBuild International, and Catholic Relief Services 2010). Emerging Adulthood (19–21): Practicing PRACTICE Development of socioemotional skills can, and should, continue into early adult- hood. Although the neurobiological and psychosocial bases are well set by early adulthood, new experiences—higher education, jobs, marriage—provide oppor- tunities for practicing and developing new skills that, according to empirical studies, modify socioemotional skills (Roberts 1997; Roberts, Caspi, and Moffitt 2003). This is particularly important for the PRACTICE skills, because young adults’ primary context is the labor market, which provides them with daily opportunities to refine and consolidate these skills. Research on programming for young adults is sparse. Cognitive and Socioemotional Learning Processes Like socioemotional skills, cognition is a function of neurobiological develop- ment, situational stimuli, and various actors. The basis for cognition begins soon after birth, with physical changes in the brain and body and interaction with the environment (via smell, vision, and other senses). Building cognitive skills is a continuous, progressive process of acquiring increasingly higher-level skills, a process that becomes more and more complex. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 56 Policy and Programming for Socioemotional Skill Development Consider, for example, the development of the ability to compare objects. Very young children compare different shapes or colors. In adolescence ­ comparison might include comparing the digestive and respiratory systems, a complex comparison that involves more elements, such as functionality and shape. Complexity deepens as cognitive development intertwines with socio- emotional development. Social interactions, first with parents and later with teachers and peers, foster socioemotional skills that are crucial to develop the capacity to explore and learn from experience, a key prerequisite for acquiring cognitive skills. Skills development strategies should aim to develop a range of cognitive skills. The analytical work in this book shows that basic skills (including reading profi- ciency, verbal fluency, numeracy, and memory) are correlated with labor market outcomes. Employers also value advanced cognitive abilities, such as communica- tions, problem-solving, and critical thinking skills (Cunningham and Villaseñor 2016). Skills development strategies must therefore not only teach basic skills but also help people master the following advanced cognitive skills: • Remembering (the process in which information is recalled from long-term memory) • Understanding (constructing meaning based on previous knowledge) • Applying information and previous experiences to solve problems • Analyzing (the process of deconstructing material into its parts and determin- ing the relationship among them) • Evaluating (the capacity to form a judgment based on criteria and standards) • Creating or combining elements to form a new coherent concept Development of these skills begins in early childhood and continues through adulthood. The most basic form of these skills can be fostered at a very young age. The learning process and the nature of the skills become more complex as children grow and interact with different people and environments, gain new experience, and accumulate knowledge. As with the development of socioemo- tional skills, the lifelong work of acquiring these skills takes different forms. The pedagogical literature on teaching basic cognitive skills is so vast that we will not attempt to summarize it here. The methods for teaching advanced cognitive skills is less developed and mirrors methods for socio- emotional skill instruction. We can apply the concepts of neurobiological, psychosocial, and situational readiness of the learner to direct the reader toward learning processes that are best suited to age-readiness. And we pro- vide insights into the practice of socioemotional skills to enhance advanced cognitive skills development. Early Years (0–5): Play Play is the work of children. Through experimentation and creativity, children begin to build basic and advanced cognitive (and socioemotional) skills. Play is particularly effective for young children, among whom a range of play activities Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 57 develops cognition. In the first years of life, parents and caregivers stimulate children mentally through songs, colors, and playful interactions (such as peeka- boo) to foster cognitive development. Later, they use storytelling, the showing and reading of books, make-believe games, and animation to attract children’s interest and maintain their focus. These techniques can be used in both homes and institutional settings, as demonstrated by many well-evaluated programs (box 4.7). Tools of the Mind (box 4.1) is a curriculum for preschoolers that approaches learning as socially mediated by peers and focus on play. It has shown substan- tial short-term effects on cognitive control among vulnerable children in the United States (Diamond and others 2007). Chile is experimenting with Tools of the Mind in some preschools in Santiago. Many countries in Latin America have started designing and implementing similar programs. Examples include Chile Crece Contigo (Chile Grows with You), Uruguay Crece Contigo (Uruguay Grows with You), Cuna Más (Cradle More; Peru), and Quizqueya Empieza Contigo (Quizqueya Begins with You; the Dominican Republic). These programs contribute to children’s school readi- ness by ensuring that cognitive, socioemotional, and motor skills milestones are achieved. Box 4.7  Fostering Active Learning in Preschool Educators can foster active learning in preschool in a variety of ways: • Use objects as toys. Children should be given plenty of opportunities to make, explore, and manipulate materials and interact with the environment. Materials include everyday objects like shells and cardboard boxes; substances with different types of surfaces, like sand, water, and paste; large, heavy materials, such as wooden blocks; and easy-to-handle materials, such as toy figures and Lego® blocks. • Offer space for play. Children need an organized environment in which to play. Teachers and caregivers should establish a daily routine that includes time for children to play with objects they choose, small-group playtime, large-group activity time for songs or movement, and outdoor time to experience and explore materials from nature. • Seek to understand intentions. Trying to discern children’s intentions reinforces their sense of initiative and control. Having children reflect on their thoughts or feelings and put them into words improves their verbal skills. • Listen and encourage thinking. Listening to and encouraging each child’s way of thinking strengthens their ability to reason. Talking with children about what they are doing and thinking fosters self-esteem. • Encourage doing for themselves. Encouraging children to solve the problems they face offers them learning opportunities. Educators stand patiently and wait while children sort out issues by themselves, providing feedback and support only when necessary. Source: Hohmann and Weikart 1995. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 58 Policy and Programming for Socioemotional Skill Development Middle Childhood (6–11): Guided Discovery and Experimentation Active learning is more effective than passive learning. Pedagogical practice in Latin America generally focuses on passive learning based on repetition, low engagement, and copying from blackboards. Not only do students not absorb the information, they also do not practice the cognitive skills acquired earlier. The role of the teachers in the classroom needs to change. Teachers need to shift from being transmitters of information to being facilitators, to guid- ing children of all ages in acquiring knowledge (Bonwell and Eison 1991). Instruction should foster active learning; students should be engaged in activities such as reading, discussing, writing, and analyzing, guided by the teacher (box 4.8). Pedagogical methods should put more emphasis on students’ exploration and attitudes. Classroom instruction should provide students with learning and study techniques that boost memory; help stu- dents transform information into knowledge by experience (learning by doing); capture students’ attention; frame information in meaningful ways (so that it can be related to previously acquired knowledge); and use a range of formats (reading, writing, visual, sound) to provide learning opportunities to students with different learning styles. It should engage students in the metacognitive process of evaluating their own work and incentivize them to think about their thinking and the way they learn; provide feedback; continuously assess learning (to foster memory); help them organize and categorize verbal information so that they can recall it more efficiently; and provide extensive practice with procedural knowledge and problem- solving (that is, how to do something), so that students are able to decode text, write, carry out simple math operations efficiently, and teach others cognitive strategies (methods of thinking that improve learning). For all these methods, the teacher is the facilitator, using her own socioemotional Box 4.8  Active Learning through Colombia’s Escuela Nueva Approach Escuela Nueva (New School) is a pedagogical method based on active learning that started in  rural communities in the 1970s in Colombia. Based on active, child-centered learning, it encourages students to work in groups. Rather than focusing exclusively on transmitting information, teachers facilitate and guide students. Escuela Nueva has increased student retention and achievement, reduced grade repeti- tion, and improved behaviors. As documented by the Colombian government, the approach results in better outcomes than conventional schooling for the most disadvantaged children. The approach has been implemented under different names in nearly 20 countries, including Mexico, India, and Vietnam. Source: Schiefelbein 1993, Colbert 2009. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 59 and advanced cognitive skills to guide discussion toward the learning objec- tives through continued questioning. Several strategies can help the region’s teachers transform themselves into learning facilitators. First, teachers, parents, policy makers, and students need to shift their understanding of the role of the teacher and of what constitutes appro- priate pedagogical practice. Second, teacher training needs to be redesigned to provide teachers with both the content and modern pedagogical strategies to dramatically change children’s experience in the classroom. Teachers need to be trained in the importance of active learning for skills development and experi- ence these concepts through real classroom practice from the moment they begin training. Third, continuous teacher career development and support are needed to incorporate new pedagogical methods as they emerge (Bruns and Luque 2014). Adolescence (12–18): Development of Advanced Cognitive Skills through Independent Discovery Advanced cognitive skills that are best taught in secondary school—such as metacognition (thinking about thinking)—may require different teaching strate- gies than were employed for basic cognitive skills instruction. During adoles- cence, the brain is neurobiologically and psychosocially ready to self-guide such that teacher facilitation plays a lesser role. The adolescent has the cognitive bases and the socioemotional skills (perseverance, independent work, problem solving) to utilize external resources for learning. The teacher becomes a partner in dis- covery, nudging the students toward higher levels of cognition and socioemo- tional mastery. Emerging Adulthood (19–21): Learning by Doing ­ ractice-oriented Most learning by adults occurs through on-the-job experience and p training institutes (OECD 2014). Skills development strategies usually stop at the factory door using the assumption is that skills development ­ happens only in a classroom. Given that people in Latin America spend an average of 10 years in school and 50 years in the labor market and that cognitive (and socioemotional) skills are malleable over a lifetime, modalities for acquiring new skills in the workplace are a new area for research. Technical training models need enhanced pedagogical practices to teach a range of advanced cognitive skills that employers demand (OECD 2014, Fawcett, El Sawi, and Allison 2014). Training should be delivered through structured experience-based (rather than classroom) learning that uses real-world practice to reinforce the cognitive skills learned in school. Although experiential learning is common practice in European technical train- ing systems (Fawcett, El Sawi, and Allison 2014), employers in other countries can find room for improvement. Latin American executives identify work study programs that bring students into the workplace as the primary way that firms can contribute to skills development processes (Ogier 2009). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 60 Policy and Programming for Socioemotional Skill Development Notes 1. The workplace may provide explicit socioemotional training, although such training is rare. Most firms believe that such training is the role of the individual or the public sector (Martin and others 2008). 2. “Foundational” indicates that skills developed in this period form the basis for core skill-building later on. “Optimal” refers to periods of maximum sensitivity when it is easiest for individuals to acquire specific skills. “Reinforcement” indicates that a skill acquired during the optimal period needs to be practiced intensively to be mastered. References Angrist, J., S. Dynarski, T. Kane, P. Pathak, and C. Walters. 2010. “Who Benefits from KIPP?” NBER Working Paper 15740, NBER, Cambridge, MA. Barr, D. J. 2010. Continuing a Tradition of Research on the Foundations of Democratic Education: The National Professional Development and Evaluation Project. Facing History and Ourselves National Foundation, Inc. http://www.facinghistory.org/sites​ /­default/files/Continuing_a_Tradition_v93010_0.pdf. Blair, C., and C. C. Raver. 2014. Closing the Achievement Gap through Modification of Neurocognitive and Neuroendocrine Function: Results from a Cluster Randomized Controlled Trial of an Innovative Approach to the Education of Children in Kindergarten. PLoS ONE 9 (11): e112393. Bodrova, E., and D. Leong. 2015. Tools of the Mind: The Vygotskian Approach to Early Childhood Education. Geneva: Pearson. Bonwell, C. C., and J. A. Eison. 1991. Active Learning: Creating Excitement in the Classroom. ASHE-ERIC Higher Education Report 1, George Washington University, Washington, DC. Boys & Girls Clubs of America. 2004. Proven Results: A Compendium of Program Evaluations from Boys & Girls Clubs of America 1985–Present. Atlanta: Boys & Girls Clubs of America. Bradshaw, C. P., M. M. Mitchell, and P. J. Leaf. 2010. “Examining the Effects of Schoolwide Positive Behavioral Interventions and Supports on Student Outcomes: Results from a Randomized Controlled Effectiveness Trial in Elementary Schools.” Journal of Positive Behavior Interventions 12 (3): 133–48. Bruns, B., and J. Luque. 2014. Great Teachers: How to Raise Student Learning in Latin America and the Caribbean. World Bank, Washington, DC. CASEL (Collaborative for Academic, Social, and Emotional Learning). 2013. Effective Social and Emotional Learning Programs: Preschool and Elementary School Edition. Chicago: CASEL. ______. 2015. Effective Social and Emotional Learning Programs: Middle and High School Edition. Chicago: CASEL. Colbert, V. 2009. “Improving Education Quality and Access in Colombia through Innovation and Participation: The Escuela Nueva Model.” Journal of Education for International Development 3 (3): 1–8. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 61 Cunha, F., and J. Heckman. 2008. “Formulating, Identifying and Estimating the Technology of Cognitive and Noncognitive Skill Formation.” Journal of Human Resources 43 (4): 738–82. Cunningham, W., and P. Villaseñor. 2016. “Employer Voices, Employer Demands, and Implications for Public Skills Development Policy Connecting the Labor and Education Sectors.” World Bank Research Observer 31 (1): 102–34. Cunningham, W., L. McGinnis, R. García Verdu, C. Tesliuc, and D. Verner. 2008. Youth at Risk in Latin America and the Caribbean: Understanding the Causes, Realizing the Potential. Directions in Development. Washington, DC: World Bank. Deporte y Desarollo. 2009. Fútbol con Corazón: Goles de paz y conviviencia. (Football with Heart: Goals of Peace and Coexistence.) http://www.deporteydesarrollo.org​ /­proyectos_AnotandoGolesdePazyConvivencia_Colombia.pdf. Diamond, A., W. S. Barnett, J. Thomas, and S. Munro. 2007. “Preschool Program Improves Cognitive Control.” Science 318: 1387–88. Durlak, J. A., R. Weissberg, A. Dymnicki, R. Taylor, and K. Schellinger. 2011. “The Impact of Enhancing Students’ Social and Emotional Learning: A Meta-Analysis of School- Based Universal Interventions.” Child Development 82 (1): 405–32. Fawcett, Caroline, Gwen El Sawi and Christine Allison. 2014. TVET Models, Structures, and Policy Reform: Evidence from the Europe & Eurasia Region. USAID, Washington DC. Fazio, M. V. 2011. “Análisis de la percepción de los empleadores acerca de las pas- antías y las perspectivas de inserción laboral en esas empresas (Analysis of the Perception of Employers about Internships and Employment Prospects in These Companies).” Nota Técnica IDB-TN-240, Inter-American Development Bank, Washington, DC. Gertler, P., J. J. Heckman, R. Pinto, A. Zanolini, S. Walker, and S. Grantham-McGregor. 2014. “Labor Market Returns to an Early Childhood Stimulation Intervention in Jamaica.” Science 344 (6187): 998–1001. Guerra, N., K. Modecki, and W. Cunningham. 2014. “Social-Emotional Skills Development across the Life Span: PRACTICE.” Policy Research Working Paper 7123, World Bank, Washington, DC. Hohmann, M., and D. P. Weikart. 1995. “Educating Young Children: Active Learning Practices for Preschool and Child Care Programs.” In Educating Young Children, edited by M. Hohmann and D. P. Weikart. Ypsilanti, MI: High/Scope Educational Research Foundation. Horner, R., G. Sugai, K. Smolkowski, L. Eber, J. Nakasato, A. W. Todd, and J. Esperanza. 2009. “A Randomized, Wait-List Controlled Effectiveness Trial Assessing School-Wide Positive Behavior Support in Elementary Schools.” Journal of Positive Behavior Interventions 11 (3): 133–44. Ibarrarán, P., L. Ripani, B. Taboada, J. M. Villa, and B. Garcia. 2014. “Life Skills, Employability and Training for Disadvantaged Youth: Evidence from a Randomized Evaluation Design.” IZA Journal of Labor and Development 3: 1–24. International Youth Foundation, Youthbuild International, and Catholic Relief Services. 2010. Executive Summary: International Youth Foundation Follow-Up Study “Central American Youth Builders Program. Baltimore, MD: International Youth Foundation, Youthbuild International & Catholic Relief Services. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 62 Policy and Programming for Socioemotional Skill Development Kagitcibasi, C., D. Sunar, and S. Berkman. 1988. Comprehensive Preschool Education Project. International Development Research Centre, Ottawa. Kautz, T., J. J. Heckman, R. Diris, B. T. Weel, and L. Borghans. 2014. “Fostering and Measuring Skills: Improving Cognitive and Non-Cognitive Skills to Promote Lifetime Success.” OECD Education Working Paper 110, OECD Publishing, Paris. Kuhn, D. 2006. “Do Cognitive Changes Accompany Developments in the Adolescent Brain?” Perspectives on Psychological Science 1: 59–67. Martin, R., F. Villeneuve-Smith, L. Marshall, and E. McKenzie. 2008. Employability Skills Examined: Shedding Light on the Views of Stakeholders and Employers. Learning Skills Network, London. Martinez, S. 2011. “Impacts of the Dominican Republic Youth Employment Program: Hard Skills or Soft Skills.” World Bank, Washington, DC. http://siteresources.­ w o r l d b a n k . o r g / I N T L M / R e s o u r c e s / 3 9 0 0 4 1 - 1 1 4 1 1 4 1 8 0 1 8 6 7 / 2 2 7 5 3 6 4​ -1313438221557​/­PJE_DR_PPT.pdf. OECD (Organisation for Economic Co-operation and Development). 2014. Skills Beyond School: Synthesis Report. OECD Reviews of Vocational Education and Training. Paris: OECD Publishing. Ogier, Thierry. 2009. Skills to Compete: Post Secondary Education and Business Sustainability in Latin America. Economist Intelligence Unit, London. Paredes, G. 2014. “Diplomado en educación socioemocional para la convivencia escolar: Propuesta y resultados.” (Graduate in Socioemotional Education for School Conviviality: Proposal and Results). Dirección Académica de Responsabilidad Social (Academic Board of Social Responsibility) of the Pontificia Universidad Católica (Pontifical Catholic University of Peru) of Peru. Lima. Peacock, S., S. Conrad, E. Watson, D. Nickel, and N. Nuhajarine. 2013. “Effectiveness of Home Visiting Programs on Child Outcomes: A Systematic Review.” BMC Public Health 13: 17. DOI:10.1186/1471-2458-13-17. Roberts, B. W. 1997. “Plaster or Plasticity: Are Work Experiences Associated with Personality Change in Women?” Journal of Personality 65: 205–32. Roberts, B. W., A. Caspi, and T. Moffitt. 2003. “Work Experiences and Personality Development in Young Adulthood.” Journal of Personality and Social Psychology 84 (3): 582–93. Schiefelbein, E. 1993. “In Search of the School of the XXI Century.” UNICEF-UNESCO. Santiago, Chile. Schweinhart, L. J., J. J. Heckman, L. Malofeeva, R. Pinto, S. Moon, and A. Yavitz. 2010. The Cost-Benefit Analysis of the Preschool Curriculum Comparison Study. Final Report to the John D. and Catherine T. MacArthur Foundation. Ypsilanti, MI: HighScope. Schweinhart, L., J. Montie, Z. Xiang, W. Barnett, C. Belfield, and M. Nores. 2005. Lifetime Effects: The High/Scope Perry Preschool Study Through Age 40. Ypsilanti, MI: High Scopes Press. Shiner, R. L. 2000. “Linking Childhood Personality with Adaptation: Evidence for Continuity and Change across Time into Late Adolescence.” Journal of Personality and Social Psychology 78: 310–21. Tierney, J., and J. Baldwin Grossman. 2000. Making a Difference: An Impact Study of Big Brothers Big Sisters. Public/Private Ventures, Philadelphia. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Policy and Programming for Socioemotional Skill Development 63 Vezza, E., B. García, G. Cruces, and J. Amendolaggine. 2014. Programa Juventud y Empleo: Informe de evaluación de impacto cohortes 2008–2009 [Youth and Employment Program: Report of the Impact Evaluation on 2008–09 Cohorts]. World Bank and the Ministry of Labor of the Dominican Republic, Washington, DC. Webster-Stratton, C., M. J. Reid, and M. Stoolmiller. 2008. “Preventing Conduct Problems and Improving School Readiness: Evaluation of the Incredible Years Teacher and Child Training Programs in High-Risk Schools.” Journal of Child Psychology and Psychiatry 49 (5): 471–88. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Chapter 5 Conclusions This study sets out to depict how different types of skills affect labor market outcomes in Latin America. The results are based on findings from new data on Bolivia, Colombia, El Salvador, and Peru, combined with the scarce literature available from two other Latin American countries (Argentina and Chile). It gives particular emphasis to the study of socioemotional skills and the role that they play in labor market outcomes. Based on the analysis, the study presents a life-cycle framework for when and how to acquire the skills valued by Latin America’s labor market. It also points toward pedagogical methods that simulta- neously enhance both socioemotional and cognitive skills. Both Cognitive and Socioemotional Skills Are Associated with Labor Market and Tertiary Education Outcomes Cognitive skills are positively associated with better jobs. Reading proficiency (in Bolivia and Colombia), math skills (in El Salvador and Peru) and verbal fluency (in Peru) are positively correlated with higher wages, formal employment, or tertiary education. Of the four countries studied, cognitive skills are associated with employment only in El Salvador. Socioemotional skills are associated with employment patterns and a range of labor market outcomes. Skills associated with achieving goals (conscientiousness, grit) are correlated with employment in Latin America. This finding emerged in all four countries in the sample; it is also evident in high-income countries. Skills associated with all three of the socioemotional skill sets explored in this study (achieving goals, managing emotions, and working with others) are correlated with higher labor earnings and, to a lesser degree, formal employment and wage employment. Only one skill (in the working with others category—extroversion) was uncorrelated with any labor market outcome. As in higher-income countries, socioemotional skills are correlated with tertiary school attendance in the four countries studied. Skills associated with achieving goals, working with others, and managing emotions predict whether students pursue tertiary education, with perseverance playing a particularly important role. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   65   66 Conclusions Instruction in Both Cognitive and Socioemotional Skills Is Both Possible and Necessary to Better Prepare Latin American Workers for the Labor Market Socioemotional skills must be taught from early childhood through early adult- hood and beyond. Important foundational socioemotional skills are developed in the early years. They include social problem solving when playing with peers; impulse control; and the development of confidence, a set of ethics, and the abil- ity to work with others. The young brain is highly malleable and is being actively wired in this stage, creating a foundation for later skill development. But some skills cannot be taught in early childhood, because the brain is not yet con- structed to learn certain skills, the child is not psychologically developed to understand certain concepts, and the social environment does not allow the child to practice these skills. Middle childhood is an optimal period for developing the socioemotional skills most valued by the labor market, because children this age (6–11) have the neurobiological wiring and psychosocial maturity needed and are in an environment that permits practice and learning. Adolescence is also a period of optimal skill development, allowing for more complex development of skills initiated in middle-childhood. These skills are practiced, refined, and rein- forced during adulthood. Programs targeted to different age levels, engaging appropriate actors, and utilizing age-appropriate concepts and methodologies are showing that the development of socioemotional skills can occur in structured environments outside the home, including in programs run by the public sector. Public policy has a role in defining, monitoring, regulating, and even providing socio- emotional skills development, as it traditionally has in the case of cognitive skills development. Schools are the ideal place to teach socioemotional skills. They are the public institution that has greatest influence over children and youth in this period. Socioemotional development should continue through the high school years and not be limited to at-risk youth, the current practice. These findings highlight the importance of keeping children in school through adolescence. Methodologies for teaching socioemotional skills to children and adolescents can be grouped into five modalities: • Training teachers in socioemotional skills. Training teachers can equip them with skills they can model in their classrooms. The teachers themselves learn to develop their own skills and to use these skills in their professional and per- sonal lives. • Improving the school climate. Schools can create a climate that is conducive to positive socioemotional skills by defining and rewarding positive behaviors. The school environment is a space to practice positive behaviors. • Weaving instruction into regular classes. Socioemotional instruction can be woven into pedagogical methods including through interactive instruction, group learning, and problem solving. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Conclusions 67 • Using a special curriculum. Schools can allocate class time to a socioemotional school curriculum. • Teaching socioemotional skills after school. Extracurricular activities can be designed not only to be fun but also to teach and allow students to practice a range of socioemotional skills. This type of intervention also reaches out-of- school youth. All of these methodologies have been implemented and evaluated across the world (appendix F). The pedagogical methods and actors used in socioemotional skill development can also address lagging cognitive skills. Cognitive skills need to be understood more broadly as both knowledge and a range of thinking and analytical skills. Research has shown that active learning is more effective than passive learning for all age groups (Hohmann and Weikart 1995, Schiefelbein 1993, Colbert 2009). A shift in pedagogical method is therefore needed, toward teachers as facilitators in a knowledge-discovery process that the learner leads. Methods for developing cognitive skills should be adapted based on the neurobiological, social, and situational contexts of the learner. Young children learn best through play, older children and adolescents through guided discovery and experimenta- tion, and adults by doing. This is not to say that all skills must be taught in an identical manner, but that the teaching of cognitive skills should take the context of the learner into consideration. Research Is Needed to Guide Policy Design There is growing consensus that both cognitive and socioemotional skills are critical inputs to labor market success. But the evidence to guide policy design is still limited. Some specific areas for research include the following: • Longitudinal studies in Latin America, such as the recent project of Longitudinal Study of Children’s Social and Emotional Skills in Cities, launched by the Organisation for Economic Co-operation and Development (OECD 2015), identify the skills that are most important for labor force outcomes. However, most studies, including this book, use single cross-section data which cannot establish the direction of causality. It is not clear if skills lead to the observed labor force outcomes or if causality may actually go in the other direction. Panel data that measure skills, education, and labor force patterns across time may sort out this possible endogeneity and yield cleaner results. • Long-term studies to determine which interventions generate permanent change. Most evaluations consider short-term program impacts (Kautz and others 2014). The few that do allow measurement of long-term impacts find mixed results; several celebrated programs with short-term impacts have been found to have no effect after a few years. A better understanding of the inter- ventions that permanently affect cognitive and socioemotional skills will allow for more efficient policy design. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 68 Conclusions • Evidence on interventions to generate socioemotional skills via family (for early childhood), schools (for adolescents who are not at risk), community (for adoles- cents and young adults) and, especially, workplaces (for young adults). Thousands of socioemotional interventions have been implemented, but very few studies empirically establish the impact of these interventions. Although there is some evidence that center-based early child development programs, primary school interventions, and programs for at-risk youth produce results, virtually no evi- dence supports many other potentially crucial types of programs. • Scientific evidence on biological and psychosocial readiness for socioemotional skill development. This study reports findings from the neurobiological and psycho- logical research that points to the importance of readiness. The tools for imag- ing the brain and translating the findings to practical learning processes are in their infancy. Much more work needs to be done to understand readiness concepts. References Colbert, V. 2009. “Improving Education Quality and Access in Colombia through Innovation and Participation: The Escuela Nueva Model.” Journal of Education for International Development 3 (3): 1–8. Hohmann, M., and D. P. Weikart. 1995. “Educating Young Children: Active Learning Practices for Preschool and Child Care Programs.” In Educating Young Children, edited by M. Hohmann and D. P. Weikart. Ypsilanti, MI: High/Scope Educational Research Foundation. Kautz, T., J. J. Heckman, R. Diris, B. T. Weel, and L. Borghans. 2014. “Fostering and Measuring Skills: Improving Cognitive and Non-Cognitive Skills to Promote Lifetime Success.” OECD Education Working Paper 110, Organisation for Economic Co-operation and Development, Paris. OECD (Organisation for Economic Co-operation and Development). 2015. Longitudinal Study of Children’s Social and Emotional Skills in Cities (LSEC). Paris. www.oecd​ .org/edu/ceri/lsec.htm. Schiefelbein, E. 1993. “In Search of the School of the XXI Century.” UNICEF-UNESCO. Santiago, Chile. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Appendix A Abstracts of Background Papers Developing Social-Emotional Skills for the Labor Market: The PRACTICE Model. 2014. Nancy Guerra, Kathryn Modecki, and Wendy Cunningham. Policy Research Working Paper 7123, World Bank, Washington, DC. Although there is a general agreement in the literature on the importance of socialemotional skills for labor market success, there is little consensus on the specific skills that should be acquired or how and when to teach them. Psychology, economics, policy research, and program implementation touch on these issues, but this work is not sufficiently integrated to provide policy direc- tion. This paper provides a coherent framework and related policies and pro- grams that bridge the psychology, economics, and education literature related to skills employers value, noncognitive skills that predict positive labor market outcomes, and skills targeted by psychoeducational prevention and intervention programs. The paper classifies social emotional skills employers value into eight subgroups (summarized by the acronym PRACTICE). It then uses the psychol- ogy literature—drawing from the concepts of psychosocial and neurobiological readiness and age-appropriate contexts—to map the age at and context in which each skill subset is developed. The paper uses examples of successful interven- tions to illustrate the pedagogical process. It concludes that the social emotional skills employers value can be effectively taught when aligned with the optimal stage for each skill development, that middle childhood is the optimal stage for development of PRACTICE skills, and that a broad international evidence base on effective program interventions at the right stage can guide policy makers in incorporating social emotional learning into their school curriculum. Employer Voices, Employer Demands, and Implications for Public Skills Development Policy Connecting the Labor and Education Sectors. 2016. Wendy Cunningham and Paula Villaseñor. World Bank Research Observer 31(1): 102–34. (An earlier version of this paper was published in 2014 as World Bank Policy Research Working Paper 6853.) Educators believe that they are adequately preparing youth for the labor mar- ket while at the same time employers lament the students’ lack of skills. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   69   70 Abstracts of Background Papers A ­possible source of the mismatch in perceptions is that employers and educa- tors have different understandings of the types of skills valued in the labor mar- ket. Drawing on the economics and psychology literature to define four skills sets—socioemotional, higher-order cognitive, basic cognitive, and technical—this paper reviews the literature that quantitatively measures employers’ demand for skills, as reported in a preference survey. A sample of 27 studies reveals remark- able consistency across the world in the skills demanded by employers. Although employers value all skill sets, their demand for socioemotional skills and higher- order cognitive skills is greater than their demand for basic cognitive or technical skills. These results are robust across regions, industries, occupations, and educa- tion levels. Employers perceive that the greatest gaps are in socioemotional and higher-order cognitive skills. These findings suggest the need to reconceptualize the public sector’s role in preparing children for a future labor market. Namely, technical training is not equivalent to job training; instead, a broad range of skills, many of which are best taught long before labor market entry, should be included in school curricula from the earliest ages. The fact that the skills most demanded by employers—higher-order cognitive skills and socioemotional skills—are largely learned or refined in adolescence argues for general education well into secondary school until these skills are formed. The public sector can provide programming and incentives to nonschool actors, namely parents and employers, to encourage them to invest in the skills development process. Cognitive and Non-Cognitive Skills for the Peruvian Labor Market: Addressing Measurement Error through Latent Skills Estimations. 2016. Wendy Cunningham, Mónica Parra Torrado, and Miguel Sarzosa. Policy Research Working Paper 7550, World Bank, Washington, DC. Evidence from developed-country data suggests that cognitive and noncogni- tive skills contribute to improved labor market outcomes. This paper tests this hypothesis in a developing country by using an individual-level data set from Peru that incorporates modules to measure cognitive and noncognitive skills. It estimates a structural latent model with unobserved heterogeneity to capture full ability rather than just measured skill. It also applies standard ordinary least squares techniques for comparison. The analysis confirms that cognitive and noncognitive skills are positively correlated with a range of labor market out- comes in Peru. In particular, cognitive skills correlate positively with wages and the probability of being a wage worker, white-collar, or formal worker, with ver- bal fluency and math ability playing particularly strong roles. The results are robust to methodology. The patterns are less uniform for noncognitive skills. For instance, perseverance of effort (grit) emerges strongly for most outcomes regardless of methodology, but plasticity—an aggregation of openness to experi- ence and resilience—is correlated only with employment and only when using the structural latent model. The ordinary least squares method also finds that the disaggregated noncognitive skills of kindness, cooperation, and openness to expe- rience are significant, mostly for the wage estimates. The different results derived from the ordinary least squares and the structural model with latent skills suggest Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Abstracts of Background Papers 71 strong measurement bias in most noncognitive skills measurement. These find- ings, although only correlational because of the use of a single cross-section, sug- gest that recent efforts by the Peruvian government to incorporate noncognitive skill development into the school curriculum are justified. Beyond Qualifications: Returns to Cognitive and Socio-Emotional Skills in Colombia. 2015. Pablo A. Acosta, Noël Muller, and Miguel Sarzosa. Policy Research Working Paper 7430, World Bank, Washington, DC. This paper examines the relationship between individuals’ skills and labor market outcomes for the working-age population in urban areas of Colombia. Using a 2012 unique household survey, it finds that cognitive skills (aptitudes to perform mental tasks such as comprehension or reasoning) and socioemotional skills (personality traits and behaviors) are associated with favorable labor market outcomes in the Colombian context, although they have distinct roles. Cognitive skills are strongly correlated with higher earnings, holding a formal job, and work- ing in a high-qualified occupation. By contrast, socioemotional skills appear to have little direct influence on these outcomes but play a strong role in labor market participation. Both types of skills, especially cognitive skills, are strongly associated with tertiary education. The analysis applies standard econometric techniques as a benchmark and structural estimations to correct for the measure- ment error of skill constructs. Ethnicity and Labor Market Returns to Cognitive and Socio-Emotional Skills in Urban Bolivia. 2015. Juan D. Barón, Miguel Sarzosa, and José Mola. Background study. Using a unique survey collected in 2012, this paper estimates the returns to cognitive and socioemotional skills on labor and educational outcomes by ethnic- ity in urban Bolivia. Results from structural maximum likelihood estimation, which takes into account the latent nature of skills, indicate that (a) irrespective of ethnicity group, cognitive and socioemotional skills matter for labor and edu- cational outcomes in urban Bolivia, with cognitive skills appearing to be more important for labor market outcomes than socioemotional skills; (b) the skills profiles of indigenous and nonindigenous groups seem to differ for both types of skills; and (c) there is substantial heterogeneity in the relationship between labor market outcomes and skills by ethnicity. Skills and Labor Market Outcomes in El Salvador. 2015. Ana María Oviedo and Noël Muller. Background study. Do adults with higher levels of cognitive and socioemotional skills achieve better labor market and tertiary education outcomes in El Salvador? Using data from the 2013 El Salvador Household Skill Survey, this background study finds that both types of skills are indeed associated with labor market outcomes in this context, especially socioemotional skills. Socioemotional skills linked to achieving goals (conscientiousness, openness to experience) and managing emo- tions (resilience, decision-making style) are largely and positively correlated with Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 72 Abstracts of Background Papers indicators of good jobs (higher labor earnings, formal job, and high-skilled occupation). Agreeableness is strongly negatively correlated to labor earnings, ­ a pattern also observed for the labor force of Peru and for women in Germany. By contrast, using more complex math operation at work and in daily life (a proxy for the stock of cognitive skills) is only correlated with being a high- skilled worker and not with labor earnings or job formality. Both cognitive skills (use of numeracy) and socioemotional skills (conscientiousness, grit, agreeable- ness, and decision-making style) are strongly correlated with employment and being involved in a productive activity (working, looking for a job, or studying). Both types of skills are also strongly correlated with tertiary education atten- dance, though through a different set of socioemotional skills (openness to expe- rience, agreeableness, and hostile attribution bias). As a caveat, we expect that the large correlations of socioemotional skills measures with labor market and tertiary education outcomes are overestimations due to the absence of well-​ measured cognitive skill controls (such as a reading test score like in the STEP ­ Household Survey). Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Appendix B Methodologies Used in This Study The empirical analysis in this study seeks to assess the degree to which certain types of cognitive and socioemotional skills affect labor market outcomes and tertiary education trajectories—that is, to determine how those skills, however acquired, influence outcomes. It uses two approaches. Estimating correlations between disaggregated measures of skills and outcomes. (Method A) The first approach follows a standard micro-econometric specification to esti- mate the relationship between a given labor market or schooling outcome and a set of skills: Yi = α + β1A i + β2 X i + ε i(B.1) where Yi is a labor market or schooling outcome (for example, wages or tertiary education attendance); Ai represents all ability (skills) that affects the outcome (for example, verbal literacy or conscientiousness); and Xi is a set of other factors that affect Yi (for example, gender or age). The true abilities Ai of individuals are unobserved (latent). A common proxy for abilities is years of schooling or school levels. However, educational attain- ment is a poor measurement of actual ability, because many of the skills and personality traits that shape an individual’s success are acquired outside the classroom and at every level of schooling students acquire skills very differently across schools and countries (Hanushek and Woessman 2008). Nonetheless, the data used in the four case studies provide a set of measured test scores, Ti, that capture various dimensions of cognitive and socioemotional skills. Assuming that Ti measures all skills captured in Ai in equation (B.1), the equation can be rewritten as follows: Yi = α + β1Ti + β2 X i + v i (B.2) Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   73   74 Methodologies Used in This Study Under the assumption that our sets of Ti perfectly measure vector Ai, we can estimate equation (B.2) using ordinary least squares (for wages) or logit (for discrete labor outcomes) regressions without any ability bias, and b1 will yield the return to each skill captured by vector Ti. However, a growing body of literature shows that measured skills Ti, capture Ai with error, so there may be a dependence between Ti and the error term, ni (that is, Cov (Ti,ni) ≠ 0). In that case, measurement errors and omitted variables could produce biased estimates of b1. Estimating correlations between unobservable aggregated skills and outcomes (Method B) The second approach is derived from a structural estimation of latent skills (Heckman, Stixrud, and Urzúa 2006; Sarzosa and Urzúa 2016). It provides con- sistent estimates even in the presence of measurement error. The outcomes of interest, Yi, are a function of latent skills and other factors influencing them, as shown in the following reduced-form equation: Yi = α Ai θ A + α Bi θB + βi X i + e i (B.3) where the θ’s are the latent skills factors that include measured Ti and dimen- sions of unobserved heterogeneity, and Xi are observable controls. Available test scores in the data sets, Ti, are only proxies of the true “latent variables” to be used for the estimation. They are treated as realizations of a “score-production func- tion,” as presented in equation (B.4), whose inputs are observable and unobserv- able characteristics: Ti = α Aiθ A + α Biθ B + βi X i + µ i (B.4) A system of production functions of test scores (equation B.4) can then be used to nonparametrically identify the distributions of the latent abili- ties. The main restrictions to this process are the assumptions that (a) latent skills factors θ’s are orthogonal and independent to one another and (b) the system includes at least three test scores per skill (Carneiro, Hansen, and Heckman 2003). Once the distributional parameters are estimated, the association between latent skills and outcomes can be estimated using a maximum-likelihood method. Given its demanding data requirements, a practical limitation of this approach is that only a limited number of aggregated skills factors can be retrieved from the data sets and studied. Four factors could be constructed using the Peru data: aggregated cognitive skills (math and verbal skills); “stability” personality traits (traits related to consistency in motivation, mood, and social interactions and including resilience, agreeableness, and conscientiousness); “plasticity” personality traits (traits related to striving toward personal growth and including extrover- sion and openness to experience); and grit. For Bolivia and Colombia, two factors Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Methodologies Used in This Study 75 were constructed: an aggregated factor of cognitive skills (reading and language skills) and an aggregated factor of socioemotional skills (related to achieving goals, working with others, managing emotions and including openness to experi- ence, conscientiousness, extroversion, agreeableness, resilience, grit, decision making, and hostile attribution bias). Use of Both Approaches and Caveats The structural estimation (Method B) of latent skills is a state-of the-art method that more fully captures the association of skills and labor market outcomes, but it does so on a limited number of factors (four for Peru, two each for Bolivia and Colombia) given the data used for this study. Analyses derived from stan- dard econometric computations (Method A) yield less precise but more detailed information on the contribution of specific measures of traits or cogni- tive abilities (twelve for Peru, eight for Bolivia and Colombia, and seven for El Salvador). Neither method allows us to claim causation between our measures of skills and improved labor market outcomes. Because of the cross-sectional nature of the data, both the outcomes of interest and the measures of skills (or latent skills factors) are observed simultaneously; employment status or work type could thus also influence skills, particularly socioemotional ones. Although skills are popularly thought to be set by early adulthood, emerging evidence suggests otherwise. New studies from the United States and Germany suggest that participation in the labor market affects emotional stability and provokes significant change in agreeableness, conscientiousness, and openness to experience (Gottschalk 2005; Boyce and others 2015). References Boyce, C. J., A. M. Wood, M. Daly, and C. Sedikides. 2015. “Personality Change Following Unemployment.” Journal of Applied Psychology 100 (4): 991–1011. Carneiro, P., K. T. Hansen, and J. J. Heckman. 2003. “2001 Lawrence R. Klein Lecture: Estimating Distributions of Treatment Effects of Uncertainty on College Choice.”. International Economic Review 44: 361–422. Gottschalk, P. 2005. “Can Work Alter Welfare Recipients’ Beliefs?” Journal of Policy Analysis and Management 24 (3): 485–98. Hanushek, E. A., and L. Woessmann. 2008. “The Role of Cognitive Skills in Economic Development.” Journal of Economic Literature 46 (3): 607–68. Heckman, J. J., J. Stixrud, and S. Urzúa. 2006. “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior.” Journal of Labor Economics 24 (3): 411–82. Sarzosa, M., and S. Urzúa. 2016. “Factor Models for Unobserved Heterogeneity in Stata: The Heterofactor Command.” Stata Journal 16 (1): 197–228. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Appendix C Summary of Associations between Measures of Skills and Labor Market and Tertiary Education Outcomes in Bolivia, Colombia, El Salvador, and Peru Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   77   78 Table C.1 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in Bolivia, 2012 Labor Formal High-skilled Wage Active in labor Tertiary education Type of skills Dimension Skill earnings job occupation worker Employment market or studying attendance Basic cognitiveBasic academic Reading proficiency (ability to ♦ ♦ ♦ knowledge understand, evaluate, use, and and reasoning engage with written text) Socioemotional Achieving goals Conscientiousness (tendency to be ♦ ♦ organized, responsible, and hardworking) Grit (perseverance and passion for ♦ long-term goals) Openness to experience (appreciation ♦ ♦ for art, learning, and unusual ideas) Working with Agreeableness (pro-social behavior, others cooperative orientation to others) Extroversion (sociability and dominance in social situations) Managing Resilience (management of negative emotions emotions) Hostile attribution bias (tendency to ♦ ♦ ♦ perceive hostile intents in others) Decision making (manner in which ♦ individual approaches decision situations) Table C.2 Summary of Associations between Aggregated Measures of Skills and Labor and Tertiary Education Outcomes in Bolivia, 2012 Active in labor Tertiary Labor High-skilled market or education Type of skills Dimension Skills earnings Formal job occupation Employment studying attendance Basic cognitive Basic academic Cognitive skills (reading ♦ ♦ ♦ ♦ knowledge and and language skills) reasoning Socioemotional Achieving goals, working Socioemotional skills ♦ ♦ ♦ with others, and (achieving goals, managing emotions working with others, managing emotions) Source: Bolivia STEP Household Surveys (2012). Note: Calculations for labor earnings control for gender, age, mother education, cities of living and their metropolitan areas, industry, and ethnicity. Calculation for other labor market outcomes control for the same variables and a self-reported categorical variable on parents’ involvement in ones’ education at the age of 12. 79 80 Table C.3 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in Colombia, 2012 Active in Tertiary Labor Formal High-skilled Wage labor market education Type of skills Dimension Skills earnings job occupation worker Employment or studying attendance Basic cognitive Basic academic Reading proficiency (ability to ♦ ♦ ♦ ♦ ♦ knowledge understand, evaluate, use, and and reasoning engage with written text) Socioemotional Achieving goals Conscientiousness (tendency to be ♦ ♦ organized, responsible, and hardworking) Grit (perseverance and passion for long-term goals) Openness to experience (appreciation ♦ ♦ ♦ ♦ for art, learning, and unusual ideas) Working with Agreeableness (pro-social behaviors, ♦ others cooperative orientation to others) Extroversion (sociability and dominance in social situations) Managing Resilience (management of negative ♦ emotions emotions) Hostile attribution bias (tendency to ♦ ♦ perceive hostile intents in others) Decision making (manner in which ♦ ♦ individual approaches decision situations) Table C.4 Summary of Associations between Aggregated Measures of Skills and Labor and Tertiary Education Outcomes in Colombia, 2012 Active in Tertiary Labor Formal High-skilled Wage labor market or education Type of skills Dimension Skills earnings job occupation worker Employment studying attendance Basic cognitive Basic academic Cognitive skills (reading ♦ ♦ ♦ ♦ ♦ ♦ knowledge and and language skills) reasoning Socioemotional Achieving goals, Socioemotional skills ♦ ♦ working with others, (achieving goals, and managing working with others, emotions managing emotions) Source: Colombia STEP Household Survey (2012). Note: Calculations for labor earnings control for gender, age, mother education, and cities of living and their metropolitan areas. Calculation for other labor market outcomes control for the same variables and a self-reported categorical variable on parents’ involvement in ones’ education at the age of 12. 81 82 Table C.5 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in El Salvador, 2013 Active in Tertiary Labor Formal High-skilled Wage labor market education Type of skills Dimension Skills earnings job occupation worker Employment or studying attendance Basic cognitive Basic academic Math ability (use and complexity of ♦ ♦ ♦ ♦ ♦ knowledge basic math operations used at and and reasoning outside of work) Socioemotional Achieving goals Conscientiousness (tendency to be ♦ ♦ ♦ ♦ organized, responsible, and hardworking) Grit (perseverance and passion for ♦ long-term goals) Openness to experience ♦ ♦ (appreciation for art, learning, and unusual ideas) Working with Agreeableness (pro-social behaviors, ♦ ♦ ♦ others cooperative orientation to others) Extroversion (sociability and ♦ dominance in social situations) Managing Resilience (management of negative ♦ ♦ ♦ emotions emotions) Hostile attribution bias (tendency to perceive hostile intents in others) Decision making (manner in which ♦ ♦ individual approaches decision situations) Source: El Salvador Skills Household Survey (2013). Note: Calculations for labor earnings control for gender, age, mother education, and cities of living and their metropolitan areas. Calculation for other labor market outcomes control for the same variables and a self-reported categorical variable on parents’ involvement in ones’ education at the age of 12. Table C.6 Summary of Associations between Disaggregated Measures of Skills and Labor and Tertiary Education Outcomes in Peru, 2010 Labor White-collar Wage Type of skills Dimension Skills earnings Formal job occupation worker Employment Basic cognitive Basic academic Memory (short-term memory, representative of knowledge and working memory) reasoning Math ability (basic math operations) ♦ ♦ Verbal ability (receptive vocabulary and verbal ability ♦ ♦ ♦ ♦ of adult subjects) Verbal fluency (speed and ease with which individual ♦ ♦ ♦ accesses words from memory) Socioemotional Achieving goals Conscientiousness (tendency to be organized, responsible, and hardworking) Grit (perseverance and passion for long-term goals) ♦ ♦ ♦ Openness to experience (appreciation for art, learning, ♦ and unusual ideas) Working with others Agreeableness (pro-social behaviors, cooperative ♦ orientation to others) Extroversion (sociability and dominance in social situations) Managing emotions Resilience (management of negative emotions) ♦ 83 84 Table C.7 Summary of Associations between Aggregated Measures of Skills and Labor and Tertiary Education Outcomes in Peru, 2010 Labor White-collar Wage Type of skills Dimension Skills earnings Formal job occupation worker Employment Basic cognitive Basic academic knowledge Cognitive skills (reading and language skills) ♦ ♦ ♦ ♦ and reasoning Socioemotional Achieving goals Perseverance personality traits (perseverance of ♦ ♦ effort and goals, and consistency of interest) Achieving goals and Plasticity personality traits (striving toward ♦ working with others personal growth; include extroversion and openness to experience) Achieving goals, working Stability personality traits (consistency in ♦ with others, and motivation, mood, and social interactions; managing emotions include resilience, agreeableness, and conscientiousness) Source: Peru ENHAB (2010). Note: Estimates control for gender, age, speaking indigenous language as mother tongue, being first-born child, region, industry, parents’ education, and distance to school at young age. Appendix D Regression Results from Country Studies In all tables, formal workers are workers who receive social security benefits through their job. High-skilled workers are senior officials and managers, profes- sionals, or technicians. Low- and middle-skilled workers are clerks, service work- ers, machine operators, or laborers. Classification is based on the International Labour Organization’s 1988 International Standard Classification of Occupations (ISCO). Table D.1 Structural Estimates of Conditional Correlations between Labor Market and Tertiary Education Outcomes and Latent Skills in Bolivia, 2012 Working, looking Tertiary Log of hourly Formal High-skilled for a job, or in education labor earnings worker worker Employed school attainment Item (1) (2) (3) (4) (5) (6) Cognitive skills 0.1491*** 0.1765** 0.3381*** –0.0509 0.0638 0.9251*** (0.044) (0.069) (0.052) (0.061) (0.075) (0.084) Socioemotional skills 0.0679 0.8237** 0.0675 –0.0809 –0.6540* 0.4994* (0.194) (0.330) (0.219) (0.254) (0.385) (0.290) Number of observations 1,057 1,448 1,807 1,807 1,807 1,378 Source: Bolivia STEP Household Survey (2012). Note: Calculations for labor earnings are estimations from ordinary least squares regressions that control for gender, age, mother’s education, and city of residence and metropolitan area. Calculations for other labor market outcomes are estimations from probit regressions that control for same variables and a self-reported categorical variable that captures parents’ involvement in child’s education at age 12. Estimations are raw coefficients and were produced using Sarzosa and Urzúa (2016). Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   85   86 Regression Results from Country Studies Table D.2 Conditional Correlations between Labor Earnings, Formality, and Occupational Status and Measures of Skills and Schooling in Bolivia, 2012 Log of hourly labor earnings Formal worker High-skilled worker Without With Without With Without With schooling schooling schooling schooling schooling schooling Item (1) (2) (3) (4) (5) (6) Cognitive skill Reading proficiency 0.086* 0.016 0.022 −0.263* 0.375** 0.155 (0.05) (0.05) (0.12) (0.15) (0.16) (0.16) Socioemotional skill Extroversion −0.022 −0.031 0.049 0.016 −0.070 −0.087 (0.04) (0.04) (0.11) (0.10) (0.11) (0.12) Conscientiousness −0.053 −0.054 0.121 0.137 0.145 0.168 (0.04) (0.04) (0.13) (0.11) (0.11) (0.11) Openness to experience 0.110*** 0.095*** 0.178 0.130 0.018 −0.062 (0.03) (0.03) (0.12) (0.13) (0.09) (0.10) Emotional stability −0.013 −0.014 0.070 0.058 −0.065 −0.062 (0.03) (0.03) (0.10) (0.10) (0.09) (0.10) Agreeableness 0.004 0.011 −0.033 0.007 −0.142 −0.139 (0.04) (0.04) (0.10) (0.11) (0.12) (0.12) Grit 0.077** 0.077** 0.007 −0.025 0.051 0.029 (0.04) (0.04) (0.11) (0.11) (0.11) (0.11) Hostile attribution bias −0.082* −0.077* −0.161 −0.162 0.039 0.073 (0.04) (0.04) (0.11) (0.11) (0.10) (0.09) Decision making −0.026 −0.031 0.032 0.011 0.022 0.014 (0.03) (0.03) (0.10) (0.09) (0.11) (0.11) Education Below primary 0.244 1.462* −0.647 (0.19) (0.78) (0.56) Secondary 0.495*** 1.999*** 0.310 (0.14) (0.63) (0.42) Upper-secondary 0.376** 2.038*** 0.547 (0.19) (0.67) (0.51) Vocational tertiary 0.603*** 2.814*** 1.501*** (0.17) (0.65) (0.56) General tertiary 0.861*** 3.491*** 2.502*** (0.18) (0.69) (0.50) Number of observations 1,274 1,274 1,445 1,445 1,445 1,445 Source: Bolivia STEP Household Survey (2012). Note: Calculations for log of hourly labor earnings are ordinary least squares regressions that control for gender, age, age-squared, mother’s education, city and metropolitan area, industry, and ethnicity. Calculations for being a formal or a high-skilled worker control for the same variables and a self-reported categorical variable that captures parents’ involvement in child’s education at age 12 (three levels); results are odds ratios. Measures of reading proficiency and socioemotional skills are standardized. Regressions coefficients and standards errors of reading proficiency are average of 10 estimations using plausible values. Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Regression Results from Country Studies 87 Table D.3 Conditional Correlations between Employment, Activity, and Educational Trajectory and Measures of Skills and Schooling in Bolivia, 2012 Working, looking for Tertiary education Employed a job, or in school attendance Without With schooling schooling Without schooling Without schooling Item (1) (2) (3) (4) Cognitive skill Reading proficiency 0.030 −0.074 −0.039 0.537*** (0.12) (0.11) (0.11) (0.20) Socioemotional skill Extroversion 0.092 0.053 0.113 −0.105 (0.08) (0.08) (0.09) (0.15) Conscientiousness 0.138* 0.132* 0.137* −0.099 (0.07) (0.07) (0.08) (0.14) Openness to experience −0.098 −0.129 −0.015 0.463*** (0.08) (0.08) (0.09) (0.16) Emotional stability −0.040 −0.047 0.006 0.222 (0.10) (0.10) (0.11) (0.17) Agreeableness −0.014 0.003 −0.045 −0.033 (0.08) (0.08) (0.09) (0.17) Grit −0.055 −0.066 0.045 −0.099 (0.09) (0.09) (0.11) (0.17) Hostile attribution bias 0.147* 0.121 0.222** 0.147 (0.09) (0.09) (0.11) (0.15) Decision making −0.087 −0.060 −0.115 −0.124 (0.09) (0.09) (0.10) (0.16) Education Below primary −0.798 (0.53) Secondary −0.352 (0.38) Upper-secondary −0.150 (0.45) Vocational tertiary 0.327 (0.51) General tertiary 1.183** (0.52) Number of observations 1,976 1,976 1,976 1,063 Source: Bolivia STEP Household Survey (2012). Note: Results are odds ratios of logit regressions that control for gender, age, age-squared, mother’s education, city and metropolitan area, industry, ethnicity, and a self-reported categorical variable that captures parents’ involvement in child’s education at age 12 (three levels). Measures of reading proficiency and socioemotional skills are standardized. Regression coefficients and standard errors of reading proficiency are average of 10 estimations using plausible values. Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 88 Regression Results from Country Studies Table D.4 Structural Estimates of Conditional Correlations between Labor Market and Tertiary Education Outcomes and Latent Skills in Colombia, 2012 Log of Working, Tertiary hourly labor Formal High-skilled looking for a job, education earnings worker worker Employed or in school attendance Items (1) (2) (3) (4) (5) (6) Cognitive skills 0.134*** 0.276*** 0.252*** 0.023 0.112** 0.988*** (0.032) (0.052) (0.04) (0.042) (0.047) (0.076) Socioemotional skills −0.026 −0.004 0.046 0.013 0.143*** 0.170*** (0.028) (0.044) (0.035) (0.04) (0.045) (0.049) Number of observations 1,363 1,560 2,328 2,089 2,328 1,692 Source: Colombia STEP Household Survey (2012). Note: Calculations for labor earnings are ordinary least squares regressions that control for gender, age, mother’s education, and city and metropolitan area. Calculations for other labor market outcomes are probit regressions that control for same variables and a self-reported categorical variable that captures parents’ involvement in child’s education at age 12 (three levels). Estimations are raw coefficients and were produced using Sarzosa and Urzúa (2016). Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Table D.5 Conditional Correlations between Labor Earnings, Formality, and Occupational Status and Measures of Skills and Schooling in Colombia, 2012 Log of hourly labor earnings Formal worker High-skilled worker Without With Without With Without With schooling schooling schooling schooling schooling schooling Item (1) (2) (3) (4) (5) (6) Cognitive skill Reading proficiency 0.161*** 0.065 0.063*** 0.014 0.141*** 0.061*** (0.05) (0.06) (0.02) (0.02) (0.02) (0.02) Socioemotional skill Extroversion −0.009 0.000 0.001 0.005 −0.004 −0.000 (0.04) (0.04) (0.02) (0.02) (0.01) (0.01) Conscientiousness −0.034 −0.034 −0.003 −0.003 0.001 −0.002 (0.04) (0.04) (0.02) (0.02) (0.01) (0.01) Openness to experience 0.082** 0.078** −0.020 −0.022 0.020 0.017 (0.03) (0.03) (0.02) (0.02) (0.02) (0.01) Emotional stability 0.008 −0.015 0.027 0.016 0.006 −0.007 (0.04) (0.04) (0.02) (0.02) (0.01) (0.01) Agreeableness 0.023 0.015 −0.011 −0.014 −0.007 −0.003 (0.03) (0.03) (0.02) (0.02) (0.01) (0.01) Grit −0.030 −0.043 −0.020 −0.025 0.013 0.007 (0.04) (0.04) (0.02) (0.02) (0.01) (0.01) Hostile attribution bias −0.003 0.023 −0.041** −0.030* −0.019 −0.003 (0.03) (0.03) (0.02) (0.02) (0.02) (0.01) Decision making 0.013 −0.007 0.012 0.002 0.019 −0.002 (0.04) (0.04) (0.02) (0.02) (0.01) (0.01) table continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Regression Results from Country Studies 89 Table D.5  Conditional Correlations between Labor Earnings, Formality, and Occupational Status and Measures of Skills and Schooling in Colombia, 2012 (continued) Log of hourly labor earnings Formal worker High-skilled worker Without With Without With Without With schooling schooling schooling schooling schooling schooling Item (1) (2) (3) (4) (5) (6) Education Below primary 0.015 0.103 0.091 (0.11) (0.08) (0.07) Upper secondary 0.289*** 0.198*** 0.163*** (0.10) (0.05) (0.04) Vocational tertiary 0.371*** 0.263*** 0.278*** (0.11) (0.05) (0.04) General tertiary 0.880*** 0.348*** 0.566*** (0.15) (0.06) (0.05) Number of observations 1,372 1,372 1,576 1,576 1,801 1,801 R-squared 0.11 0.16 Source: Colombia STEP Household Survey (2012). Note: Calculations for labor earnings are ordinary least squares regressions that control for gender, age, mother’s education, and cities of residence and their metropolitan areas. Calculations for other labor market outcomes are logit regressions that control for the same variables and a self- reported categorical variable that captures parents’ involvement in child’s education at age 12 (three levels). Average marginal effects are reported for logit regressions and reflect the changes in the probability of being observed in a labor or school participation situation with respect to the variables evaluated at the mean. Measures of reading proficiency and socioemotional skills are standardized. Regressions coefficients and standard errors of reading proficiency are the average of 10 estimations using plausible values. Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Table D.6 Conditional Correlations between Employment, Activity, and Educational Trajectory and Measures of Skills and Schooling in Colombia, 2012 Working, looking for Tertiary education Employed a job, or in school attendance Without With schooling schooling Without schooling Without schooling Item (1) (2) (3) (4) Cognitive skill Reading proficiency 0.003 −0.009 0.021* 0.199*** (0.02) (0.02) (0.01) (0.02) Socioemotional skill Extroversion −0.007 −0.007 0.009 −0.009 (0.02) (0.02) (0.01) (0.01) Conscientiousness 0.044*** 0.045*** 0.023** 0.002 (0.02) (0.02) (0.01) (0.01) Openness to experience 0.011 0.010 0.018* 0.045*** (0.02) (0.01) (0.01) (0.01) Emotional stability 0.018 0.015 0.009 0.048*** (0.02) (0.02) (0.01) (0.01) table continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 90 Regression Results from Country Studies Table D.6  Conditional Correlations between Employment, Activity, and Educational Trajectory and Measures of Skills and Schooling in Colombia, 2012 (continued) Working, looking for Tertiary education Employed a job, or in school attendance Without With schooling schooling Without schooling Without schooling Item (1) (2) (3) (4) Agreeableness −0.016 −0.016 −0.019* 0.001 (0.02) (0.02) (0.01) (0.01) Grit 0.005 0.004 0.003 0.003 (0.02) (0.02) (0.01) (0.01) Hostile attribution bias −0.010 −0.008 −0.009 −0.049*** (0.01) (0.01) (0.01) (0.01) Decision making −0.031* −0.033** −0.003 0.055*** (0.02) (0.02) (0.01) (0.01) Education Below primary −0.063 (0.06) Upper-secondary 0.003 (0.04) Vocational tertiary 0.053 (0.05) General tertiary 0.066 (0.07) Number of observations 2,117 2,117 2,356 1,717 Source: Colombia STEP Household Survey (2012). Note: Calculations are logit regressions that control for age, mother’s education, and city and metropolitan area, and a self-reported categorical variable that captures parents’ involvement in child’s education at age 12 (three levels). Average marginal effects are reported for logit regressions and reflect the changes in the probability of being observed in a labor or school participation situation with respect to the variables evaluated at the mean. Measures of reading proficiency and socioemotional skills are standardized. Regressions coefficients and standard errors of reading proficiency are the average of the 10 estimations using plausible values. Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Regression Results from Country Studies 91 Table D.7 Conditional Correlations between Labor Earnings, Formality, and Occupational Status and Measures of Skills and Schooling in El Salvador, 2013 Log of hourly labor earnings Formal worker High-skilled worker Without With Without With Without With schooling schooling schooling schooling schooling schooling Item (1) (2) (3) (4) (5) (6) Cognitive skill Math ability 0.118 −0.010 0.022 −0.008 0.059*** 0.020 (0.14) (0.14) (0.02) (0.02) (0.02) (0.01) Socioemotional skill Extroversion 0.169 0.161 −0.000 −0.001 −0.007 −0.004 (0.13) (0.13) (0.02) (0.02) (0.01) (0.01) Conscientiousness 0.279** 0.260* 0.033* 0.026 0.001 −0.002 (0.14) (0.14) (0.02) (0.02) (0.01) (0.01) Openness to experience −0.064 −0.127 0.031* 0.012 0.027* 0.002 (0.14) (0.14) (0.02) (0.02) (0.02) (0.01) Emotional stability 0.265** 0.224* 0.028* 0.017 0.029** 0.012 (0.13) (0.13) (0.02) (0.02) (0.01) (0.01) Agreeableness −0.419*** −0.368*** −0.001 0.009 −0.020 −0.005 (0.14) (0.13) (0.02) (0.02) (0.01) (0.01) Grit −0.108 −0.119 −0.025 −0.024 −0.010 −0.015 (0.13) (0.13) (0.02) (0.02) (0.01) (0.01) Hostile attribution bias 0.151 0.185 −0.010 −0.002 −0.018 −0.002 (0.14) (0.14) (0.02) (0.02) (0.01) (0.01) Decision making 0.240* 0.245* 0.012 0.013 0.016 0.016 (0.13) (0.13) (0.02) (0.02) (0.01) (0.01) Education Primary 0.369 0.313*** 0.065 (0.36) (0.06) (0.09) General secondary 1.240*** 0.343*** 0.303*** (0.47) (0.07) (0.08) Vocational tertiary 1.013** 0.379*** 0.336*** (0.43) (0.06) (0.08) Tertiary 2.121*** 0.663*** 0.579*** (0.54) (0.07) (0.08) Number of observations 1,105 937 1,197 1,196 1,225 1,028 Source: El Salvador Skills Household Survey (2013). Note: Calculations for labor earnings are ordinary least squares regressions that control for gender, age, mother’s education, and cities of residence and their metropolitan areas. Calculation for other labor market outcomes are logit regressions that control for the same variables and a self- reported categorical variable that captures parents’ involvement in child’s education at age 12 (three levels). Average marginal effects are reported for logit regressions and reflect the changes in the probability of being observed in a labor or school participation situation with respect to the variables evaluated at the mean. Measures of socioemotional skills are standardized. Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 92 Regression Results from Country Studies Table D.8 Conditional Correlations between Employment, Activity, and Educational Trajectory with Measures of Skills and Schooling in El Salvador, 2013 Working, Tertiary looking for a education Employed job, or in school attendance Without With Without Without schooling schooling schooling schooling Item (1) (2) (3) (4) Cognitive skill Math ability 0.044*** 0.039*** 0.087*** 0.057*** (0.01) (0.01) (0.01) (0.01) Socioemotional skill Extroversion 0.001 0.002 0.002 −0.003 (0.01) (0.01) (0.01) (0.01) Conscientiousness 0.029** 0.027** 0.027*** 0.012 (0.01) (0.01) (0.01) (0.01) Openness to experience 0.009 0.006 0.008 0.023* (0.01) (0.01) (0.01) (0.01) Emotional stability −0.005 −0.007 0.008 0.014 (0.01) (0.01) (0.01) (0.01) Agreeableness −0.010 −0.008 −0.020** −0.022* (0.01) (0.01) (0.01) (0.01) Grit 0.022* 0.021* 0.029*** 0.009 (0.01) (0.01) (0.01) (0.01) Hostile attribution bias 0.004 0.005 0.004 −0.024** (0.01) (0.01) (0.01) (0.01) Decision making −0.037*** −0.038*** −0.024*** −0.005 (0.01) (0.01) (0.01) (0.01) Education Primary −0.001 (0.04) General secondary 0.053 (0.05) Vocational tertiary 0.009 (0.04) Tertiary 0.116** (0.05) Number of observations 1,787 1,490 2,064 1,081 Source: El Salvador Skills Household Survey (2013). Note: Calculations are logit regressions that control for age, mother’s education, and city and metropolitan area, and a self-reported categorical variable that captures parents’ involvement in child’s education at age 12 (three levels). Average marginal effects are reported for logit regressions and reflect the changes in the probability of being observed in a labor or school participation situation with respect to the variables evaluated at the mean. Measures of socioemotional skills are standardized. Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Regression Results from Country Studies 93 Table D.9 Structural Estimates of Conditional Correlations between Labor Market Outcomes and Latent Skills Factors in Peru, 2010 Log of hourly Formal White-collar Wage labor earnings worker worker worker Employed Items (1) (2) (3) (4) (5) Cognitive skills 0.1344*** 0.2223*** 0.3087*** 0.2693*** −0.0019 (0.037) (0.065) (0.072) (0.069) (0.054) Grit 0.0532 0.0667 0.1535** 0.1510** 0.0600 (0.038) (0.063) (0.070) (0.068) (0.053) Plasticity −0.0752 −0.2337 −0.3733 −0.3735 0.6729** (0.196) (0.329) (0.380) (0.327) (0.275) Stability 0.3986* −0.0261 0.0633 −0.3769 −0.0993 (0.227) (0.332) (0.457) (0.383) (0.291) Number of observations 748 789 789 789 1,265 Source: Peru ENHAB (2010). Note: Calculations are ordinary least squares regressions (for hourly labor earnings) and probit regressions (for other labor market outcomes) that control for gender, age, speaking indigenous language as mother tongue, being first born, region, industry, parents’ education, and distance to school. Estimates are raw coefficients and were produced using Sarzosa and Urzúa (2016). Standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 94 Table D.10 Conditional Correlations between Labor Outcomes and Measures of Skills in Peru, 2010 Log of hourly wage Formal worker White-collar worker Wage worker Employed (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Without With Without With Without With Without With Without With Item schooling schooling schooling schooling schooling schooling schooling schooling schooling schooling Math ability 0.096*** 0.047 −0.008 −0.022 0.046** 0.016 0.013 −0.011 0.012 0.008 (0.035) (0.035) (0.021) (0.022) (0.018) (0.018) (0.020) (0.020) (0.014) (0.014) Working memory 0.053* 0.042 0.025 0.022 0.018 0.012 0.013 0.008 0.007 0.006 (0.030) (0.030) (0.019) (0.019) (0.015) (0.015) (0.018) (0.018) (0.013) (0.013) Verbal ability 0.103*** 0.039 0.050** 0.032 0.045** 0.008 0.061*** 0.030 −0.011 −0.017 (0.035) (0.036) (0.021) (0.022) (0.019) (0.019) (0.022) (0.023) (0.015) (0.016) Verbal fluency 0.049 0.028 0.016 0.010 0.026* 0.012 0.025 0.014 0.027** 0.025* (0.033) (0.032) (0.018) (0.018) (0.015) (0.015) (0.017) (0.017) (0.013) (0.013) Conscientiousness −0.043 −0.044 −0.004 −0.004 0.002 0.002 −0.005 −0.005 −0.001 −0.001 (0.036) (0.035) (0.021) (0.021) (0.018) (0.018) (0.019) (0.019) (0.015) (0.015) Kindness −0.093*** −0.086** 0.000 0.002 −0.007 −0.004 0.006 0.009 −0.008 −0.008 (0.035) (0.034) (0.019) (0.019) (0.016) (0.015) (0.019) (0.018) (0.013) (0.013) Cooperation −0.058* −0.056* 0.018 0.019 −0.011 −0.009 −0.026 −0.024 0.002 0.003 (0.030) (0.030) (0.018) (0.018) (0.016) (0.016) (0.017) (0.016) (0.013) (0.013) Emotional stability 0.074** 0.067** −0.021 −0.023 0.001 −0.003 −0.014 −0.017 −0.003 −0.003 (0.032) (0.031) (0.019) (0.019) (0.017) (0.017) (0.018) (0.018) (0.013) (0.013) Extroversion 0.025 0.021 −0.007 −0.008 0.013 0.011 −0.010 −0.012 0.010 0.010 (0.034) (0.033) (0.019) (0.019) (0.016) (0.016) (0.018) (0.018) (0.014) (0.014) Openness to experience 0.015 0.009 −0.008 −0.009 −0.032* −0.035** −0.016 −0.019 0.015 0.014 (0.032) (0.031) (0.020) (0.020) (0.017) (0.016) (0.019) (0.018) (0.014) (0.014) Perseverance of effort 0.058 0.044 0.035* 0.030 0.050*** 0.039** 0.054*** 0.045*** 0.033** 0.032** (0.037) (0.037) (0.019) (0.019) (0.016) (0.016) (0.017) (0.017) (0.014) (0.014) table continues next page Table D.10  Conditional Correlations between Labor Outcomes and Measures of Skills in Peru, 2010 (continued) Log of hourly wage Formal worker White−collar worker Wage worker Employed (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Without With Without With Without With Without With Without With Item schooling schooling schooling schooling schooling schooling schooling schooling schooling schooling Consistency of interest 0.004 0.011 −0.025 −0.024 0.002 0.005 −0.004 −0.001 0.005 0.005 (0.031) (0.030) (0.018) (0.018) (0.016) (0.015) (0.016) (0.016) (0.012) (0.012) Years of schooling 0.065*** 0.018*** 0.039*** 0.032*** 0.006 (0.012) (0.007) (0.006) (0.006) (0.005) Number of observations 822 822 865 865 865 865 865 865 1,390 1,390 R-squared 0.212 0.242 0.112 0.119 0.348 0.383 0.253 0.275 0.233 0.234 Source: Peru ENHAB (2010) . Note: Regressions are ordinary least squares regressions that control for gender, age, speaking indigenous language as mother tongue, being first born, region, industry, parents’ education, and distance to school. Robust standard errors are in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. 95 Appendix E Cross-Country Variations in Associations between Skills Dimensions and Labor Market and Tertiary Education Outcomes Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   97   98 Table E.1 Cross-Country Variations in Associations between Skills Dimensions and Labor Market and Tertiary Education Outcomes in Bolivia, Colombia, El Salvador, and Peru Active in Tertiary High-skilled Wage labor market education Type of skill Dimension Skill Earnings Formal job occupation worker Employment or studyinga attendancea Basic cognitiveBasic academic Memory (short-term memory, Bolivia, Colombia, Bolivia, El Salvador, El Salvador, Colombia, Bolivia, knowledge representative of working memory) Colombia, Peru Colombia, Peru Peru El Salvador Colombia, and reasoning Math ability (basic math operations) Peru El Salvador, El Salvador Verbal ability (receptive vocabulary and Peru verbal ability of adult subjects) Verbal fluency (speed and ease with which individual accesses words from memory) Reading proficiency (ability to understand, evaluate, use, and engage with written text) Socioemotional Achieving Conscientiousness (tendency to be Bolivia, El Salvador El Salvador, Colombia, Bolivia, Colombia, Bolivia, goals organized, responsible, and Colombia, Peru Peru Colombia, El Salvador Colombia, hardworking) El Salvador El Salvador, El Salvador Grit (perseverance and passion for Peru long-term goals) Openness to experience (appreciation for art, learning, and unusual ideas) Working with Agreeableness (pro-social behaviors, El Salvador, None None None None Colombia, El Salvador others cooperative orientation to others) Peru El Salvador Extroversion (sociability and dominance in social situations) Managing Resilience (management of negative Bolivia, Colombia, El Salvador Bolivia Bolivia. Bolivia, Colombia, emotions emotions) El Salvador, El Salvador Colombia, El Salvador El Salvador Hostile attribution bias (tendency to perceive Peru El Salvador hostile intents in others) Decision making (how individuals approach decision situations) Sources: Bolivia and Colombia: STEP Household Surveys (2012); El Salvador: El Salvador Skills Survey 2013; Peru: ENHAB 2010. Note: Table shows statistically significant associations with outcome (at the 10 percent, 5 percent, or 1 percent levels) for given dimensions of skills. Associations were identified using ordinary least squares or logit regressions controlling for a range of characteristics. See appendixes D for details by country. a. Outcome not studied for Peru. Appendix F Inventory of Promising Interventions to Foster Socioemotional Skills Table F.1 Promising Programs That Foster Socioemotional Skills, by Target Age Group Program/country Target age group Early years 0–5 Carolina Abecedarian Project (United States) 0–5 Supplementation Study (Jamaica) 1–2 Early childhood development component of the national conditional cash transfer program Famílias en Acción (Families in Action) (Colombia) 1–2 Perry Preschool (United States) 3–4 Tools of the Mind (Chile, United States) 3–4 Head Start (United States) 3–5 Middle childhood 6–11 Incredible Years (United States, Western Europe, elsewhere) 3–7 Project STAR (Student-Teacher Achievement Ratio) (United States) 5–6 Seattle Social Development Project (United States) 6–7 Montreal Longitudinal Experimental Study (Canada) 7–9 Escuela Nueva (New School) (Colombia, Vietnam, elsewhere) 6–11 Fútbol con Corazón (Football with Heart) (Colombia) 5–17 Escuela Amiga (Friendly School) (Peru) 6–17 Adolescence 12–18 Knowledge Is Power Program (KIPP) (United States) 5–19 Big Brothers Big Sisters (United States) 10–16 Empresários Pela Inclusão Social (EPIS) (Entrepreneurs for Social Inclusion) (Portugal) 13–15 Becoming a Man (United States) 15–16 Chicago One-Goal (United States) 15–18 Construye T (Build Yourself) (Mexico) 15–18 National Guard ChalleNGe (United States) 16–18 table continues next page Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5   99   100 Inventory of Promising Interventions to Foster Socioemotional Skills Table F.1  Promising Programs That Foster Socioemotional Skills, by Target Age Group (continued) Program/country Target age group Emerging adulthood 19–29 Youthbuild (United States, Central America, others) 16–24 Job Corps (United States) 16–24 Year-up (United States) 18–24 Jóvenes en Acción (Youth in Action) (Colombia) 18–25 Juventud y Empleo (Youth and Employment) (Dominican Republic) 16–28 Galpão Aplauso (Applause Warehouse) (Brazil) 15–29 Sustainable Transformation of Youth (Liberia) 18–35 Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Table F.2 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) Location Component Age Parental Guidance, Parental Teacher Social Health Program Purpose target School Home involvement counselor training training services services Skills targeted Carolina Abecedarian Reduce poverty 0–5 ♦ ♦ ♦ ♦ ♦ Reduction of aggressive Project (United States) behavior, prevention of antisocial behavior, academic skills Supplementation Study Improve health 1–2 ♦ ♦ ♦ ♦ Self-esteem, emotional (Jamaica) regulation, reduction of antisocial and oppositional behavior Early childhood Improve health 1–2 ♦ ♦ ♦ ♦ ♦ Self-management, social development relationships, regulation of component of the antisocial behavior, national conditional cognition, language cash transfer program (expressive and receptive), Famílias en Acción motor skills (Families in Action) (Colombia) Perry Preschool Reduce poverty 3–4 ♦ ♦ ♦ Reduction of aggressive (United States) behavior, academic motivation, IQ Tools of the Mind (Chile, Improving 3–4 ♦ ♦ ♦ Problem-solving, self- United States) academic management, learning, learning time management, teamwork, literacy, math Head Start (United Reduce poverty 3–5 ♦ ♦ ♦ ♦ ♦ ♦ Social relationships, self- States) concept, self-efficacy, self-regulation, emotional regulation, academic skills 101 102 Table F.3  Description of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) Program Purpose Age target Target population Description Start year References Carolina Abecedarian Reduce poverty 0–5 Disadvantaged Preschool intervention for 6-week-old babies 1972 Barnett and Masse Project (United States) African-American followed by a school-age treatment. Preschool (2007), Temple and children component was full-day childcare 5 days a week, Reynolds (2007), 50 weeks a year. School curriculum focused on Campbell and educational games but also included medical and others (2014) nutritional components. During grades 1–3, teachers and parents interacted biweekly. Supplementation Study Improve health 1–2 Growth-stunted Psychosocial stimulation based on weekly one-hour 1986 Gertler and others (Jamaica) toddlers from poor visits from community health workers over a (2014) neighborhoods two-year period that taught parenting skills and encouraged mothers and children to interact in ways (such as playing) that develop cognitive and socioemotional skills. Home-made toys were left after each visit. Program also provided nutritional supplements (milk formula) to subset of children. Early childhood Improve health 1–2 Children from Psychosocial stimulation based on weekly home visits 2009 Attanasio, Cattan, and development households with play demonstrations by female community others (2015) component of the participating in the leaders. Subset of children receives only national conditional conditional cash micronutrient sprinkles, given daily. Another cash transfer program transfer program subset receives both treatments. Intervention lasts Famílias en Acción (poorest 20 percent 18 months. (Families in Action) of households) (Colombia) Perry Preschool (United Reduce poverty 3–4 Low-income African- Two and a half hours a day of center-based preschool 1962 Heckman and others States) American children education by certified teachers five days a week for (2010), Heckman, (initial IQs below 85 two years. Children are taught social skills in daily Pinto, and Savelyev at age 3) sequence in which they plan, execute, and review a (2013) task with teachers and fellow students. Children learn to work with others when problems arise. Average child-teacher ratio is 6:1. Teachers also provide weekly 1.5-hour home visits to promote parent-child relationships, involve parents in educational process, and help implement preschool curriculum at home. table continues next page Table F.3  Description of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) (continued) Program Purpose Age target Target population Description Start year References Tools of the Mind (Chile, Improving 3–4 Children from Teaching and classroom strategies to help children 1993 Bodrova and Leong United States) academic low-income families regulate their behaviors. Program considers (1996) learning (78 percent of self-regulation (such as paying attention and families have annual remembering) integral to quality and quantity of income of less than academic learning. In this view, academic content $25,000) (such as literacy and numeracy) is a means for practice, not the sole goal of learning; educational goals are more broadly defined as child achievement, engagement, and social competence. Program elements consist of preschool and kindergarten curricula encouraging children to role-play and learn in groups, a teaching approach, and a professional development program for teachers. Head Start (United States) Reduce poverty 3–5 Children from Preschool education combined with medical, dental, 1965 Carneiro and Ginja low-income families and mental health care; nutrition services; and (2014) (below the federal efforts to help parents foster their children’s poverty level) and development. Services have changed substantially children with since program’s introduction. Implementation disabilities differs greatly across sites. Program is mostly center based but also exists in home and family childcare settings (or combination of these). 103 104 Table F.4 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) Average Age Number of annual cost Program Purpose target Effects participants per participant Returns References Carolina Abecedarian Project Reduce poverty 0–5 Program yielded lasting improvements in IQ, 111 $13,900 (2002 Benefit-cost Barnett and Masse (United States) especially in girls. It increased educational dollars) ratio of (2007), Temple attainment, reduced participation in 3.78: 1 and Reynolds criminal activity, decreased substance (2007), Campbell abuse, and improved anxiety, depression, and others and aggressive behavior in women who (2014) participated as girls. Among men in their mid-30s who had participated as boys, it improved employment and health outcomes, significantly improved socioemotional skills, and reduced the prevalence of risk factors for cardiovascular and metabolic disease. Supplementation Study Improve health 1–2 Psycho-simulation component increased 129 — — Gertler and others (Jamaica) earnings by 25 percent after 20 years (2014) (enough for beneficiary to catch up to earnings of a nonstunted comparison group); improved skill development; increased educational attainment; and reduced criminal activity. Nutritional supplementation had no long-term effects. Early childhood Improve health 1–2 Psychosocial stimulation had significant 1,420 $500 (2011 — Attanasio, Cattan, development component impacts on cognition and receptive dollars), which and others of the national conditional language. Nutritional supplementation had could be (2015) cash transfer program no effects, and there were no interactions reduced at Famílias en Acción between stimulation and supplementation. scale (Families in Action) (Colombia) table continues next page Table F.4  Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in the Early Years (Children 0–5) (continued) Average Age Number of annual cost Program Purpose target Effects participants per participant Returns References Perry Preschool (United Reduce poverty 3–4 Program had large and lasting effects on 123 $11,300 (2007 Benefit-cost Barnett and Masse States) education, employment, earnings, dollars) ratio of (2007), Temple marriage, health, healthy behaviors, and 7.16:1 and Reynolds criminal activity by age 27. Rate of return of (2007), Heckman 7–10 percent a year for both boys and girls and others is attributed to the increase in (2010), socioemotional skills, as there was no Heckman, Pinto, lasting improvement in IQ. and Savelyev (2013) Tools of the Mind (Chile, Improving 3–4 Program improved classroom quality and — About $3,000 per — Bodrova and Leong United States) academic children’s cognitive control and reduced classroom (1996) learning conduct problems. Results on academic (2008 dollars; achievements and language were mixed. estimate) depending on program’s size Head Start (United States) Reduce poverty 3–5 Program reduces incidence of behavioral 1 million $7,800 (2009 Internal rate Carneiro and Ginja problems, health problems, and obesity in (2009) dollars) of return (2014) boys at ages 12 and 13; decreases of at depression and obesity among adolescents; least 4 and reduces engagement in criminal percent activities and idleness among young adults. It raises IQ and achievement test scores, but improvements fade by age 10. Note: Estimated average costs and benefits may not be comparable across countries because different components may have been included in the computations. See references for details on methodology. — = Not available. 105 106 Table F.5 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) Location Components Age Parental Guidance, Parental Teacher Performance Health Program Purpose target School Center involvement Mentoring counselor training training activities services Skills targeted Incredible Years (United Prevent violence 3–7 ♦ ♦ ♦ ♦ Social competences, States, Western and improve emotional self- Europe, elsewhere) student regulation, persistence, learning interpersonal problem solving, anger management, communication, pro-social behavior Project STAR (Student- Enhance quality 5–6 ♦ ♦ ♦ Student effort, initiative, Teacher of education self-value in classroom, Achievement Ratio) pro-social behaviors, IQ (United States) Seattle Social Prevent crime 6–7 ♦ ♦ ♦ ♦ ♦ Communication, decision Development making, negotiation, Project (United and conflict resolution States) Montreal Longitudinal Prevent crime 7–9 ♦ ♦ ♦ Social and behavioral Experimental Study skills; positive (Canada) interactions with teachers, parents, and peers; problem solving; self-regulation table continues next page Table F.5  Objectives and Components of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) (continued) Location Components Age Parental Guidance, Parental Teacher Performance Health Program Purpose target School Center involvement Mentoring counselor training training activities services Skills targeted Escuela Nueva (New Enhance quality 6–11 ♦ ♦ Teamwork and School) (Colombia, of education cooperative learning, Vietnam, elsewhere) decision making, self-management, creativity and innovative thinking, leadership, communication, self-confidence Fútbol con Corazón Develop social 5–17 ♦ ♦ ♦ ♦ ♦ ♦ Leadership, social (Football with Heart) skills and problem-solving, (Colombia) prevent resilience, self-control, violence teamwork, initiative, confidence, ethics Escuela Amiga (Friendly Prevent violence 6–17 ♦ ♦ ♦ Self-awareness, self- School) (Peru) and improve regulation, resilience, student social awareness and learning connectedness, social decision making Sources: Kautz and others 2014; OECD 2015, and authors’ elaboration. 107 Table F.6  Description of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) 108 Age Program Purpose target Target population Description Start year References Incredible Years Prevent violence 3–7 Clinic-referred School curriculum offering 20- to 30-minute lessons on 1982 Webster-Stratton, (United States, and improve children socioemotional skills twice a week for 15 weeks. Lessons focus Reid, and Western Europe, student diagnosed with on recognizing and understanding feelings, getting along with Stoolmiller (2008) elsewhere) learning oppositional friends, regulating emotions, solving problems, and behaving defiant disorder well at school. They are reinforced by practicing skills in or early-onset 20-minute sessions after completing the structured curriculum conduct and continuing to practice the skills throughout the school day problems and at home. Program is supplemented by parent training that focuses on positive discipline and engaging in children’s school lives via family homework. Project STAR Enhance quality 5–6 Children from From kindergarten through third grade, children and teachers 1985 Krueger (1999), (Student-Teacher of education low-income were randomly assigned to classrooms of differing class sizes of Chetty and others Achievement schools 13–17 or 20–25 students to study the impact of greater quality (2011) Ratio) (United kindergarten in smaller classes on short-term grades and States) longer-term outcomes. Seattle Social Prevent crime 6–7 Public elementary Training of teachers and parents to promote parent-child 1981 Hawkins and others Development schools in attachment and teacher-child interactions. Elementary school (2008) Project (United high-crime areas teachers received five days of training a year that included States) proactive classroom management, interactive teaching, and cooperative learning. First-grade teachers also received extra training on curriculum teaching children to resolve conflicts with peers. Parents were offered a seven-session course on behavioral management training (offered to parents of first and second graders), a four-session course to support their children’s academic achievement (offered to parents of second and third graders), and a five-session course to reduce drug use (offered to parents of fifth and sixth graders). Montreal Prevent crime 7–9 Most disruptive Two-year intensive training program for boys, parents, and teachers 1974 Algan and others Longitudinal boys in schools taught social skills (for example, saying hello, doing things (2014) Experimental located in areas together, saying no politely) and self-control (for example, how Study (Canada) of low to wait when impatient, follow rules, react to teasing) in 19 socioeconomic sessions of small groups of children (one disruptive boy for every status three nondisruptive ones). Family visits aim to teach parents to reinforce social skills and self-control (monitoring child, managing punishment and conflicts, and so forth). table continues next page Table F.6  Description of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) (continued) Age Program Purpose target Target population Description Start year References Escuela Nueva (New Enhance quality 6–11 Rural children Teaching and learning practices in classroom that promote active 1985 in Psacharopoulos, School) of education (Colombia) learning. Students guide themselves through textbooks and Colombia; Rojas, and Velez (Colombia, work in groups while interacting with one another. Teachers are 2012 in (1993), Vietnam, trained to promote active learning and use it themselves. Vietnam Schiefelbein (1993) elsewhere) Flexible promotion and individualized instruction allow students to advance at their own pace (instruction is multigrade). Students learn democratic behavior by participating in student government. Mastery learning (peer instruction) is practiced as older students tutor younger students. Fútbol con Corazón Develop social 5–17 Children from Out-of-school program uses soccer to attract participants and 2007 Deporte y Desarollo (Football with skills and disadvantaged teach them socioemotional skills. Rules of game are changed in (2009) Heart) (Colombia) prevent neighborhoods order to teach a range of socioemotional skills, including social violence who keep on problem solving, resilience, self-control, teamwork, initiative, formal schooling confidence, and ethics. Children play in gender-mixed teams and are supported by mentors, who are trained to reinforce the socioemotional objectives of the program. Intervention also includes a nutrition program, health and life skills workshops, links to vocational training, parental engagement, counseling, and community outreach. Escuela Amiga Prevent violence 6–17 Pupils from poor Year-long training session, taught by university psychology staff, 2013 Gertler and Kudo (Friendly School) and improve neighborhoods for midcareer teachers and school principals; develops their (2015) (Peru) student of Lima socioemotional skills so that they can apply them in the learning classroom. Participants can attend 2-hour daily sessions during the week while actively working the schools or meet for 10 hours every Saturday (384 hours of class time in 2 semesters). Course is delivered through lectures, paper exercises, role- playing, and group interactions that focus on recognizing and managing a range of skills. Course incorporates real-time issues into curricula, drawn from participants’ work lives. After course ends, program’s roving support teams regularly visit schools to continue supporting teachers and principals in applying course tools. 109 110 Table F.7 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) Age Number of Average annual Program Purpose target Effects participants cost per participant Returns References Incredible Years Prevent violence 3–7 After six months, participants were — $1,200–$3,000, — Webster-Stratton, (United States, and improve significantly more likely to display depending on Reid, and Western Europe, student learning emotional self-regulation and social project Stoolmiller elsewhere) competence; they also had fewer components (2008) conduct problems and engaged in less off-task behavior (disengagement). Effect was particularly strong among students in classrooms with lowest initial scores on these skills. Project STAR Enhance quality of 5–6 Children placed in improved kindergarten 11,571 $9,355 (2009 dollars; $1,520 (9.6 percent) Krueger (1999), (Student-Teacher education classrooms had significantly higher cost of 33 at age 27 for class Chetty and Achievement earnings in early adulthood and were percent- quality, $368 (2.3 others (2011) Ratio) (United more likely to attend college, save more reduction of class percent) at age States) for retirement, and live in better size for 2 years) 27 for class size neighborhoods. Improvements in test scores faded out but gains in socioemotional measures persisted. Seattle Social Prevent crime 6–7 Participants had significantly better 808 $4,355 Benefit-cost ratio of Hawkins and others Development educational and economic attainment, 4.25: 1 (2008) Project (United mental health, and sexual health by age States) 27. Hypothesized effects on substance use and crime were not found at ages 24 or 27. Montreal Prevent crime 7–9 Program increased full-time employment 250 $6,500 (2011 $14 per dollar Algan and others Longitudinal or school enrolment at ages 17–26 by dollars) invested (2014) Experimental 11 percentage points, and secondary Study (Canada) school graduation by ages 23–24 by 19 percentage points. It reduced the probability of having a criminal record by ages 23–24 by 11 percentage points. The program also boosted measures of non-cognitive skills and grades during adolescence. table continues next page Table F.7  Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Middle Childhood (6–11) (continued) Age Number of Average annual Program Purpose target Effects participants cost per participant Returns References Escuela Nueva (New Enhance quality of 6–11 Program improved academic 551,749 in — — Psacharopoulos, School) education achievement and behavior. Vietnam Rojas, and Velez (Colombia, (1993), Vietnam, Schiefelbein elsewhere) (1993), Colbert (2009) Fútbol con Corazón Develop social 5–17 — 300 children — — Deporte y Desarollo (Football with skills and per soccer (2009) Heart) prevent violence field (Colombia) Escuela Amiga Prevent violence 6–17 Program helped build teachers’ and Nearly 15,000 Teacher training — Gertler and Kudo (Friendly School) and improve principals’ knowledge of classroom costs about (2015) (Peru) student learning socioemotional skills and how to use teachers $4,000 per them in a school setting. Trained school and 81 teacher (2013 professionals felt more able to manage principals dollars); roaming the classroom but also better learned teams costs about themselves and how to manage about $20,000 their own children, marriages, and (2013 dollars) per neighbors: 90 percent of teachers and school school principals felt that they were better able to manage their classrooms, 93 percent believed that they were better equipped to manage conflict in the school setting, and 50 percent felt that their professional relationships had improved. Note: Estimated average costs and benefits may not be comparable across countries because different components may have been included in the computations. See references for details on methodology. — = Not available. 111 112 Table F.8 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) Location Components Age Parental Guidance, Teacher Sports/art Health Program Purpose target School involvement Mentoring counselor training activities services Skills targeted Knowledge Is Power Increase college 5–19 ♦ ♦ ♦ ♦ ♦ Zest, grit, optimism, self- Program (KIPP) graduation management, gratitude, (United States) rate social intelligence, curiosity Big Brothers Big Sisters Reduce poverty 10–16 ♦ ♦ ♦ Self-confidence, motivation, (United States) social acceptance and behavior, prevention of antisocial behavior, academic skills Empresários Pela Reduce dropout 13–15 ♦ ♦ ♦ ♦ Motivation, self-control, Inclusão Social (EPIS) rate problem-solving, social skills (Entrepreneurs for Social Inclusion) (Portugal) Becoming a Man Reduce dropout 15–16 ♦ ♦ ♦ ♦ Social-cognitive skills: impulse (United States) rate and control, emotional prevent self-regulation, conflict violence resolution, raising aspirations for the future, sense of personal responsibility Chicago One-Goal Increase college 15–18 ♦ ♦ ♦ ♦ Mindsets, perseverance, (United States) graduation self-advocacy, academic skills rate Construye T (Build Reduce dropout 15–18 ♦ ♦ Self-awareness, self-regulation, Yourself) (Mexico) rate and resilience, social awareness enhance and connectedness, social quality of decision making education National Guard ChalleNGe Reduce dropout 16–18 ♦ ♦ Confidence and responsibility, (United States) rate feeling of self-control, sense of leadership and potential, academic skills Sources: Kautz and others 2014, OECD 2015, and authors’ elaboration. Table F.9  Description of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) Program Purpose Age target Target population Description Start year References Knowledge Is Power Increase college 5–19 Low-income students Recruitment of graduates of elite colleges to teach in 1994 Angrist and others Program (KIPP) graduation who qualify for free low-performing school districts. Program uses a (2010), Tuttle and (United States) rate or reduced-price “character” report card, in which teachers assess students’ others (2013) lunch (95 percent success in demonstrating a range of socioemotional skills. are black or Latino) Teacher’s feedback is aggregated into a score for each skill and reported in the card. During student performance reviews with students and parents, character report card is discussed in same way as standard report card. Big Brothers Big Sisters Reduce poverty 10–16 Children living in Volunteer mentors who meet regularly in lengthy one-on- 1977 Tierney, Baldwin- (United States) unstable family one meetings with mentees for one year, on average. Grossman, and environments, Program allows mentees and mentors to form strong Resch (1995) which generally are attachments and may help avoid negative peer effects of single-parent grouping at-risk youth together. households Empresários Pela Reduce dropout 13–15 Pupils with poor Large private sector program consisting of one-on-one 2007 Martins (2010) Inclusão Social (EPIS) rate academic results after-school meetings with trained staff member or (Entrepreneurs for and most at risk to meetings in small groups. Sessions aim at improving the Social Inclusion) drop out socioemotional skills (for example, study skills, motivation, (Portugal) self-esteem) of the worst-performing students. Becoming a Man Reduce dropout 15–16 At-risk boys (poor Mentorship programs focusing on teaching socioemotional 2009 Cook and others (United States) rate and academic results, skills (through 27 weekly hour-long cognitive behavioral (2014), Heller prevent dropouts, or training sessions) and academic skills (through daily and others violence engaged in crime) hour-long tutoring sessions in math). Intervention is (2015) delivered in groups to help control costs. Groups kept small to help develop relationships (maximum of 15 boys, with average youth-to-adult ratio of 8:1. Students skip an academic class in order to participate in program, one of the draws to attend. Program is documented by manuals and can be delivered by college-educated mentors without specialized training in psychology or social work, although there is a preference for such training in selecting program providers. Program seeks mentors with ability to keep youth engaged. After-school sports component reinforces conflict-resolution skills and social and emotional learning objectives of in-school curriculum. 113 table continues next page 114 Table F.9  Description of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) (continued) Program Purpose Age target Target population Description Start year References Chicago One-Goal Increase college 15–18 Motivated, High school teachers who help students improve their 2003 Kautz and Zanoni (United States) graduation disadvantaged high grades and test scores; apply to colleges (helping them fill (2014) rate school students out financial aid forms, write essays, and make college (selected through choices); and remain in college. Daily classes start in 11th applications and grade and continue for two years. Specific socioemotional interviews) from skills are taught in contexts in which they can be readily Chicago’s applied. Mentoring relationship continues throughout low-income schools first year of college. in Chicago, most of which have college enrolment rates of less than 50 percent Construye T (Build Reduce dropout 15–18 High school staff and Training of teachers and school principals and use of toolkits 2013 SEMS (2014) Yourself) (Mexico) rate and students of activities to develop socioemotional skills and improve enhance school climate. quality of education National Guard Reduce dropout 16–18 High school dropouts Seventeen-month intervention occurs in a residential facility 1993 Millenky and others ChalleNGe (United rate removed from the usual environments of participants. (2012), States) Program features a 2-week residential orientation and Perez-Arce and assessment period; a 20-week residential program, often others (2012) conducted at a military base; and a 1-year nonresidential mentoring program. Table F.10 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) Age Number of Average annual cost Program Purpose target Effects participants per participant Returns References Knowledge Is Power Increase college 5–19 Program improved academic 27,000 (2010) $4,200–$17,200 (2014 — Angrist and others Program (KIPP) graduation achievement, especially dollars), (2010), Tuttle schools (United rate among students with limited depending on the and others States) English proficiency and special model (2013) education needs. Big Brothers Big Sisters Reduce poverty 10–16 Program has positive impacts on 126,000 — — Tierney, Baldwin- (United States) academic outcomes for girls Grossman, and but not boys. Studies on Resch (1995) long-term effects are needed. Empresários Pela Reduce dropout 13–15 Program reduced grade retention 84 schools and €1,000– €3,250 (2010) — Martins (2010) Inclusão Social (EPIS) rate by 10 percentage points. more than (Entrepreneurs for 15,000 students Social Inclusion) (10 percent of (Portugal) the country) Becoming a Man (United Reduce dropout 15–16 Program improves math skills 2,740 $1,100 Benefit-cost ratios Cook and others States) rate and and reduces arrests for violent on the order of (2014), Heller prevent crime by 44 percent in the 3:1 to 31:1 just and others violence short term. from reductions (2015) in crime during program year Chicago One-Goal Increase college 15–18 Program improved academic 4,000 by 2017 $1,351 in 2011 — Kautz and Zanoni (United States) graduation results in high school; ($5,092 in 2007) (2014) rate increased high school graduation rates; and raised college enrollment by 10–20 percentage points, with about 15–30 percent of the increase related to improvements in socioemotional skills table continues next page 115 116 Table F.10  Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Adolescence (12–18) (continued) Age Number of Average annual cost Program Purpose target Effects participants per participant Returns References Construye T (Build Reduce dropout 15–18 — 2,500 schools in — — SEMS (2014) Yourself) (Mexico) rate and 32 states (2 enhance million students quality of and 99,000 education teachers) National Guard Reduce dropout 16–18 Program increased likelihood of 113,000 (total, $11,633 (2010 dollars) $2.66 per dollar Millenky and ChalleNGe (United rate earning a General Education 2015) invested (return others (2012), States) Degree (GED) or high school on investment of Perez-Arce and diploma and being employed 166 percent; others (2012) and reduced likelihood of internal rate of being arrested or convicted. return 6.4 Three years after intervention, percent) effects for criminal behavior and high school graduation declined and became statistically insignificant, however. Initial reduction in crime likely occurs because participants are housed in a residential program (incapacitation effect). Note: Estimated average costs and benefits may not be comparable across countries because different components may have been included in the computations. See references for details on methodology. Table F.11 Objectives and Components of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) Location Components Age Guidance, Sports/art Cash/ Work Health Program Purpose target School Center Work Mentoring counselor activities voucher training services Skills targeted Job Corps Reduce poverty 16–24 ♦ ♦ ♦ ♦ Interpersonal communication, (United States) problem solving, social and management skills, technical skills, academic skills Youthbuild (United Prevent violence 16–24 ♦ ♦ ♦ ♦ ♦ Teamwork, communication, States, Central and improve initiative, confidence, self- America, South student esteem Africa, elsewhere) learning Year-up Increase 18–24 ♦ ♦ ♦ ♦ Time management, teamwork, (United States) employability problem solving, conflict resolution Jóvenes en Acción Increase 18–25 ♦ ♦ ♦ ♦ ♦ Initiative, teamwork, creativity, (Youth in Action) employability openness to feedback, (Colombia) communication (verbal and written), problem-solving, decision making, growth mindset, goal setting Juventud y Empleo Increase 16–28 ♦ ♦ ♦ ♦ ♦ Self-esteem, problem solving, (Youth and employability decision making, conflict Employment) resolution, empathy, (Dominican cooperation, responsibility, Republic) emotional control, reduced risk behavior, communication, creative thinking Galpão Aplauso Increase 15–29 ♦ ♦ ♦ ♦ Cooperation, trustworthiness, (Applause employability leadership, literacy, numeracy, Warehouse) technical skills (Brazil) Sustainable Prevent violence 18–35 ♦ ♦ ♦ Self-control, self-esteem, regulation Transformation of and criminality of antisocial behavior, and Youth (Liberia) growth mindset 117 Sources: Kautz and others 2014; OECD 2015, and authors’ elaboration. 118 Table F.12  Description of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) Age Start Program Purpose target Target population Description year References Job Corps Reduce poverty 16–24 Poor high school dropouts and One-year center-based academic, vocational, and social 1964 Schochet, Burghardt, (United States) other poor youth residing in skills training, including counseling. Program provides and McConnell disruptive environments health services and a stipend during enrollment. Largest (2008), Flores and who need training or residential training program in the United States for others (2012) education; participants must at-risk youth. be citizens or permanent residents and not on parole Youthbuild (United Prevent violence 16–24 Low-income criminal offenders Training in construction and other skills involved in 1978 IYF, Youthbuild States, Central and improve community improvement projects while indirectly International, and America, South student emphasizing improvements in socioemotional skills. Catholic Relief Africa, elsewhere) learning Services (2010), Cohen and Piquero (2015) Year-up Increase 18–24 Low-income young adults In-classroom training of technical and professional skills 2001 Grobe, Rosenblum, (United States) employability (first six months) and internship at one of program’s and Weissman corporate partners (second six months). Participants earn (2010), Roder and college credits and a weekly stipend. They are supported Elliot (2014) by staff advisors, professional mentors, social services staff, and a network of community-based partners. Jóvenes en Acción Increase 18–25 Unemployed urban youth in Three-month classroom training followed by three months 2002 Attanasio, Kugler, (Youth in Action) employability bottom fifth of income of on-the-job training. Classroom training, provided by and Meghir (Colombia) distribution private institutions, focuses on providing skills (2011), Attanasio, demanded in the local labor market. On-the-job training Guarín and others is provided through unpaid internships with companies (2015), Kugler and that specialize in manufacturing, retail sales and trade, or others (2015) services (5.2 hours of work a day on average). Participants are also required to develop a life project. Throughout the six months of the program, they receive stipend of $2.20 a day to cover transportation and food costs. Women with children under seven receive an additional $0.80 a day to help cover the costs of childcare. table continues next page Table F.12  Description of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) (continued) Age Start Program Purpose target Target population Description year References Juventud y Empleo Increase 16–28 Youth from households earning In-classroom training, internship, and benefits (stipend and 2001 Ibarrarán and others (Youth and employability less than $120 a month who insurance). Vocational skills module included 150 hours of (2014), Ibarrarán Employment) experience difficulty finding training (in sectors such as sales, tourism and hospitality, and others (2015) (Dominican jobs and have not and carpentry). Life skills component included module of Republic) completed secondary 75 hours that covered self-esteem and self-realization, education communication, conflict-resolution, life planning, time management, team work, decision making, hygiene and health, and coaching on risky behaviors. Once in-classroom training phase was completed, all participants were assigned to 240-hour apprenticeships or internships at private companies, for which they received a daily stipend of about $2 and basic insurance. During this period, participants received oversight and job counseling. Galpão Aplauso Increase 15–29 At-risk youth from low-income Combination of vocational, academic, and life skills training, 2005 Calero and others (Applause employability households (earning less delivered through a pedagogic method that uses arts (2014) Warehouse) (Brazil) than two minimum wages) and dance, and job placement services. Program lasts living in slums about six months, five hours a day, five days a week, delivered in three shifts—morning, afternoon, and evening. Includes 300 hours of vocational training (mainly construction related, soldering, or wood shop); 180 hours of training on academic and basic skills, including remedial courses in both math and Portuguese; and 120 hours of training on life skills. Sustainable Prevent violence 18–35 High-risk men (for example, Eight-week course of group cognitive behavior therapy 2009 Blattman, Jamison, Transformation of and criminality men engaged in petty crime (CBT) focused on developing self-control, such as the and Sheridan Youth (Liberia) and drug dealing, homeless tendency to be planful, responsible, and resistant to (2015) men, and poorly temptation. Therapy sought to foster nonviolent, reintegrated excombatants) noncriminal self-image and set of values. Sessions from five mixed-income employed a variety of techniques, from lectures and areas of Monrovia group discussions to various forms of practice, including role-playing in class, homework that required practicing tasks, exposure to real-world situations, and in-class processing of experiences of executing tasks. As in many CBT programs, tasks began simply and became more 119 difficult over time. In a second treatment arm, half of participants received a $200 cash transfer. 120 Table F.13 Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) Age Number of Average cost per Program Purpose target Effects participants participant Returns References Job Corps Reduce poverty 16–24 Program increases earnings, but 60,000 a year $26,551 (2009 $22.10 in weekly Schochet, Burghardt, (United States) marginal value of longer exposure dollars) earnings after and McConnell decreases. 48 months in (2008), Flores and average with others (2012) heterogeneity across race: $46.20 for whites, $22.80 blacks, and $15.10 for Hispanics Youthbuild Prevent violence 16–24 Program increased self-esteem and — $13,000–$24,000 Benefit-cost ratio: IYF, Youthbuild (United States, and improve life skills and decreased (the lower 13:1 to 22:1 International, and Central America, student delinquency. estimate based on social Catholic Relief South Africa, learning excludes trainee costs; 7: 1 to 12:1 Services (2010), elsewhere) stipends and cost based on Cohen and of building program costs; Piquero (2015) materials) return on investment: $7.20–$21.60 Year-up (United Increase 18–24 Program increased employment and 10,000 (2014) $24,562, of which — Grobe, Rosenblum States) employability earnings. one quarter is and Weissman weekly stipend (2010), Roder and ($150–$250) Elliot (2014) Jóvenes en Acción Increase 18–25 Ten years after intervention, trainees, 80,000 in seven $750 (plus Net gains in Attanasio, Kugler, (Youth in employability especially women, were largest cities opportunity cost women’s future and Meghir Action) significantly more likely to be over 2002–06 of $62 from income of (2011), Attanasio, (Colombia) employed in formal sector and being out of $666–$2,993 Guarín, and others have higher formal earnings (no labor force (2015), Kugler and data on informal sector). during program) others (2015) table continues next page Table F.13  Effects, Costs, and Benefits of Promising Interventions Fostering Socioemotional Skills in Young Adulthood (19–29) (continued) Age Number of Average cost per Program Purpose target Effects participants participant Returns References Juventud y Empleo Increase 16–28 Short-term employment and wage 27,500 over $400 — Ibarrarán and others (Youth and employability gains for women were large for 2002–08; (2014), Ibarrarán Employment) men; effects dissipate in longer 38,000 since and others (2015) (Dominican run. Effects on risky behavior were 2013 Republic) mixed: Pregnancy declined, but smoking increased. Effect on motivation and expectations was strong. Galpão Aplauso Increase 15–29 Five months after program ended, Peak of 10,000 in $385 a month, or — Calero and others (Applause employability impacts were very large: 33 early years; four $2,225 for entire (2014) Warehouse) percent increase in probability of to five cohorts curriculum (2012 (Brazil) being employed and 24 percent of 100 youth a dollars) increase in earnings. No impacts year in 2009–13 found for shorter periods nor changes on personality traits. Sustainable Prevent violence 18–35 A year after completing program, 999 $530 ($14 for — Blattman, Jamison, Transformation and criminality self-control and noncriminal registration, $189 and Sheridan of Youth values improved, leading to large, for therapy, $216 (2015) (Liberia) sustained declines in crime and for grant, and violence. Impacts were greatest $111 for when followed by cash, likely administration because it reinforced behavioral costs) changes through prolonged practice. Intervention had no effect on employment or earnings. Note: Estimated average costs and benefits may not be comparable across countries because different components may have been included in the computations. See references for details on methodology. 121 122 Inventory of Promising Interventions to Foster Socioemotional Skills References Algan, Y., E. Beasley, F. Vitaro, and R. E. Tremblay. 2014. “The Impact of Non-Cognitive Skills Training on Academic and Non-academic Trajectories: From Childhood to Early Adulthood.” Manuscript. Sciences Po, Paris. Angrist, J., S. Dynarski, T. Kane, P. Pathak, and C. Walters. 2010. “Who Benefits from KIPP?” NBER Working Paper 15740, National Bureau of Economic Research, Cambridge, MA. Attanasio, O., S. Cattan, E. Fitzsimons, C. Meghir, and M. Rubio-Codina. 2015. “Estimating the Production Function for Human Capital: Results from a Randomized Control Trial in Colombia.” NBER Working Paper 20965, National Bureau of Economic Research, Cambridge, MA. Attanasio, O., A. Guarín, C. Medina, and C. Meghir. 2015. “Long Term Impacts of Vouchers for Vocational Training: Experimental Evidence for Colombia.” NBER Working Paper 21390, National Bureau of Economic Research, Cambridge, MA. Attanasio, O., A. Kugler, and C. Meghir. 2011. “Subsidizing Vocational Training for Disadvantaged Youth in Colombia: Evidence from a Randomized Trial.” American Economic Journal: Applied Economics 3 (3): 188–220. Barnett, W. S., and L. N. Masse. 2007. “Comparative Benefit-Cost Analysis of the Abecedarian Program and Its Policy Implications.” Economics of Education Review 26 (1): 113–125. Blattman, C., J. C. Jamison, and M. Sheridan. 2015. “Reducing Crime and Violence: Experimental Evidence on Adult Noncognitive Investments in Liberia.” NBER Working Paper 21204, National Bureau of Economic Research, Cambridge, MA. Bodrova, E., and D. Leong. 1996. Tools of The Mind: The Vygotskian Approach to Early Childhood Education. Geneva, Switzerland. Calero, C., C. H. Corseuil, V. Gonzales, J. Kluve, and Y. Soares. 2014. “Can Arts-Based Interventions Enhance Labor Market Outcomes among Youth? Evidence from a Randomized Trial in Rio de Janeiro.” IZA Discussion Paper 8210, Institute for the Study of Labor, Bonn. Campbell, F., G. Conti, J. J. Heckman, S. H. Moon, R. Pinto, L. Pungello, and Y. Pan. 2014. “Early Childhood Investments Substantially Boost Adult Health.” Science 343: 1478–85. Carneiro, P., and R. Ginja. 2014. “Long-Term Impacts of Compensatory Preschool on Health and Behavior: Evidence from Head Start.” American Economic Journal: Economic Policy 6 (4): 135–73. Chetty, R., J. N. Friedman, N. Hilger, E. Saez, D. Whitmore Schanzenbach, and D. Yagan. 2011. “How Does Your Kindergarten Classroom Affect Your Earnings? Evidence from Project Star.” Quarterly Journal of Economics 126 (4): 1593–660. Cohen, M. A., and A. R. Piquero. 2015. Benefits and Costs of a Targeted Intervention Program for Youthful Offenders: The YouthBuild USA Offender Project. YouthBuild USA. Somerville, MA. Colbert, V. 2009. “Improving Education Quality and Access in Colombia through Innovation and Participation: The Escuela Nueva Model.” Journal of Education for International Development 3 (3): 1–8. Cook, P. J., K. Dodge, G. Farkas, R. G. Fryer Jr., J. Guryan, J. Ludwig, S. Mayer, H. Pollack, and L. Steinberg. 2014. “The (Surprising) Efficacy of Academic and Behavioral Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Inventory of Promising Interventions to Foster Socioemotional Skills 123 Intervention with Disadvantaged Youth: Results from a Randomized Experiment in Chicago.” NBER Working Paper 19862, National Bureau of Economic Research, Cambridge, MA. Deporte y Desarollo. 2009. Fútbol con corazón: Goles de paz y convivencia [Soccer with Heart: Goals of Peace and Conviviality]. Deporte y Desarollo. Bogota, Colombia. Flores, C., A. Flores-Lagunes, A. Gonzalez, and T. Neumann. 2012. “Estimating the Effects of Length of Exposure to Instruction in a Training Program: The Case of Job Corps.” Review of Economics and Statistics 94 (5): 153–71. Gertler, P., J. J. Heckman, R. Pinto, A. Zanolini, C. Vermeerch, S. Walker, S. Chang, and S. Grantham-McGregor. 2014. “Labor Market Returns to an Early Childhood Stimulation Intervention in Jamaica.” Science 344: 998–1001. Gertler, P., and I. Kudo. 2015. “Socio-Emotional Learning for At-Risk Students in Urban Schools in Lima.” Manuscript. World Bank and Innovation for Poverty Action, Lima. Grobe, T., E. Rosenblum, and T. Weissman. 2010. Dollars and Sense: How ‘Career First’ Programs Like Year Up Benefit Youth and Employers. Jobs for the Future, Boston. Hawkins, J. D., R. Kosterman, R. F. Catalano, K. G. Hill, and R. D. Abbott. 2008. “Effect of Social Development Intervention in Childhood 15 Years Later.” Archives of Pediatrics and Adolescent Medicine 162 (12): 1133–41. Heckman, J. J., S. H. Moon, R. Pinto, P. Savelyev, and A. Yavitz. 2010. “Analyzing Social Experiments as Implemented: A Reexamination of the Evidence from the High Scope Perry Preschool Program.” Journal of Quantitative Economics 1 (1): 1–46. Heckman, J. J., R. Pinto, and P. A. Savelyev. 2013. “Understanding the Mechanisms through Which an Influential Early Childhood Program Boosted Adult Outcomes.” American Economic Review 103 (6): 1–35. Heller, S. B., A. K. Shah, J. Guryan, J. Ludwig, S. Mullainathan, and H. A. Pollack. 2015. “Thinking, Fast and Slow? Some Field Experiments to Reduce Crime and Dropout in Chicago.” NBER Working Paper 21178, National Bureau of Economic Research, Cambridge, MA. Ibarrarán, P., J. Kluve, L. Ripani, and D. Rosas Schady. 2015. “Experimental Evidence on the Long Term Impacts of a Youth Training Program.” IZA Discussion Paper 9136, Institute for the Study of Labor, Bonn. Ibarrarán, P., L. Ripani, B. Taboada, J. Villa, and B. Garcia. 2014. “Life Skills, Employability and Training for Disadvantaged Youth: Evidence from a Randomized Evaluation Design.” IZA Journal of Labor and Development 3 (10). IYF (International Youth Foundation), Youthbuild International, and Catholic Relief Services. 2010. Executive Summary: International Youth Foundation Follow-Up Study “Central American Youth Builders Program.” Baltimore, MD: IYF, Youthbuild International, and Catholic Relief Services. Kautz, T., J. J. Heckman, R. Diris, B. T. Weel, and L. Borghans. 2014. “Fostering and Measuring Skills: Improving Cognitive and Non-Cognitive Skills to Promote Lifetime Success.” OECD Education Working Paper 110, Organisation for Economic Co-operation and Development, Paris. Kautz, T., and W. Zanoni.2014. “Measuring and Fostering Non-Cognitive Skills in Adolescence: Evidence from Chicago Public Schools and the OneGoal Program.” University of Chicago, Chicago. Krueger, A. B. 1999. “Experimental Estimates of Education Production Functions.” Quarterly Journal of Economics 114 (2): 497–532. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 124 Inventory of Promising Interventions to Foster Socioemotional Skills Kugler, A., M. Kugler, J. Saavedra, and L. O. Herrera Prada. 2015. “Long-Term Direct and Spillover Effects of Job Training: Experimental Evidence from Colombia.” NBER Working Paper 21607, National Bureau of Economic Research, Cambridge, MA. Martins, P. S. 2010. “Can Targeted, Non-Cognitive Skills Programs Improve Achievement?” IZA Discussion Paper 5266, Institute for the Study of Labor, Bonn. Millenky, M., D. Bloom, S. Muller-Ravett, and J. Broadus. 2012. Staying on Course: Three- Year Results of the National Guard Youth Challenge Evaluation. Technical Report, MDRC, Washington, DC. OECD (Organisation for Economic Co-operation and Development). 2015. Skills for Social Progress: The Power of Social and Emotional Skills.OECD Skills Studies. Paris: OECD Publishing. Perez-Arce, F., L. Constant, D. S. Loughran, and L. S. Karoly. 2012. A Cost-Benefit Analysis of the National Guard Youth ChalleNGe Program. RAND Corporation, Arlington, VA. Psacharopoulos, G., C. Rojas, and E. Velez. 1993. “Achievement Evaluation of Colombia’s “Escuela Nueva”: Is Multigrade the Answer?” Comparative Education Review 37 (3): 263–76. Roder, A., and M. Elliott. 2014. Sustained Gains: Year Up’s Continued Impact on Young Adults’ Earnings. Technical Report, Economic Mobility Corporation, New York. Schiefelbein, E.1993. In Search of the School of the XXI Century: Is Colombian Escuela Nueva the Right Pathfinder? UNICEF-UNESCO, Paris. Schochet, P. Z., J. Burghardt, and S. McConnell. 2008. “Does Job Corps Work? Impact Findings from the National Job Corps Study.” American Economic Review 98 (5): 1864–86. SEMS (Subsecretaría de Educación Media Superior de México) [Undersecretariat of Middle Education of Mexico]. 2014. Programa Construye T (2014–2018): Fortalecer las capacidades de la escuela para promover el desarrollo integral de los jóvenes [Program Build Yourself (2014–2018): Strengthening the Capabilities of the School to Promote the Comprehensive Development of the Youth]. Report, Undersecretariat of Middle Education of Mexico, Mexico. Temple, J. A., and A. J. Reynolds. 2007. “Benefits and Costs of Investments in Preschool Education: Evidence from the Child-Parent Centers and Related Programs.” Economics of Education Review 26 (1): 126–44. Tierney, J. P., J. Baldwin-Grossman, and N. L. Resch. 1995. Making a Difference: An Impact Study of Big Brothers Big Sisters. Public/Private Ventures, New York. Tuttle, C. C., B. Gill, P. Gleason, V. Knechtel, I. Nichols-Barrer, and A. Resch. 2013. KIPP Middle Schools: Impacts on Achievement and Other Outcomes. Mathematica Policy Research, Washington, DC. Webster-Stratton, C., M. J. Reid, and M. Stoolmiller. 2008. “Preventing Conduct Problems and Improving School Readiness: Evaluation of the Incredible Years Teacher and Child Training Programs in High-Risk Schools.” Journal of Child Psychology and Psychiatry 49 (5): 471–88. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Environmental Benefits Statement The World Bank Group is committed to reducing its environmental footprint. In support of this commitment, the Publishing and Knowledge Division lever- ages electronic publishing options and print-on-demand technology, which is located in regional hubs worldwide. Together, these initiatives enable print runs to be lowered and shipping distances decreased, resulting in reduced paper consumption, chemical use, greenhouse gas emissions, and waste. The Publishing and Knowledge Division follows the recommended standards for paper use set by the Green Press Initiative. The majority of our books are printed on Forest Stewardship Council (FSC)–certified paper, with nearly all containing 50–100 percent recycled content. The recycled fiber in our book paper is either unbleached or bleached using totally chlorine-free (TCF), processed chlorine-free (PCF), or enhanced elemental chlorine-free (EECF) processes. More information about the Bank’s environmental philosophy can be found at http://www.worldbank.org/corporateresponsibility. Minds and Behaviors at Work  •  http://dx.doi.org/10.1596/978-1-4648-0884-5 Latin America has shown impressive growth in educational attainment over the past two decades— but that education has failed to yield the expected benefits. A mounting body of research and policy debates suggests that the quantity of education is not an adequate metric of human capital acquisition. Rather, individuals’ skills—what people actually know and can do—should stand as policy targets and be fostered across the life cycle. Evidence from around the world suggests that employers require both cognitive and socioemotional skills and that both types of skills are associated with a range of positive employment and educational attainment outcomes. Minds and Behaviors at Work: Boosting Socioemotional Skills for Latin America’s Workforce synthesizes original empirical research on the role of cognitive and socioemotional skills in shaping adults’ labor market outcomes in Bolivia, Colombia, El Salvador, and Peru. This work is put in perspective with insights from similar studies in other Latin American countries and high-income countries. The findings show that cognitive skills matter for reaping labor market gains in terms of higher wages and job formality in Latin America but so do socioemotional skills. Moreover, socioemotional skills seem to have a particularly strong effect on labor force participation and tertiary education attendance as a platform to build knowledge. Minds and Behaviors at Work also presents a policy framework for developing skills by providing insights from developmental psychology about when people are neurobiologically, socioemotionally, and situationally ready to develop socioemotional skills and provides examples of interventions that combine socioemotional learning and cognitive development. This book will be of importance to policy makers, researchers, and anyone else interested in human development, from Latin America and beyond. In particular, this book will be most valuable for the curious minds wondering how our mental abilities and behaviors shape our education and employment trajectories, and how to foster these abilities along our lives. ISBN 978-1-4648-0884-5 SKU 210884