88262 The World Bank PREMnotes 2014 NUMBER 30 Special Series on Using National Education Management Information Systems to Make Local Service Improvements: The Case of Pakistan Ritu Nayyar-Stone Education management information systems (EMISs), usually located within the ministry of education, are tools that can help governments improve education system administration by providing information that can be used in strategic planning, resource allocation, and monitoring and evaluation. Frequently, however, they are underutilized and become merely a reporting mechanism. Using the data at the point of collection—usu- ally individual schools in a decentralized environment—and feeding them into service improvement action plans can circumvent problems with the national EMIS, and allow the data to become instrumental in improving local education service delivery outcomes. The Advent of Education Donors and United Nations agencies such as Management Information the World Bank, the United Nations Educational, Systems Scientific, and Cultural Organization (UNESCO), the United Nations Children’s Fund (UNICEF), Health and education expenditures in most coun- and the United States Agency for International De- tries across the world usually account for 15–35 velopment (USAID) also began to focus on EMISs. percent of the national budget. In many countries, For example, from 1998 to 2011, World Bank part of this money is also spent on establishing and education projects with an EMIS component were maintaining a health and education management implemented in 82 countries around the world, information systems (HMIS/EMIS) that collect and in countries as diverse as Afghanistan, Armenia, store important output and outcome data to assist Côte d’Ivoire, Djibouti, Iraq, Kenya, Kosovo, the with sector policy, management, and evaluation. Lao People’s Democratic Republic, Latvia, Leso- In the 1980s and 1990s, several events helped tho, Mozambique, Nigeria, Panama, Vietnam, the highlight the role and potential of EMISs, which Republic of Yemen, and Zambia. Adoption of the then began attracting the attention of governments Millennium Development Goals by the United around the world: the shift in education goals, from Nations in September 20031 motivated countries access to the quality and performance of individual and governments to improve their performance schools; the availability of low-cost computers, easy- and accountability on education service delivery. to-use databases, and statistical analysis software; In early 2011, the Bank also launched the Systems and the decentralization of education provision Approach to Better Education Results (SABER), (Cassidy 2005; Powell and Trucano 2006). which incorporated the EMIS as one of its policy FROM THE POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK domains, focusing on the quality of education data and making. Some districts in Pakistan began following this the degree to which resulting information is used in policy path, but these efforts were undone when the Local Gov- planning and dialogue. ernment Ordinance (LGO) of 2001, which created and In parallel, initial costs of establishing an EMIS have de- gave decision-making autonomy to local governments, clined with the development of systems such as OpenEMIS, lapsed and was discontinued after 2010. Piloted in the which provides generic and open source EMIS.2 Launched Khyber Pakhtunkhwa (KP) province and expanded suc- by UNESCO, it is a tool designed to be quickly and easily cessfully to Punjab and Sindh provinces,3 44 districts and adapted to the needs of information producers and users approximately 921 schools used the EMIS. Pakistan’s at national and subnational levels. It also provides seamless experience provides lessons that can be relevant for other integration with DevInfo, the database system endorsed by countries. the United Nations for tracking country progress toward The next section outlines Pakistan’s NEMIS and its reaching the Millennium Development Goals and other data collection process. A brief description of the local national priorities. government context follows, including local authority The wealth of existing literature on EMISs focuses under decentralization and challenges in the education on its purpose; management and operation; information system. Pakistan’s success in using its EMIS to improve and communication technology (ICT) requirements; data education indicators is then highlighted, including the steps collection analysis and verification; and dissemination of followed by district governments in developing and using its outputs and utilization. The literature also highlights the performance management tool (PMT) for education, lessons and challenges regarding EMISs that are particularly which incorporated EMIS data. The final section discusses relevant for developing countries: the results of the PMT for education, factors for success, • The need to build capacity and generate demand for and the potential for replication. EMIS data so that they are used in policy making. • Poor data quality and limited data verification decrease the demand for data. Pakistan’s Education Management • Provision of timely information is not easy; there are Information System substantial lags between initial data collection activi- Pakistan’s NEMIS is based within the Academy of Educa- ties and the publication of final results. tion Planning and Management (AEPAM), an autonomous • Substantial donor and sustained high-level support/ organization within the federal Ministry of Education. political will are essential. AEPAM produces an annual report that summarizes • A top-down EMIS approach, with national ministries education information for the entire country, based on specifying their information requirements, is not data collected from the four provincial EMISs—each hav- sufficient. ing its own questionnaire and organizational structure, as The above challenges are also relevant to Pakistan, well as data from the Federally Administered Tribal Areas, which established a national EMIS (NEMIS) in 1993 to Islamabad Capital Territory, Azad Jammu and Kashmi- collect, maintain, and disseminate data to support policy rand Gilgit Baltistan. The NEMIS also provides advisory making, planning, and management at each level of gov- services to provincial EMISs, but has no formal authority ernment. There are extensive challenges with Pakistan’s over their management and functioning. District EMIS system. Zaidi (2003) refers to a series of problems with cells are responsible for the collection and transmission of the EMIS at the district (lack of funding, shortage of hu- data to provincial units/wings, but ultimate responsibility man capacity, absence of EMIS vision, and organizational for filling out the questionnaire (often referred to as an issues), provincial (delays in district data submission, data annual school census form) rests with the principal and entry problems, and resource constraints), and federal levels head teacher of each school. (lack of authority, leadership, and financial and human Figure 1 shows the flow of EMIS data from the prin- resources). A more recent assessment of the EMIS at the cipal and teachers of each school to the NEMIS. The data federal, provincial, and district levels in 2008 and 2011 flow is often complicated by the following factors: shows that the challenges identified by Zaidi still exist • Use of existing administrative records from individual (Government of Pakistan 2011). schools, some of which are not regularly updated. This note argues that using EMIS data directly at the • Limited data verification at each level. source, by schools and local government in a decentralized • Spotty or nonuniform consolidation and data entry of setting, can help bypass problems with the national system, all school forms at the district EMIS cell. create the right incentives for improving data quality, and • Irregular verification of district data during consolida- improve education indicators due to informed decision tion and entry into provincial databases. 2 PREMNOTE • Limited meta data or reconciliation of data fields in the Local Government Context and NEMIS, because each province has a different EMIS Authority under Decentralization questionnaire. The PMT for education was conceived and implemented • Lack of understanding of the value of available infor- when a local government system was still operational in mation. Pakistan. Under that system, districts had some autonomy • Inadequate capacity to use EMIS data; when used, de- and authority, as specified under the LGO 2001, the cisions are focused primarily on facility construction/ brainchild of then President Pervez Musharraf.5 One of expansion and textbook provision. the first acts of Musharraf after the October 1999 coup, A final challenge has been the large time lag between which removed the government of Nawaz Sharif, was the annual data collection and the printing and distribution of announcement of a “seven-point agenda” for reform that the statistical reports from the NEMIS. In the first few years included a devolution plan to be initiated with a series of lo- after the start of the NEMIS, there was a time lag of 26–30 cal government elections. The elections were held between months from the end of data collection to the publication December 2000 and September 2001 and resulted in ap- of results. In the early 2000s, this time lag dropped to 18 proximately 200,000 new officials. Other elements of the months and was gradually reduced even further; the most devolution plan included autonomy of district departments recent Pakistan Education Statistics Report was published in and checks and balances through such monitoring mecha- 2013 and includes information for 2011–12. nisms as school management committees, parent-teacher Even though each province has its own EMIS ques- associations, and citizen community boards. tionnaire, similar information is collected in all four. For Figure 2 illustrates the administrative structure of the example, EMIS indicators reported on in all provinces districts and district officials involved in educational service include: the school building and its condition; enrollment; delivery between December 2000 and September 2010 in repeaters (by grade and subject); number of students per- Pakistan. Below the district mayor and district coordination manently absent; number of teachers in the school (as well officer there were 10 departments, one of which was edu- as their qualifications); results of board exams; facilities cation. Each of the 10 departments, including education, available in the school; the school management committee was headed by an executive district officer (EDO).6 While or parent-teacher councils; provision of free text books; the federal Ministry of Education maintained overall re- planned construction; and stipends for girls. sponsibility for education policy, planning and curriculum development, service delivery was devolved to the provinces and districts. Provincial departments of education coor- dinated and supported education at the district level and were headed by the provincial minister of education. At the Figure 1. Flow of EMIS Data in Pakistan district level, the EDO for education had Director National EMIS National EMIS substantial authority. Director EMIS Provincial EMIS unit Challenges in Education Service Delivery DO Under the 2001 LGO, districts were District (EMIS) responsible for all primary (grades 1–5), secondary (grades 6–10) and higher secondary (grades 11–12) edu- Municipal authority DDO/ADO cation, and could recruit both primary and secondary school teachers up to basic pay scale (BPS) 16.7 Authority Union council Supervisor and LCOs over hiring, firing, and transfers for BPS 17 and above remained with the provincial government. Funding for School Principal, head teacher, teachers education came from the national, provincial and district governments, Source: Zaidi 2003. but most of the expenditures were for Note: District officer (DO); deputy district officer (DDO); assistant district officer (ADO); local salaries—sometimes as much as 95 to circle offices (LCOs). LCOs are local teachers resource centers, providing opportunities for in- service training for teachers. 97 percent; leaving very little for other PREMNOTE 3 Figure 2. Administrative Structure of Districts • Free cooking oil to families as an incentive to enroll for Education Service Delivery girls in school. • A monthly stipend for female students, Rs 100 per District Mayor month in KP for primary education, and Rs 200 per month in Punjab for female students in secondary Council Deputy District Mayor schools (grades 6–8) in 16 districts. • Local financing under Khushal Pakistan Programs I and Audit II (KPP-I and KPP-II). Under KPP-I, each member of the District Coordination National Assembly (lower house) and the Senate (upper Officer (DCO) house) was allocated approximately US$80,000 each year to carry out minor development projects in his or Deputy District Officers her constituencies. KPP-I covered multiple sectors such • Civil defense as health, education, sanitation, roads, electrification of Executive District • Human resource villages, gas, and telephone service. KPP-II (also called Officer (EDO) Education management the Roshan Pakistan Program) was managed by the • Boys’ schools Prime Minister’s Secretariat in Islamabad and included • Girls’ schools development work under the prime minister’s directive. • Technical education • Colleges Donor-funded reforms:8 • Athletic education • Focused mostly on primary education, by decreasing • Special education the number of vacant posts in primary education as well as teacher absenteeism. Source: Nayyar-Stone et al. 2006. • Worked on building the skills and capacity of teach- Note: Dotted line denotes direct subordination. ers (Asian Development Bank [ADB], Decentralized Elementary Education Project; USAID, Education expenses. The result was inadequate school repair and Sector Reform Assistance; Aga Khan, Canadian In- maintenance as well as inadequate furniture, equipment, ternational Development Agency [CIDA], and the electricity, boundary walls, and sometimes even drinking United Kingdom’s Department for International water for the students. Development [DfID]). Public schools had high drop-out rates, with varying • Provided performance-based grants for education policies aimed at keeping students in school. For example, (World Bank, Education Sector Reform Program). in Thatta district (Sindh province), only 50 percent of the • Helped form district education management teams students who entered first grade moved onto second grade. (DEMTs), high-level management coordination and In Punjab, there was a 40–50 percent drop-out rate between decision-making bodies to help smooth implementa- primary and secondary school. In response, KP province ad- tion of education activities in the targeted districts opted a policy of not failing any female students in primary (USAID, Districts That Work [DTW] project). school. In Punjab, some districts had a policy of not failing any • Provided training on EMIS to improve data quality, students in grades 1–3; and in both KP and Punjab provinces, initiate evidence-based management and decision schools held a “0 period,” where teachers came in early to making, and enhance data analysis and interpretation tutor poorly performing students. Many districts also faced (USAID, DTW). other serious problems, including poor school infrastructure, teacher absenteeism, and unfilled teacher vacancies. In addi- Use of EMIS to Improve tion, poorly targeted or limited teacher training exacerbated Education Indicators the existing problems in the education sector. Although the The PMT for education was developed after the KP’s provin- LGO permitted district governments to pay monetary incen- cial education department asked USAID’s DTW for assis- tives and bonuses to staff for good performance, such awards tance in strengthening its existing EMIS. In response, during were rare due to limited financial resources. a two-day workshop (in December 2008), representatives Within this context, several education sector reforms, from 10 districts and the provincial EMIS cell identified both federal and donor funded, were initiated in Pakistan the 30 worst-performing primary schools in each district during President Musharraf’s administration. based on EMIS data. By the end of the workshop, the group Federally funded reforms included: had outlined improvement targets for the 10 districts, and • Free text books for female students up to grade 10. drafted action plans to achieve these goals. 4 PREMNOTE After the workshop for KP province, workshops were select the key performance indicators from the EMIS to also held for 13 districts from Punjab province (in March be used to identify and rank the low-performing schools 2009) and 10 districts from Sindh (in April 2009). By in the district. The selection of indicators and ranking of the end of 2009, seven additional districts from KP and schools, with some variation across the three provinces, was four from Sindh also began using the PMT for educa- conducted by a group comprising provincial and district tion—bringing the number of districts using this tool to education officers, head principals from a few schools, EMIS 44. Start-up workshops for each district/province were data programmers and operators, and parents of children followed by: attending primary school. This group examined and con- • Meetings with district education officials to review sidered several sources and indicators, including Millen- their action plans. nium Development Goal indicators for primary education • Sharing of workshop proceedings with the DEMT and core education indicators established by the federal including the list of education indicators targeted for Ministry of Education. Different districts chose different improvement in the district. education indicators. The most common set included: • Collecting data on the indicators and verifying the • Promotion rates: The proportion of students who data. successfully complete a grade and are promoted to • Holding review meetings with the head teachers of the next grade. all low-performing schools, along with the concerned • Repetition rates: The proportion of students who district education officers. repeat a grade once or twice. • Selecting two master trainers from each district (in • Drop-out rates: Proportion of students who leave consultation with the provincial education depart- school without completing the grade in which they ment) and training them on implementing the PMT were enrolled. for education in their districts. • Teacher absenteeism: Percentage of teachers absent • Training of head teachers and parent-teacher com- from school. mittee members by the master trainers, and technical • Student-teacher ratios: The number of students per assistance for developing and implementing a school teacher. action plan. • Student-classroom ratios: The number of students The design and implementation of the PMT for educa- per classroom. tion initiative across the three provinces took 15 months, Step 3: Use performance management techniques with results seen as early as 6–9 months into the process. and EMIS to identify the initial and then final set of low- Implementation included a substantial amount of training performing schools in the district. on the various components and stages of the tool (in total, • Conduct a trend analysis of provincial EMIS data 1,548 officials and teachers were trained in KP, Punjab, (examine data for the last three years) to identify low- and Sindh provinces) and use of the EMIS (61 people were performing schools in each district based on one or a trained across the three provinces). The PMT for education combination of key performance indicators. initiative was also supported by various trainer and partici- • Select common indicators from the provincial EMIS to pant guides and manuals for designing and implementing identify reasons for the low performance of identified the tool as well as using the EMIS. schools. For example, the student-teacher ratio, num- Overall, the process comprised three phases and nine ber of infrastructure facilities lacking in the schools, or steps. the number of parent-teacher meetings held in a year. Phase 1 • Re-rank the selected schools using both key and com- Step 1: Conduct situation analysis of targeted schools in mon indicators and a school scorecard. the district. The districts conducted a general situational Step 4: Develop district action plans for improving the analysis across primary schools, with each school defining low-performing schools. The district action plan, which its basic needs and priorities. The analysis also focused on also identified the targets for improvements, included the identifying gaps as well as future challenges and opportuni- actions required to achieve the targets; the person(s) respon- ties and helped the schools recognize factors that cause poor sible; resources required; source of funding; any collabora- performance (step 3). The situational analysis became the tion/participation of citizens, civil society, or others; time foundation for the schools’ action plans (step 7). frame (including start and completion time); and remarks. Step 2: Select key performance indicators from the The overall responsibility for implementing the action plan provincial EMIS. A stakeholder consultation was used to rested with the EDO of education. PREMNOTE 5 Phase II identification of the delay and of possible corrective actions Step 5: Validate school data. Since there are thousands of to get back on track. schools in each district, only data from the low-performing Step 9: Report annually and take actions for improve- schools were validated to update any missing data or incom- ments. This included reporting on the status of the low- plete forms and ensure that the baseline and targets were performing schools and highlighting those schools that accurate for the school. The verification focused on the key improved their performance. This was considered critical and common performance indicators chosen by the district in keeping stakeholders informed of improvements in the from its EMIS. school system. Step 6: Train master trainers on developing school ac- tion plans. One male and one female assistant district officer Results of the PMT for Education: were selected in each district and tasked with becoming the Factors for Success master trainers to train other officers and school staff in Table 1 presents the results of the implementation of the developing and monitoring school action plans. PMT for education across KP, Punjab, and Sindh provinces. Step 7: Develop school action plans. Developed in Table 1 confirms the findings of the 2011 NEMIS consultation with stakeholders, the school action plan was survey, that EMIS data are primarily used in improving the roadmap to achieving the objectives and targets set by school infrastructure. However, the PMT for education each of the low-performing schools in consultation with also focused on improving at least one outcome indicator. the district education officers. For each indicator and target, While KP channeled efforts on decreasing the number of the school action plan included subactivities, responsibility, “low-performing” schools—a combination of several output resource required, sources of funding, collaboration, start and outcome indicators, Punjab and Sindh provinces sought and end times, and remarks. improvement in teacher absenteeism and reduction in repetition rates, respectively. Phase III There are several factors that made the PMT for educa- Step 8: Implement and monitor district and school action tion process successful in Pakistan and in turn led to the plans. The school action plans were implemented by as- use of EMIS data in decision-making and improvement in signing responsibilities to specific individuals and ensur- education service indicators: ing sufficient resources. Monitoring helped identify gaps Championed by the government. The initial request for between planned and actual timelines and targets. If targets assistance in improving the provincial EMIS and using the or deadlines were missed, the school action plan required data in the decision-making process came from the edu- Table 1. PMT for Education Results No. of District districts action Schools with Change in implementing plans parent-teacher province- PMT for realized Improvement of school councils specific Province education (%) infrastructure activated (%) indicators KP 17 70 Percent of low-performing 92 Decrease in schools now having: “low-performing” Electricity: 66% schools: 89% Water: 72% Boundary wall: 67% Punjab 13 — Decrease in schools without: 93 Decrease in Electricity: 41% teacher Boundary wall: 35% absenteeism: 30% Drinking water: 63% Functioning toilet: 40% Functioning washroom: 50% Sindh 14 70 Decrease in schools without: 88 Reduction in Electricity: 15% repetition rate: 65% Boundary wall: 31% Source: DTW reports. Note: — = results not available. 6 PREMNOTE cation department of the KP government. Participation Tremendous support from a large, donor-funded of the provincial EMIS cell and buy-in from the districts project. The process was introduced and supported by a ensured local ownership throughout the entire process by well-funded donor project, Districts That Work, which central actors. signed memorandums of understanding with several dis- Built on existing data familiar to all officials. Since the tricts and worked with them on different fronts, including EMIS is based on an annual census, all education officials are providing technical assistance support and grants as incen- familiar with it, and most are involved in its data collection. tives. Training and technical assistance during design and By the end of a two-day workshop, officials were quickly able implementation of the PMT for education, along with to identify key and common performance indicators that necessary manuals and training guides, helped keep the needed improvement and develop practical action plans process on track. with targets. The development of a practical tool that used existing data led to its quick expansion and acceptance by Options for Replication Punjab and Sindh provinces. There are legitimate questions regarding replication of the Quick wins generated interest and replication in other Pakistan experience in other countries. This note does not districts. Clear and tangible results attracted the attention attribute improvement in education indicators solely to of other provinces, which then also requested technical the PMT for education, given the large complementary assistance in using the EMIS, and led to expansion into support provided to the education sector by USAID and additional districts. other donor projects. Nor does this note claim that similar Local authority. The 2001 LGO gave provincial and funding or technical assistance is required to see comparable district governments some autonomy over decision making use of EMIS elsewhere. In fact, the three phases and nine and resource allocation. This autonomy enabled the district steps of the PMT for education can be greatly simplified. and school action plans to be very localized, and additional The process worked in Pakistan largely due to the support support and resources were provided by the districts when provided by the DTW project at a time when districts still needed. Linking the EMIS data to district and school action had considerable local authority. Evidence suggests that this plans helped commit managers and other stakeholders to tool is no longer in use by the provinces. this process, because they were then able to see how the Nevertheless, elements of the Pakistan initiative do data benefited the planning process (Mark 2014; Powell provide lessons that can be applied elsewhere. In many and Trucano 2006). countries, schools and local governments are already fa- The EMIS originated at the district level. Each school, miliar with EMIS data because they are responsible for its and the district officer in charge of the school, was respon- collection. In a decentralized setting, where fiscal, admin- sible for collecting EMIS data. The districts did not have istrative, and policy autonomy reside at the local level, the to wait to get current NEMIS data to identify the low- EMIS can provide quality data for use in evidence-based performing schools and prepare their action plans. The lag decision making, and modest capacity-building efforts can in data reports from the national level did not affect them. be enough to make that possible. The verification step also helped ensure that the data were The majority of countries worldwide already have accurate. an EMIS. Many officials have, at the ready, information to Use of the EMIS created incentives for verification of improve education service indicators. Even data of uneven data quality. Once schools were targeted for improvement, quality can be used to make service improvements. Refin- all of the common and key performance indicators were ing the data can become an ongoing process. With the verified by the individual school before formulation of its existing reservoir of data that EMISs represent worldwide, action plan. This step ensured that baselines were accurate the potential for improvements in the education sector is and targets for improvement realistic. The demand for substantial. better data and information for decision making created incentives to improve data quality. Typically concerns about data accuracy prevent government officials from making Acknowledgments decisions based on available data. In Pakistan, “some dis- The PMT for education was spearheaded by Rukhsana tricts felt that the error rate could be as high at 30 percent. Sadiq, Director Education Program for Districts That Decision makers also show a tendency to disparage the Work, a USAID-funded local government and decentral- accuracy of the data. If they can sell the idea that the infor- ization project in Pakistan during 2006–10. Its concept mation is inaccurate then they are free to make decisions was based on performance management principles, and based on political rather than factual needs.” (Government its implementation the result of several local project of Pakistan 2011). staff. The author thanks Keith Mackay, Anya Reva, and PREMNOTE 7 Katharine Mark for valuable comments on earlier versions 7. Each BPS has a large range; BPS 16 had a minimum of of this paper. US$58–US$193 per month (2008 rates). BPS 22 had a maximum range of US$265–US$530 per month. About the Author 8. The most active donors were: USAID, World Bank, DfID, and ADB. Ritu Nayyar-Stone is Senior Research Scientist at NORC at the University of Chicago, which she joined in March 2013. Her areas of expertise include evaluations and assessments, References survey design and analysis, governance, service delivery, Cassidy, Tom, 2005. Education Management Information System monitoring, and performance management. Previously, (EMIS) Development in Latin America and the Caribbean: Lessons and Challenges. Washington, DC: Inter-American Nayyar-Stone was Senior Research Associate at The Urban Development Bank. Institute, where she worked for nearly 18 years. She was Hatry, Harry, P. 2006. Performance Measurement: Getting Results, technical manager of the Districts That Work project in Second Edition. Washington, DC: The Urban Institute Press. Pakistan, overseeing all project components and technical Mark, Katharine. 2014. “Performance Management Can Improve reports. She has a PhD in economics from Boston University. Local Services in Developing Countries: The Service Improve- E-mail: nayyarstone-ritu@norc.org. ment Action Plan.” Special Series on The Nuts & Bolts on M&E Systems Number 29, World Bank, Washington, DC. Pakistan, government of. 2011. National EMIS Survey Report: Notes Findings and Recommendations; A Baseline for Strategic De- 1. Two of the eight goals and 6 of the 48 indicators focus velopment, Second Edition. Academy of Education Planning and Management and Ministry of Professional & Technical on education. Training, funded by USAID, Islamabad. 2. The system was initially conceived by UNESCO to be ———. 2013. “Pakistan Education Statistics 2011–2012.” Na- quickly and easily customized to meet the specific needs tional Education Management Information System, Acad- of the educational systems of its member states (see www. emy of Educational Planning and Management, Ministry openemis.org). of Education, Training and Standards in Higher Education, 3. This initiative was funded by the U.S. Agency for Inter- Islamabad. Nayyar-Stone, Ritu, Robert Ebel, Sonia Ignatova, and Khalid national Development, under its US$26 million Districts Rashid, with Harry Hatry and George Peterson. 2006. “As- That Work project, and was implemented by the Urban sessment Report: Pakistan Devolution Support Project.” UI Institute from August 2006 to March 2010. Project No. 07862, Urban Institute, prepared for USAID. 4. This report includes information for 2011–12 and trend Powell, Marcus, and Michael Trucano. 2006. “Rethinking Educa- analysis for some indicators spanning six years, 2006–7 to tion Management Information Systems: Lessons from and 2011–12. There is also a note that data for the number of Options for Less Developed Countries.” infoDev, Working Paper No. 6, Washington, DC. public and private education institutions are based on the Sadiq, Rukhsana. 2009. “A Guidebook on Performance Man- latest National Education Census conducted in 2005–6. agement System for Schools.” Districts That Work Project, 5. Pakistan is a federal country, with a national govern- USAID, Pakistan. ment and four provinces: Punjab, Sindh, Baluchistan, and Urban Institute. 2010. “Final Report, Pakistan Districts That Khyber Pakhtunkhwa. The LGO 2001 formed three tiers Work, August 2006 to March 2010.” UI Project Number of government under the provinces: district, municipal, 07862-003, Washington, DC. and union councils. World Bank. 2013. “The What, Why, and How of the Systems Approach for Better Education Results (SABER).” Conference 6. The 10 district departments included: agriculture, com- Edition No. 79901, Washington, DC. munity development, education, finance and planning, Zaidi, Mosharraf. 2003. “EMIS in Pakistan: Process, People and health, information technology, law, literacy, revenue, and Structures: Examining Policy Issues in Education Data.” Edu- works and services. cation Sector Reform Assistance Program, USAID, Pakistan. This note series is intended to summarize good practices and key policy findings on PREM-related topics. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank. PREMnotes are widely distributed to Bank staff and are also available on the PREM Web site (http://www.worldbank.org/ prem). If you are interested in writing a PREMnote, email your idea to Madjiguene Seck at mseck@worldbank.org. For additional copies of this PREMnote please contact the PREM Advisory Service at x87736. This series is for both external and internal dissemination 8 PREMNOTE