47433 The World Bank in Vietnam HOW DO WOMEN FARE IN EDUCATION, EMPLOYMENT AND HEALTH? A Gender Analysis of the 2006 Vietnam Household Living Standard Survey FINAL REPORT December 2008 2 Acknowledgement This report was prepared by Sunhwa Lee (International Consultant) with programming and research assistance from Trinh Thu Nga at the Institute of Labor Science and Social Affairs (ILSSA). The project team leaders at the World Bank were Carrie Turk and Hoa Thi Mong Pham who initiated the project and provided guidance for research and preparation of the final report. Along with team leaders, many people have also contributed to this report. Ingrid Fitzgerald (UNDP), Geetanjali Narayan (UNICEF), and Marjo Ritta Tervonen (UNDP) provided guidance for research with ongoing gender discussions in Vietnam. Gillian Brown (World Bank), Tran Thi Van Anh (World Bank gender consultant), and Sandy Lieberman (World Bank) provided insightful comments on an earlier draft of the report. Nguyen Van Thuy and Nguyen The Quan (General Statistical Office in Vietnam) aided our research by clarifying several indicators in the Vietnam Household Living Standard Survey; Gian Thanh Cong (ILSSA) kindly provided data tables that were helpful in the course of research. 3 Table of Contents Acknowledgement ............................................................................................................................ 2 Table of Contents ............................................................................................................................. 3 List of Figures................................................................................................................................... 4 Introduction ...................................................................................................................................... 5 1. Demographic Characteristics of Households ............................................................................... 5 Poverty Rates................................................................................................................................ 7 Female versus Male Headed Households..................................................................................... 9 Household Characteristics of Ethnic Minorities ........................................................................ 10 2. Education .................................................................................................................................... 12 Trends in Educational Attainment.............................................................................................. 12 School Attendance by Region, Ethnicity, and Family Characteristics....................................... 14 Attendance at Extra Classes ....................................................................................................... 17 3. Employment and Earnings ........................................................................................................ 19 Labor Force Participation Rates ................................................................................................. 19 Number of Weeks Worked on Income Generating Activities.................................................... 20 Housework ................................................................................................................................. 21 Types of Employment between Women and Men ..................................................................... 22 Gender Segregation in Industry and Occupation ....................................................................... 25 Gender Gap in Earnings in Wage Employment ......................................................................... 28 4. Health Status and Access to Health Care .................................................................................. 30 Gender and Indicators of Health ................................................................................................ 31 Access to Health Care Services .................................................................................................. 34 Types of Health Care Services Used and Reasons for Their Use .............................................. 36 5. Disability and Impact on Work Participation ............................................................................ 37 Prevalence and Types of Disability ............................................................................................ 38 Causes of Disability ................................................................................................................... 41 Disability, School Attendance, and Work Participation............................................................. 42 Suggestions for Future Research .................................................................................................... 43 References ...................................................................................................................................... 48 Tables ............................................................................................................................................. 50 Appendix A-1. Classification of Ethnic Minority Groups ............................................................ 85 Appendix A-2. The 2006 VHLSS Questions on Disability Used for Analysis ............................. 86 4 List of Figures Figure 1-1 Household Characteristics in 2006 VHLSS ........................................... 6 Figure 1-2 Poverty Rates by Gender and Selected Characteristics .............................. 8 Figure 2-1 Distribution of Major Fields of Study among Men and Women in Higher Education (Current Attendees) ................................................................ 14 Figure 2-2 Percent Attending School among 15-17 Year Olds ................................... 15 Figure 2-3 Percent Attending School among 18-21 Year Olds ................................... 17 Figure 3-1 Labor Force Participation Rates in Past 12 Months .................................... 20 Figure 3-2 Percent People Reporting Not Doing Any Housework by Gender and Area ...... 22 Figure 3-3 Industry Distribution of Main Jobs by Gender (Age 18-64) ........................ 24 Figure 3-4 Industry Distribution of Main Jobs by Gender (Age 18-64) ........................ 26 Figure 3-5 The Degree of Female Concentration by Occupation ................................ 28 Figure 4-1 Indicators of Health Status by Gender .................................................. 32 Figure 4-2 Percent Reporting Absence from Normal Activities Due to Illness by Gender and Age (in Past 12 Months) ............................................................ 32 Figure 4-3 Average Number of Days Missing Normal Activities among People Reporting Illness (in Past 12 Months) ............................................................... 34 Figure 4-4 Percent People Using Health Care Services (Among People Reporting Illness in Past 12 Months) ........................................................................ 35 Figure 5-1 Prevalence of Disability by Gender and Age .......................................... 39 Figure 5-2 Causes of Disability by Gender and Age Group ...................................... 41 5 Introduction Along with remarkable achievements in reducing poverty during the past decade, Vietnam's social and economic development policies have placed much emphasis on promoting gender equality. From a perspective of gender equality, women in Vietnam are considered in a relatively favorable position compared with women in other developing countries or other developed Asian countries, with a high rate of women's labor force participation and a high degree of women's representation in political positions including the National Assembly. With the new Law on Gender Equality passed in November 2006, more policy efforts are called for in achieving gender equality in both public and private spheres of people's lives. In this context, it is all the more important to have up-to-date information on various indicators of gender equality in order to accurately assess the current situation of gender disparities in Vietnam. This task is essential for formulating policies that address specific problem areas of gender disparities, and for developing effective strategies for implementing and monitoring gender equality policies. Using data from the recent 2006 Vietnam Household Living Standard Survey (VHLSS), this report assesses the current situation of Vietnamese women vis-à-vis men in several key areas: household demographics, education, employment, health, and disability. Some indicators presented here, such as those in household demographics, education, and employment, provide updated information to previous reports which were based on the 1998 Vietnam Living Standard Survey (VLSS) and 2004 VHLSS. 1 This report presents descriptive statistics on various indicators in the above areas in order to assess not only gender differences, but also gender-related differences by region, ethnicity, living standards, and so on. Descriptive analyses presented here are based on statistical tabulations of mostly bivariate relationships; these analyses provide a primary basis for discerning important relationships between variables under consideration. It is important to note that they do not necessarily indicate causal relationships. Multivariate analyses will be required to make conclusive statements about causality in the results presented here, which is beyond the scope of this study. The main objective of this report is to examine key indicators of women's social and economic lives with the most recent data, so that we have a better understanding of the nature and the extent of gender inequality in contemporary Vietnam. This will help us to evaluate progress made thus far and identify areas for current and future challenges toward gender equality. 1. Demographic Characteristics of Households Basic demographic characteristics of sampled households in the 2006 VHLSS are presented in Table 1-1, along with those in the 1998 VLSS and 2004 VHLSS. In the 2006 survey, the total of 9,189 households and 39,071 household members were sampled, and they are the focus of this study. Throughout this study, we report descriptive statistics which are calculated using sample weights in order for the survey sample to be representative of civilian, non-institutionalized population. (Sample sizes presented throughout the study are unweighted totals to indicate the actual number of sampled households or individuals being analyzed.) 1 Data from the 1998 VLSS are analyzed in Vietnam Through the Lens of Gender: Five Years Later (by Jaikishan Desai, 2000), and data from the 2004 VHLSS are analyzed in Gender Analysis of the 2004 Vietnam Household Living Standard Survey (by Sunhwa Lee, 2006). 6 While basic demographic characteristics of Vietnamese households have not changed much between the 2004 and 2006 surveys, some characteristics represent continuing changes since the 1998 survey. Most households in 2006, about 73 percent, are in rural areas, but the proportion of urban households continues to increase (from 24 percent in 1998 to 28 percent in 2006; see Figure 1-1). Nuclear-family households--consisting of the couple and their children--have also been increasing since 1998, while vertical-family households that include grandchildren or the couple's parents have been declining. As of 2006, nearly three-quarters (73 percent) of Vietnamese households are made up of nuclear families, and less than 1 in 5 households (19 percent) are vertical families. The "other" type of households that includes other relatives or non-relatives has also increased somewhat since 1998, but these households still account for only a small proportion in 2006 (8 percent).2 Figure 1-1. Household Characteristics in 2006 VHLSS Percent (%) 100 87.8 72.5 73.0 80 60 40 27.5 18.7 20 11.3 8.2 0.9 0 rity n h ral l ar r se a he Kin ba rtic cle i ne Ru no Ot Ur Ve Nu Ch mi nic Eth Area Household Type Ethnicity The vast majority of households in 2006 consist of the Kinh (88 percent), and ethnic minority groups--other than the Chinese--make up about 11 percent of all households (see Figure 1-1 and Table 1-1). Compared with 1998, the proportion of ethnic minority households (including the Chinese) has decreased slightly, while the proportion of the majority Kinh households has increased. As for the head of household, the typical Vietnamese household is still headed by a male (75 percent), and heads of households are mostly married (81 percent). Only a very small proportion of household heads are divorced or separated (less than 3 percent).3 The poverty rate calculated for all household members indicates a continuous decline in the proportion of people living below the poverty line (Table 1-1).4 While more than 1 out of 3 2 With the 2004 or 2006 VHLSS, it is not possible to discern who these "other" household members are. In these surveys, except for parents, parents-in-laws, grandparents, and grandchildren, all others are included in the "other" category without any specification. Unlike the 2004 or 2006 surveys, the 1998 VLSS included more detailed categories (e.g., brother/sister, niece/nephew) that specify other relatives or non-relatives. 3 Compared with male heads of households, female heads of households are much more likely to be divorced or separated, as discussed later in the characteristics of male-headed versus female-headed households. 4 The poverty rate throughout this study refers to the proportion of people living below the poverty line that is calculated by General Statistical Office (GSO) in Vietnam with support from the World Bank, using household 7 people in 1998 (37 percent) lived in poverty, the rate fell to 20 percent in 2004 and it fell further to 16 percent in 2006. The rate of food poverty, which represents extreme poverty, was already low in 2004 (7.4 percent) and it declined further to 6.7 percent in 2006. We examine below the poverty situation in more detail, separately by gender as well as by other characteristics. Poverty Rates While the gender difference in the overall poverty rate is small, large disparities exist by region, ethnicity, and age. Males and Females in rural areas are five times more likely than their urban counterparts to live in poverty (20 percent vs. 4 percent). More than half of ethnic minorities still live in poverty, compared with only 1 out of 10 Kinh/Chinese majority. Nearly 1 in 4 children under age 15 live in poverty compared with 1 in 7 adults, indicating a large concentration of young children in poor households. When we consider poverty rates for all household members separately by gender, the proportion living below the poverty line is slightly higher for females (16.3 percent) than for males (15.6 percent; the difference is statistically significant at p < .05). Since the poverty rate here is estimated using consumption expenditure data collected at the household level, this suggests that females are somewhat more likely than males to be concentrated in poor households.5 Compared with the gender difference, we see more substantial differences in poverty by area, region, and ethnicity (see Figure 1-2 and Table 1-2). There is a large disparity between urban and rural areas: the poverty rate is five times higher in rural areas than in urban areas--20 percent versus 4 percent--similarly for males and females. Likewise, the regions which are still largely rural or populated with ethnic minorities tend to have higher rates of poverty: the North East, the North West, the North Central Coast, and the Central Highlands show considerably higher rates of poverty--between 25 and 50 percent--than the national average. In particular, the North West shows the highest rate of poverty with half of all males and females still living below the poverty line. By contrast, the poverty rate is less than 10 percent for the Red River Delta and the South East which include large urban areas. In the regions where the poverty rate is high--especially the North East, North West, and the North Central Coast--we also observe relatively larger gender differences than other regions. By ethnicity, while only 10 percent of the Kinh/Chinese live below the poverty line, more than half of ethnic minorities still live in poverty (see Figure 1-2 and Table 1-2). Ethnic minorities in Vietnam consist of diverse ethnic groups which differ widely in geography, endowments, cultural norms/behaviors, and socioeconomic situations (Baulch et al. 2008a; Swinkles and Turk 2006; WHO 2003). When ethnic minorities are divided into four broad groups (see Appendix A-1 for details of the four groups), poverty rates vary substantially among them. As shown in Table 1-2, the Khmer and the Cham, residing mostly in the South East and the Mekong River Delta, have the expenditure data. The (general) poverty line indicates the minimum consumption level for both food and non- food items, with food consumption necessary to secure 2100 calories per person per day. The rate of food poverty indicates the proportion of people who cannot afford the minimum consumption level to secure 2100 calories per person per day. 5 The poverty rate using the GSO-World Bank poverty line is estimated with VHLSS data on per capita expenditure, derived from consumption expenditure data collected at the household level. Per capita expenditure is obtained by dividing total household expenditure simply by the number of people in the household. Hence, everyone in the household shares the same poverty status, and larger households are more likely to be poor. This method implicitly assumes an equal distribution of resources within the household--whether it be an adult or a child--without using adult equivalence scales or economies of scale. 8 Figure 1-2. Poverty Rates by Gender and Selected Characteristics % 60 50 Male Female 40 30 20 10 0 an l al 14 4 + se ity ta -6 65 ur or To ne rb 0- 15 R in e U hi e Ag m Ag e /C Ag ic nh hn Ki Et lowest rate of poverty (34 percent overall), followed by the Tay, Thai, Muong, and Nung in the Northern Uplands (43 percent; hereafter the Tay and others). Compared to these two ethnic groups, the other Northern Mountain ethnic groups and the Central ethnic groups, consisting of many smaller ethnic groups, still exhibit a poverty rate of more than 70 percent. Compared to the Kinh/Chinese, the gender difference in the poverty rate is also slightly greater among ethnic minority groups, especially among the Tay and others, and the Northern Mountain ethnic groups. Among the Central ethnic groups, on the other hand, the poverty rate is slightly higher for males than for female. When we examine poverty rates by age, the highest rate is seen among children under age 15; nearly one-quarter of children aged 0-14 (24 percent) live below the poverty line and there is no gender difference (see Figure 1-2 and Table 1-2). (Please note that the discussion on poverty by age here is based on poverty rates estimated without using any adult equivalence scale, as usually done in studies using the GSO-World Bank poverty line.) 6 Although the overall poverty rate has fallen to 16 percent by 2006, a higher rate of poverty among children indicates that larger numbers of children are still concentrated in poor households. The poverty rate is lower for those in the prime-working age or later in their working lives, ages 35 thru 64. These are also the age groups where there is little gender difference in poverty rates. Among those aged 65 and older, on the other hand, a higher proportion of women (17 percent) live in poverty than men (14 percent). 6 The widely used poverty measure based on per capita expenditure in the VHLSS--which is estimated by dividing total household expenditure simply by the number of household members--may underestimate or overestimate poverty rates of individual members under different household settings. Many researchers use varying types of equivalences scales, which often assume the level of children's consumption to be a certain fraction of that of adults' consumption and which also account for economies of scale in larger households. There is a large body of literature discussing different types of equivalence scales, their underlying assumptions, their impacts on poverty measures, and difficulties in definition and measurement, for the context of both developed and developing countries (e.g., Betson 2004; Deaton 1997; Nelson 1993; White and Edoardo 2002;World Bank Institute 2005). Some studies indicate that children's poverty rate would be lower in Vietnam when certain equivalence scales are used (White and Edorardo 2002). Other researchers, on the other hand, point out that equivalence scales include assumptions about intra-household allocation of resources which are difficult to verify empirically (Betson 2004); which very often include policy judgments (Nelson 1993); and the results of which use are not compelling enough to be easily used in practical poverty analysis (World Bank Institute 2005). 9 With respect to household structure, Table 1-2 shows that both males and females living in nuclear families are less likely to live in poverty (14 percent) than those in vertical households or others.7 By gender of household head, the poverty rate is lower for those living in female-headed households (12 percent) than those in male-headed households (17 percent). While female- headed households in other social contexts usually refer to households headed by single females without male adults, it is important to keep in mind that female-headed households in Vietnam include a substantial proportion of married-couple households where husbands are present, as examined below. Within the same type of households, the poverty rate tends to be slightly higher for females than for males. Female versus Male Headed Households Female-headed households in Vietnam consist of rather distinct types of households, including those headed by married women whose husbands are present in the household. These households headed by married women have unusually high living standards, compared with other female-headed or male-headed households. Female-headed households make up about one-quarter of all households in Vietnam, and these households are rather distinct from male-headed households in several respects. First of all, female-headed households, albeit their smaller proportion, encompass more diverse types of households than male-headed households (see Table 1-3). The marital status of household heads shows that male heads of households are mostly married men (96 percent); only a small proportion of them are widowers (3 percent). By contrast, nearly half of female heads of households (47 percent) are widows, and 38 percent are married women. Another 8 percent of female heads of households are divorced or separated women. Secondly, female household headship in Vietnam does not necessarily mean the absence of adult males. Not only are a sizable proportion of female-headed households headed by married women whose husbands are present in the household (88 percent of married female-headed households), but even many households headed by widows also have adult male members (aged 18 and older) in the household. About 61 percent of widow-headed households have adult males aged 18 and older in the household, and about 32 percent of these households are vertical households that include their children's families. Thirdly, female-headed households are over-represented in urban areas and the South East, especially those headed by married women or divorced/separated women. More than half of households headed by married women (55 percent) and 46 percent of those headed by divorced/unmarried women are in urban areas, compared with only 23 percent of male-headed households overall. On the other hand, the majority of households headed by widows are in rural areas (70 percent), similar to those headed by widowers (68 percent). In other words, the distinctiveness of female-headed households mostly comes from the households headed by married or divorced/unmarried women, rather than from those headed by widows. (A relatively high proportion of households headed by divorced/unmarried males are also in urban areas, but the number of these households is very small.) 7 The higher rate of poverty in vertical or other households than nuclear households is in part due to differences in household size. The average number of household members is 3.8 for nuclear families, 5.3 for vertical households, and 5.6 for other households. 10 Fourthly, the distinctiveness of households headed by married women or divorced/unmarried women can be further illustrated by the level of schooling completed by these women, their employment status, and their living standards (see Table 1-3). The distribution of completed levels of schooling indicates that married women who are heads of households include the highest proportion of those who completed upper secondary or higher education (30 percent), compared with either widows (6 percent) or male heads of households (19 percent). The percentage of these married women who are employed is somewhat lower than their male counterparts (85 percent compared with 91 percent), but their proportion of wage workers among the employed (30 percent) is similar to that of their male counterparts. (As will be examined later on the section on employment, only about one-quarter of female labor force overall is represented in wage employment.) In addition, more than two-thirds of household members in married female-headed households (67 percent) are represented in the two richest quintiles. As a result, the poverty rate is the lowest at 7 percent for those in married female-headed households, in comparison with those in other female-headed or male-headed households. While the overall poverty rate appears to be lower for those living in female-headed households than those in male-headed households, this difference needs to be interpreted with great caution. When we compare poverty rates across different marital status of household heads, the poverty rate is rather similar for those within the same marital status, except for those in households headed by married women and married men. For example, the poverty rate for those in households headed by widows and by widowers is very similar, at around 15 percent. Yet, the poverty rate is much lower for those in married female-headed households than those in male- headed households (7 percent versus 17 percent), mainly because of unusually higher living standards observed in the households headed by married women. It is beyond the scope of this descriptive report to explore the concept of household headship in Vietnam or possible reasons underlying household headship by married women even when their husbands are present in the household. While married women with greater earnings power or decision making authority vis-à-vis their husbands may become the head of household in some social contexts, this does not appear to be a main reason for married women examined here. A brief examination of husbands' education and employment characteristics among married female heads of household shows that their husbands have even higher levels of education and are more likely to be employed in the wage sector, especially in the government or state-owned enterprises (SOEs), both in comparison with married women themselves or other male heads of households.8 It will be important in future research to further explore the meaning of household headship among married women, for instance, how their household headship influences actual decision making in the household or their legal rights over household assets. Household Characteristics of Ethnic Minorities The four broad ethnic groups considered in this study differ not only from the Kinh/Chinese but also among themselves, with respect to household structures and socioeconomic characteristics of household heads. Among the four ethnic groups, the Tay and others are most likely to have household heads who completed at least lower secondary 8 For instance, 13 percent of these husbands have completed higher education (junior college or more) and another 22 percent completed upper secondary education, compared with only 5 percent and 14 percent of male heads of household, respectively. Also, 51 percent of these husbands have jobs in the wage sector, compared with 31 percent of male household heads; and 28 percent of the husbands have jobs in the government or SOEs, compared with only 11 percent of male household heads overall. The average age of these husbands is 49.6, compared with 48.2 for male household heads overall. 11 schooling, while the Khmer/Cham have a higher share of household heads in wage employment than others. These two ethnic groups have a substantially lower rate of poverty than the Northern Mountain or the Central ethnic groups. Given considerably higher rates of poverty among ethnic minorities compared with the Kinh/Chinese, and also given varying rates of poverty among different ethnic groups, we examine in detail the characteristics of households among four different ethnic groups in comparison with those of Kinh/Chinese households (see Appendix A-1 for details on the four ethnic groups).9 The four broad ethnic groups presented here not only differ in geography from each other and from the Kinh/Chinese, but they also differ in household structures and socioeconomic characteristics of household heads. These household characteristics provide important backgrounds by which to understand the gender situation in their school attendance, economic activities, and health status, which will be examined closely in later sections. Table 1-4 shows that all four ethnic groups are overwhelmingly concentrated in rural areas (more than 90 percent of each group), compared with the Kinh/Chinese (70 percent). As for household structure, the majority of households from all four ethnic groups are also nuclear households, but ethnic minorities households are more likely than Kinh/Chinese households to consist of either vertical households or to include other relatives and non-relatives who are not direct family members (see the percentage of "other" households). As a result, the average household size is larger for ethnic minorities (5.0 overall) than for the Kinh/Chinese (4.1), and it is especially large for the Northern Mountain (5.6) and Central ethnic (5.5) groups. The average age of household heads, the presence of children, and the presence of elders indicate that the Tay and others, the Northern Mountain, and the Central ethnic groups consist of relatively younger households compared with the Kinh/Chinese or the Khmer/Cham. The Central ethnic groups in particular are most likely to have very young children (age under 6) in the household at 51 percent, which is twice as high as the percentage among the Kinh/Chinese or the Khmer/Cham. The Khmer/Cham, on the other hand, are more similar to the Kinh/Chinese than to the other three ethnic groups in the average age of household heads and the presence of elders aged 60 and older. The gender of household heads shows that all ethnic groups, except the Khmer/Cham, are even more likely than the Kinh/Chinese to have a male as head of household. And, most household heads in the three ethnic groups are married. Khmer/Cham households, on the other hand, are similar to the Kinh/Chinese in their share of female heads of household but have the highest proportion of widow(er)s as heads of household (20 percent). This is associated with the fact that Khmer/Cham households consist of a higher proportion of vertical households than others and include relatively older family members. The average age of household heads also suggests that despite similar presence of elders between Kinh/Chinese and Khmer/Cham households, household headship is likely to be held more by older members in Khmer/Cham households. Considerably higher rates of poverty among ethnic minorities than among the Kinh/Chinese are closely associated with socioeconomic characteristics of household heads. All ethnic minorities groups are significantly disadvantaged in their levels of schooling completed, compared with the Kinh/Chinese. In particular, household heads from the Central ethnic groups and the Khmer/Cham have the lowest level of schooling overall, with 70 percent of them having either no schooling at all or less than primary schooling. Among the four ethnic groups, household heads from the Tay and others are most likely to have completed primary or lower secondary education, 9 The Chinese are combined with the Kinh, as done in most other research and statistics in Vietnam. 12 although they are still substantially less likely than the Kinh/Chinese to have higher levels of schooling. In terms of employment characteristics, the vast majority of household heads from the Tay and others, the Northern Mountain, and Central ethnic groups have worked in the past 12 months (more than 90 percent), while the rate of employment is slightly lower for the Kinh/Chinese (85 percent) and the Khmer/Cham (84 percent). Yet, work among the three ethnic groups is mostly confined to agricultural work, and the proportion engaged in agricultural work is especially high among the Northern ethnic groups (88 percent). By contrast, a substantially smaller proportion of household heads from the Khmer/Cham have been engaged in agricultural work (41 percent). In fact, employment characteristics between the Kinh/Chinease and the Khmer/Cham are quite similar. Although all four ethnic groups are considerably more likely than the Kinh/Chinese to live in poverty, these four groups diverge in their living standards, as indicated by the distribution in expenditure quintiles as well as the poverty rate. The vast majority of the Northern Mountain and the Central ethnic groups live in poor households: more than 90 percent of them are concentrated in the two poorest expenditure quintiles, with the poverty rate of more than 70 percent. For these two groups, extreme poverty measured by food poverty is also prevalent, with nearly half of them living below the food poverty line. The Khmer/Cham and the Tay and others are substantially better off than the other two ethnic groups with a poverty rate of 34 percent and 43 percent, respectively. Yet, their poverty rate is still more than three times higher than the rate among the Kinh/Chinese, and nearly half of them are also concentrated in the poorest quintile. This indicates that great policy attention is still needed in poverty reduction strategies targeted at ethnic minorities, including re-evaluation of effectiveness of current existing policies. It is important to keep in mind these differences among ethnic minorities in order to better understand their gender situations in education, employment, and health status, which are examined next. 2. Education Vietnam's socioeconomic development plans have placed great emphasis on improving school attendance, and the results are well reflected in recent data on current school enrollment as well as the level of schooling completed by the adult population. Rapid increases in school attendance have also been accompanied by remarkable reduction in the gender differential. However, wide disparities by region and ethnicity still persist in school attendance, along with inequalities by family income. The statistics presented below provide an overview of the current status of school attendance among Vietnamese from a gender perspective. Trends in Educational Attainment Increases in school enrollment have been accompanied by a rapid narrowing of the gender gap. By 2006, females aged 15-17 and 18-21 exceeded their male counterparts in the rate of school attendance. Yet, there still remain significant gender differences in major fields of study in higher education. Table 2-1 presents the level of schooling completed among adults aged 18-64 by gender and age. The table demonstrates rapid increases in educational attainment, especially at the secondary level among the young population, together with a swift narrowing of the gender gap. Among the population aged 35 and older, a little more than 20 percent of men and 20 percent or less of women had completed upper secondary schooling or beyond. The gender differential is also 13 larger for successively older age groups. By contrast, the proportion of young adults aged 18-24 who completed at least upper secondary schooling has more than doubled, reaching well over 40 percent, and many of these young people are still attending schools. The gender difference has also decreased greatly for younger adults, with women aged 18-21 now exceeding men of the same age in the completion of upper secondary schooling or more (48 percent vs. 44 percent). Rapid increases in school enrollment, especially among females, can be further illustrated in the school-age population, by comparing data in the 2004 and 2006 VHLSS. Table 2-2 shows the rate of current school attendance and school completion for each age group in the school-age population 6 through 24. For young children under age 14, more than 90 percent were already attending either primary or lower secondary schools by 2004. For 15-17 year olds, the proportion attending the upper secondary level has increased by more than 5 percentage points just in two years between 2004 and 2006, among males and females alike. Importantly, the proportion of 15- 17 year olds who are currently attending schools (either upper secondary or lower secondary levels) is higher for females (74 percent) than for males (70 percent). Among 18-21 and 22-24 year olds, the proportion attending higher education has also increased between 2004 and 2006. Females' attendance at higher education in the 18-21 age group exceeds males' attendance in 2006 (17 percent vs. 15 percent), while it is very much similar in the 22-24 age group when we combine both those who are currently attending and those who have already completed (18.4 percent for females vs. 18.8 percent for males). It is important to note, however, that about one- third of 22-24 year olds have not advanced beyond the primary level and this percentage is higher for females (35 percent) than for males (31 percent). Despite growing attendance at higher education by females as well as by males, there is a significant degree of gender segregation in their major fields of study. According to Figure 2-1, more than half of current female attendees are concentrated in two fields of study--education (25 percent) and business (30 percent). 10 Of male attendees nearly 40 percent are majoring in technology or engineering-related fields, whereas only 7 percent of female attendees are in these fields. This gendered pattern in the fields of study in higher education very much mirrors the pattern of gender segregation in occupations, as examined in the next section on job characteristics between women and men. Importantly, however, the current pattern of gender segregation in higher education represents some improvement from the past. When we compare major fields of study between current attendees and graduates aged 30-44 (see Table 2-3), the share of education majors among current female attendees is only about half the share among recent female graduates (25 percent vs. 51 percent); the share of business majors, on the other hand, has increased for current female attendees (30 percent vs. 23 percent). This suggests important changes in young Vietnamese women's outlook for their opportunities in a rapidly changing economy; yet, women's still limited representation in other fields such as computer, technology, or engineering-related fields suggests their potentially diverging career opportunities from men's. 10 Current attendees here include those attending junior colleges or universities. 14 Figure 2-1. Distribution of Major Fields of Study among Men and Women in Higher Education (Current Attendees) % 60 Male Female 50 40 30 20 10 0 g r n w er re s e e ss te r in tic io nc nc La tu th ne pu at is ee ie ie ul O si uc at om sc sc ric in Bu st Ed ng Ag C al oc h/ ur /e at /s at gy M es N lo iti an no ch um Te H School Attendance by Region, Ethnicity, and Family Characteristics While school-aged females are now equally or more likely than males to attend schools, there are large, persisting differences by area, region, ethnicity, household expenditure level, and parents' level of schooling. While gender differences in overall school attendance seem no longer apparent due to rapid increases in females' school enrollment, considerable disparities persist by region, ethnicity, and family backgrounds. Given that a large proportion of younger children (under age 14) now attend primary and lower secondary levels, these disparities are more visible among the 15-17 and 18-21 age groups. As shown in Table 2-4, among 15-17 year olds, the overall rate of school enrollment11 for females (76 percent) now exceeds the rate for males (72 percent), whereas in 2004 it was still lower for females than for males (69 percent for females and 73 percent for males in 2004; data not shown). The urban-rural difference persists, however. The school enrollment rate is much higher for urban males and females than for their rural counterparts, and the difference is by more than 10 percentage points (see also Figure 2-2). Importantly, even in rural areas the rate is now higher for females (73 percent) than for males (69 percent). In fact, the school enrollment rate for rural males of this age group indicates a slight decrease since 2004 (from 71 percent in 2004 to 69 percent in 2006). There are large differences by region as well: the rate is the highest in the Red River Delta (84 percent overall), while it is the lowest in the Mekong River Delta (56 percent overall). In all regions, the rate is higher for females than for males, except in the North West where the rate for females is much lower than for males--by 18 percentage points. The North West is also the region with the highest poverty rate. 11 School enrollment here refers to attendance at any level of schooling in the past 12 months, including vocational schools. 15 There are important differences by ethnicity as well. The Kinh/Chinese have a much higher rate of school enrollment compared with ethnic minorities overall (see Table 2-4 and Figure 2-2).12 Within each group, females show a higher rate than males, except for one minority group. Importantly, while the rate for ethnic minority girls aged 15-17 has increased compared with the rate observed in the 2004 VHLSS (from 61 percent to 64 percent), the rate for minority boys indicates a significant drop from 2004 (from 73 percent to 59 percent; data for 2004 not presented).13 Even among the ethnic minorities, however, the rate varies greatly. The Tay and others have a rate of school attendance that is close to the Kinh/Chinese, both for females and males. On the other hand, substantially lower rates are observed for the other three ethnic groups (please note that the sample size for each of these groups is relatively small.) Across different ethnic groups, the rate tends to be higher for females than for males, except for the Northern Mountain ethnic group where the opposite pattern is observed (the difference, however, is not statistical significant).14 Figure 2-2. Percent Attending School among 15-17 Years Olds Figure 2-2. Percent Attending School Among 15-17 Year Olds % Male Female 100 80 60 40 20 0 g ic s s s an se al c s g un ar ar ar ni ar in hn ur ne rb th ol ye ye ye ye /N R et U le ho hi ng + 4 8 11 /C n tr a sc 1- 5- 12 ai ou 9- nh nt en o i/M Ki N ou C ha M T N y/ Area Ethnicity Mother's schooling Ta In addition to regional and ethnic differences, school enrollment varies to an important degree by family's living standards or socioeconomic characteristics (see Table 2-4). The rate of school 12 Figure 2-2 does not present the rates for the Khmer/Cham group, since their sample size is quite small. 13 This decline among ethnic minority boys aged 15-17 is rather large even in view of the generally stalling trend among rural males. It is difficult to make conclusive inferences from these statistics, given small sample sizes for some ethnic groups in this age group. A similar trend, albeit small, is also observed for 11-14 year olds in rural areas, regardless of ethnicity; that is, the rate of school attendance for rural ethnic minority boys aged 11-14 has also declined slightly between 2004 and 2006 (from 89 percent to 86 percent), while the rate for ethnic minority girls has increased slightly (from 83 percent to 86 percent). 14 The Northern Mountain ethnic group and the North West region where a sizable proportion of the Northern Mountain ethnic group (about 32 percent) is located are the two notable groups where females still show a lower rate of school attendance than males in the 15-17 age group. In the North West region, even other ethnic groups like the Tay and others show a slightly lower rate of school attendance for females. The Northern Mountain ethnic group located in the other region--like the North East--also shows a lower rate for females than males. Please note that sample sizes across regions for these ethnic groups of this age range are rather small (between n=24 and n=49). 16 attendance increases as family's living standards go up. Nearly 90 percent of males and females aged 15-17 from the richest quintile are attending schools, whereas only about half of those from the poorest quintile do so. The rate is higher for females than for males across all expenditure quintiles including the poorest quintile. The level of schooling completed by mothers and fathers also shows a close association with the school enrollment rate among 15-17 year olds; the higher the level of mother's or father's schooling, the higher the rate of school enrollment for their children (see also Figure 2-2). Importantly, when the mother has no schooling, the rate of school enrollment is still lower for girls than for boys, suggesting that mothers' education can be crucial for their daughters' advancement in education.15 Table 2-5 shows the rate of school enrollment among 18-21 year olds by region, ethnicity, and family background characteristics. For this age group, the overall rate of school enrollment in 2006 is very much similar between males and females (41 percent for males and 42 percent for females), whereas there was a clear gender gap in the 2004 VHLSS (43 percent for males and 36 percent for females in 2004; data not presented). In fact, between 2004 and 2006 the rate has increased for females of this age group, while a slight drop has occurred for males, similar to the pattern observed among 15-17 and 11-14 year olds. Despite improved access to higher education for females, however, we still observe gender differences among certain groups when we examine by region, ethnicity, or family backgrounds. Rural females and males aged 18-21 are significantly less likely than their urban counterparts to be enrolled in school of any level, as expected. While urban females now exceed urban males in their school attendance, rural females' attendance is similar to rural males' (see also Figure 2-3). With respect to region, in certain regions like the Red River Delta, the North West, and the South Central Coast, females continue to lag behind males in their school enrollment, despite their relatively high rates in regions like the Red River Delta and the South Central Coast. The gender gap is particularly large in the North West (41 percent for males and 27 percent for females). By ethnicity, again, the Tay and others have a similar rate of school enrollment to the Kinh/Chinese among 18-21 year olds, while the other three ethnic groups have a substantially lower rate. The Northern Mountain ethnic group, again, shows a larger gender gap than the other ethnic groups (see Figure 2-3).16 In this age group, similar to the 15-17 age group, the rate of school enrollment is higher for those from the richest quintile, and the differential by expenditure quintile is greater than in the case of enrollment among 15-17 year olds. Those aged 18-21 from the richest quintile are nearly four times as likely as those from the poorest quintile to attend schools (63 percent vs. 17 percent); yet, there is no apparent gender differential in the poorest quintile. Parents' education levels are also closely associated with the rate of school attendance. Moreover, the gender gap is still apparent when mothers have low levels of schooling or when fathers have no schooling. When the mother has less than primary schooling, not only the rate of school attendance is very low among both females and males, but the rate is also much lower for females (see Figure 2-3 and Table 2-5). When the mother has more than 12 year of schooling, nearly three-quarters attend schools at ages 18-21, and the rate is higher for females. This again suggests that mothers' educational attainment would be critical for reducing the gender gap in education among children. 15 Since the level of mother's schooling completed (as well as father's) is also likely to be associated with other factors such as age, household expenditure per capita, ethnicity, and so on, any conclusive assessment on the causal impact of mother's schooling on girls' school attendance would require a multivariate analysis. 16 Figure 2-3 does not show the rates for the Khmer/Cham group, since their small sample size is quite small. 17 Figure 2-3. Percent Attending School Among 18-21 Year Olds % 100 80 Male Female 60 40 20 0 n al se g ic ic g ar s ar s ar s ar s ba ur ne un hn hn lin Ur R hi /N et et oo ye ye ye ye C ng in tra l sc h 1- 4 5- 8 11 12 + h/ ou ta 9- Kin i/ M un C en No o ha M y/T N Area Ta Ethnicity Mother's Schooling Attendance at Extra Classes More than one-third of primary school students, half of lower secondary, and two-thirds of upper secondary students attend extra classes. While the rate of attending extra classes is generally higher for females than for males, no apparent gender difference is observed in average household spending on these classes. Large differences in average spending on extra classes are found between urban and rural areas, and by family's economic status. There is a growing concern in Vietnam regarding extra classes being offered at school or outside of school, because household expenditures for these classes are growing and this can create increasing inequalities in educational attainment by family's economic status. While many may consider extra classes as an important way of improving academic performance and for successful entrance to higher education, there is a concern that attendance at these classes is not always voluntary. Many anecdotes suggest that extra classes provide an important avenue for school teachers to earn extra income, and students attend these classes for the fear of getting bad grades if not attending. Hence, these extra classes may be regarded by parents as necessary for getting ahead or at least not falling behind in school performance. We examine here whether girls are equally attending these classes as boys, or whether they also exceed boys, similar to what is observed in the pattern of school enrollment. Table 2-6 presents the rate of extra class attendance during the past 12 months among males and females who are enrolled in primary, lower secondary, or upper secondary schools. Approximately half of the students attend extra classes, and the rate is higher for females than for males (53 percent versus 49 percent; the difference is statistically significant). Yet, the rate of extra class attendance varies by school level: at the primary level about one-third of students attend these classes, whereas more than two-thirds at the upper secondary level attend extra classes. At all levels of schooling, the rate for females is higher than for males. According to detailed information regarding extra classes in the 2006 VHLSS, most of the students attending extra classes do so during the school year (about 90 percent), while a small proportion do so only during holidays. Nearly 95 percent of students also reported that these 18 classes are taught by teachers of their current schools; the classes are also carried out mostly at school (72 percent), while some reported the classes being held at teachers' houses (27 percent). This seems to support the view that these extra classes have become part of informal school curriculum which may be imposed on students by teachers at school. Similar to the case of regular school attendance, attendance at extra classes also varies by region, ethnicity, and family's socioeconomic characteristics. The rate of attendance is much higher in urban areas than in rural areas (70 percent versus 46 percent overall), and the rate is higher for females in both areas. These classes are especially prevalent in the Red River Delta with nearly 80 percent of students attending them, followed by the South East (59 percent). In these two regions, there is little difference by gender. While extra classes appear to be quite prevalent, the rate of attendance is quite low among ethnic minorities overall (15 percent). Expectedly, the rate of extra class attendance is closely associated with family's expenditure quintiles. About one-quarter of students at the poorest quintile attend extra classes, whereas nearly three-quarters of students at the richest quintile do so. Importantly, the rate is higher for females across all quintiles (except for the 3rd quintile). Likewise, the rate varies greatly by mother's or father's level of schooling; and it is generally higher for females across all levels of parents' schooling, including those parents with no schooling. Although it is hard to tell from the data what motivates parents to send their daughters to extra classes somewhat more than their sons, this pattern is quite consistent with what we observed in the pattern of school enrollment where females are now exceeding males at the upper secondary level and reaching parity at higher education. While females are more likely than males to attend extra classes at all three levels of schooling, the average expenditure reported for these classes (per student) is similar between females and males. 17 At lower levels of schooling, the amount seems higher for males, but there is no significant difference by gender. The average expenditure on extra classes is 422,000 VND per year across all school levels. Yet, the amount is much larger for those at higher levels of schooling; the average amount at the upper secondary level (588,000 VND) is nearly twice as large as the amount at the primary level (297,000 VND). The amount also varies widely by area and by family's expenditure quintile. The average amount reported in urban areas (797,000 VND) is more than three times higher than the amount in rural areas (262,000 VND). The difference is even larger by expenditure quintile: the average amount reported for students from the richest quintile (1,055,000 VND) is nearly three times as high as the amount for students from the second richest quintile (371,000 VND), and nearly eight times higher than the amount reported for those from the poorest quintile (131,000 VND).18 The average amount spent on females' extra classes seems slightly higher than the amount on males' across all quintiles except for the poorest quintile, but the gender difference is not statistically significant. In summary, the examination of school attendance as well as attendance at extra classes demonstrates that great progress has been made in closing the gender gap in school attendance among Vietnamese. For the school-age population, especially 15-17 and 18-21 year olds, females now exceed males in the rate of school attendance. This suggests that parents are now supporting 17 Average (mean) expenditures for extra classes are calculated only among students who attended extra classes, excluding zero values. 18 The large difference by expenditure quintile in average spending on extra classes comes in part from differences in the level of schooling across quintiles among students who attend classes. Yet, even when we examine the average expenditure controlling for students' level of schooling, the pattern of differences by expenditure quintile is similar. 19 and investing in girls' education as much as boys' education. Yet, at the level of higher education, there exists visible gender differences in the fields of study which can lead to diverging opportunities between men and women in the labor market. Moreover, wide disparities observed in school attendance by area, region, ethnicity, and family's socioeconomic background raise a concern for increasing inequalities in education. The consideration of extra class attendance and expenditure on these classes also raises a concern that these disparities may become even wider in the future, to the extent that attendance at these classes influences children's advancement to the next higher level of schooling. It will be important in future research to closely examine how disparities in educational attainment play out for individuals' economic opportunities in the labor market. 3. Employment and Earnings This section examines various characteristics of women's and men's work, in order to understand the nature and the degree of gender inequality in the Vietnam's current labor market. The section presents results from descriptive analyses regarding labor force participation rates, hours and weeks worked on income generating activities, involvement with housework, wage employment versus self-employment, industry and occupational characteristics of the main job, and earnings in wage employment. Labor Force Participation Rates Between 2004 and 2006, the overall rate of labor force participation has changed little, but the participation rate has declined substantially among 15-17 year olds, especially among rural females. In the prime working-age population (age 25-64), women are less likely than men to participate in the labor force, particularly in urban areas. The gender difference is smaller in rural areas, and the participation rate is higher in rural than in urban areas. Table 3-1 presents labor force participation rates by gender, age, and area, using both 2006 VHLSS and 2004 VHLSS data. The labor force participation rate here refers to the rate of being employed at any time during the past 12-month period.19 Between 2004 and 2006, the overall rate of labor force participation has changed little when we consider all population aged 6 and older. Yet, the participation rate among the school-age population, especially among 15-17 year olds, has declined substantially between the two survey years. While this decline took place in both urban and rural areas, the decline was greater in rural areas and particularly among rural females. In 2004, nearly half of rural females aged 15-17 (48 percent) were engaged in some type of work, whereas only one-third of them (33 percent) were in 2006. In 2006, rural females in this age group were also less likely than rural males (38 percent) to be in the labor force, reversing the pattern in 2004. But their rates are still more than twice as high as the rates among their urban counterparts (15 percent for urban males and 14 percent for urban females aged 15-17). This declining labor force participation among school-aged males and females in rural as well as urban areas very much corresponds to increases in their school enrollment, as discussed above. 19 While labor statistics usually include the unemployed or those looking for a job in the rate of labor force participation, the rate here does not include those looking for a job which is very small for the whole 12-month period. In this report, we use the labor force participation rate interchangeably with the employment rate. 20 Similar to 2004, the labor force participation rate in 2006 is higher in rural areas than urban areas for all age groups, and for males and females alike (see Figure 3-1). The large difference between urban and rural areas is noticeable especially in younger age groups (ages under 25) and older age groups (ages above 54). This indicates that young people are still important part of the workforce in rural areas despite increases in their school enrollment. The predominantly agricultural nature of work in rural areas also involves many older people to continue to work at older ages. Figure 3-1. Labor Force Participation Rates in Past 12 Months Urban male Urban female Rural male Rural female % 100 80 60 40 20 0 6-10 11-14 15-17 18-24 25-34 35-44 45-54 55-64 65+ Age Females are overall less likely than males to participate in the labor force in both urban and rural areas, for all age groups. Yet, the gender difference is greater in urban areas than rural areas. In the prime working-age population (ages 25-64), 90 percent of men in urban areas have worked over the 12-month period, compared with 80 percent of women. In rural areas, the gender difference is much smaller. The largest gender difference in urban areas is shown in the 55-64 age group, which is likely due to the earlier retirement age for women (age 55) than for men (age 60) in the formal sector. Accordingly, among women aged 55-64, a large rural-urban difference is also observed: nearly 80 percent of rural women aged 55-64 have worked over the 12-month period, compared with only 48 percent of urban women in the same age group.20 Number of Weeks Worked on Income Generating Activities Among people who have worked in the past year, the average number of weeks worked is very similar between women and men overall--about 45 weeks in urban areas and 34 weeks in rural areas. Only women aged 25-34 who were in the prime years of childbearing and childrearing were less likely than men to work--by 2 weeks--both in urban and rural areas. 20 There are also some regional differences among women aged 55-64. The rate of labor force participation is especially lower for women in the South East and the Mekong River Delta. Restricting to rural areas, the labor force participation rate for women aged 55-64 in the South East is 66 percent, and for those in the Mekong River Delta is 69 percent. The rate for urban women aged 55-64 is the lowest in the South East at 32 percent, followed by those in the Red River Delta at 43 percent. (The rate for urban women in the Mekong River Delta is 59 percent, but the sample size is small n=71) 21 Table 3-2 presents the average (mean) number of weeks worked during the 12-month period by gender, age, and area, among people who reported doing any work.21 The table also includes the percentage of those who did not participate in any income-generating activities during the year. While the overall rate of labor force participation is higher in rural areas than urban areas for both males and females, the average number of weeks worked over the past year is much higher in urban areas (by more than 10 weeks), mostly likely due to the seasonal nature of agricultural work in rural areas. The average number of weeks worked among urban males and females is about 46 weeks, compared with 35-36 weeks for rural males and females. In both areas, the number of weeks worked has increased by about a week since 2004. Despite the lower rate of labor force participation for females than for males, especially in urban areas, the average number of weeks worked is very much similar between urban females and males who have worked (46 weeks for both; see Table 3-2). This gender similarity is true for all age groups in urban areas, except for those aged 25-34 where women worked on average two weeks less than men over the year. This is likely due to that women aged 25-34 are in the prime years of childbearing and childrearing. For all other age groups in urban areas, women work as much as men throughout the year, once they are engaged in any income generating work. In rural areas, women's number of weeks worked is about one week less than men's, for most age groups. Housework Despite a high rate of women's labor force participation, housework remains mostly women's work. One-third to nearly half of urban men (age 25-64) report not doing any housework. Rural men are slightly more likely than urban men to do housework. While women in Vietnam exhibit a very high rate of labor force participation and most women work as much as men throughout the year, data on the division of household labor suggest that housework still remains mainly women's responsibility. Table 3-3 shows the percentage of people who did not do any housework and the average number of hours spent on housework per day among those who did any housework.22 Overall, close to half of all males (age 6 and above) report not doing any housework at home, compared with only one in five females. Across all ages including young children--regardless of urban or rural areas--females are more likely to do housework than males. When we consider adults aged 25 and older only, the burden of housework on women becomes more apparent (see also Figure 3-2). For women aged 25-64, less than 10 percent report not doing any housework with little difference by area or age. For men aged 25-64, on the other hand, urban men are substantially less likely than rural men to be engaged in housework, although the percentage of men not doing any housework decreases somewhat over ages up to age 55-64. The largest gender gap in housework can be seen among those aged 25-34 where nearly half of urban men and more than one-third of rural men report not doing any housework. Given the labor force participation rate of more than 85 percent for urban women and 95 percent for rural women in this age group, 21 The average number of weeks worked is calculated based on the total number of hours worked during the year, assuming 8 hours of work per day and 6 days of work per week. The number of weeks presented in Table 3-2 is calculated only among those who did any work, excluding zero values. 22 The analysis of housework is based on two simple questions regarding housework in the 2006 VHLSS. One asks whether the person did any housework that includes cleaning, shopping, cooking, washing, water and wood fetching, repair work in the house, etc. The other asks about the average number of hours spent per day on housework. 22 this suggests that urban married women in particular bear double burden of doing both housework and income generating activities due to their husbands' lack of involvement.23 Figure 3-2. Percent People reporting Not Doing Any Housework by Gender and Area Urban Male Urban Female Rural Male Rural Female 60 % 50 40 30 20 10 0 25-34 35-44 45-54 55-64 65+ Age Even among males who do any housework, their average number of hours spent on housework is only 1.5 hours per day, compared with 2.2 hours for females overall (see Table 3-3). There is little difference in average hours spent on housework between urban and rural areas. Moreover, the average number of hours spent on housework in 2006 indicates virtually no change from 2004, for both males and females. Types of Employment between Women and Men Wage employment has become quite common in urban areas, with 61 percent of urban men and 48 percent of urban women holding their main job in wage employment over the 12- month period. For urban women, nonagricultural self-employment is also common. In rural areas, agricultural work is still dominant, but rural men are more likely than rural women to experience wage employment. With rapid economic development, wage jobs in Vietnam have been increasing both in urban and rural areas. The comparison of data from the 2004 and 2006 VHLSS shows a small increase in wage jobs during the two-year period. Yet, men are still more likely than women to be engaged in wage employment, and there are large differences in the type of employment between urban and rural areas. Table 3-4 presents the percentage of men and women aged 18-64 who were engaged in wage employment, self-employment, or both during the previous 12-month period. The table also presents the data from the 2004 VHLSS. For the country as a whole, self-employment--which includes both agricultural and nonagricultural work--still remains a major type of work for adults 23 The large gender gap in housework is very much similar even when we confine our examination only to married men and women, who make up nearly 70 percent of men and 80 percent of women in the 25-34 age group. 23 aged 18-64 in 2006, although wage work has been increasing. When we look at urban and rural areas separately, however, wage employment has emerged as the major type of work in urban areas, especially for men. More than half of urban men (54 percent) in 2006 were engaged solely in wage employment over the 12-month period, and 42 percent of urban women were as well. For urban women, the percentage engaged only in self-employment (50 percent) is still higher than the percentage in wage employment. Between 2004 and 2006, the percentage of women and men solely engaged in wage employment over the 12-month period has increased slightly in both urban and rural areas, while the percentage of those who combined wage work and self-employment has decreased. Although the change is small, this suggests the growing importance of wage employment. In rural areas, however, wage employment is still very limited and it tends to be combined with other self- employment work: only 17 percent of rural men and 12 percent of rural women were involved solely in wage work. Rural women are also less likely than rural men to be engaged in any wage employment. Table 3-5 examines more closely the type of employment among adult men and women (age 18- 64), focusing on the main job held over the 12-month period and distinguishing self-employment into agricultural and nonagricultural work. The table also examines how working in different types of employment is associated with age, marital status and education. While a relatively similar percentage of women and men participate in the labor force in Vietnam, women and men are engaged in very much different types of employment. Overall, 42 percent of adult men have their main job in the wage sector, whereas only 27 percent of adult women do. There are even greater differences between urban and rural areas (see also Figure 3-3). In urban areas, wage employment is now the predominant type of work for both men (61 percent) and women (48 percent), although the percentage is substantially higher for men than for women. For urban women, nonagricultural self-employment is also an important type of work (41 percent), compared with urban men (30 percent). In rural areas, agricultural self-employment is still the dominant type of work for men (49 percent) and women (60 percent). After agricultural work, wage employment is the next common type of work for rural women as well as rural men, but men are substantially more likely than women to have wage work (35 percent of rural men versus 20 percent of rural women). Only a small percentage of rural men and women are engaged in nonagricultural self-employment (20 percent or less). Similar to urban-rural differences, regions which include large urban areas show higher percentages of men and women in wage employment. The South East has the highest percentage of men (56 percent) and women (48 percent) in wage employment, followed by the Red River Delta and the South Central Coast. These regions, along with the Mekong River Delta, also have a relatively higher percentage of men and women in nonagricultural self-employment than other regions. The North West, on the other hand, has the lower percentage of men and women engaged in wage employment but the highest percentage in agricultural self-employment. Across all regions, women are less likely to be in wage employment than men. 24 Figure 3-3. Type of Employment for Main Job by Gender and Area (Age 18-64) Wage employment Nonagricultural self emp Agricultural self emp % 100 80 60 40 20 0 Male Female Male Female Urban Area Rural Area The type of employment also tends to differ greatly by ethnicity, given a large concentration of ethnic minority households in rural areas. More than 70 percent of ethnic minority men overall and more than 80 percent of ethnic minority women are still confined to agricultural self- employment; only a small proportion is engaged in wage employment. Yet, there are considerable differences among ethnic minorities themselves. The Khmer/Cham have a higher percentage of men and women in wage employment, which is similar to the percentage of the Kinh/Chinese. The Khmer/Cham women are even more likely than the Kinh/Chinese to be in wage employment, although their number is relatively small. Compared with other ethnic groups, the vast majority of the Northern Mountain ethnic men and women are still engaged in agricultural work (89 percent of men and 95 percent of women). For ethnic minorities, the proportion engaged in nonagricultural self-employment is even smaller than the proportion in wage employment. The next two panels of Table 3-5 present the distribution of types of employment among men and women by age, marital status, and educational attainment, separately for urban and rural areas. Although wage employment is less common in rural areas, the association between wage employment and selected characteristics are quite similar between the two areas. Wage employment is closely associated with age: younger women and men are more likely to have wage employment than older people, while older women and men more likely to have agricultural self-employment and nonagricultural self-employment to some extent. Accordingly, unmarried men and women--who tend to be younger--or divorced/separated men and women are more likely to have wage employment than the married or widows. Widows or widowers in rural areas--who tend to be older--are most likely to have agricultural work compared with other marital status groups, while widows in urban areas are most likely to be engaged in nonagricultural self-employment. 25 The level of schooling also shows a close association with the likelihood of wage employment.24 The higher the level of schooling completed, the more likely one is to be working in wage employment. The majority of those who completed upper secondary or higher education have wage employment (especially in urban areas), and approximately 90 percent of women and men who completed higher education all have wage employment both in urban and rural areas. At lower levels of education, wage employment is still a lot more common for men than for women in both areas, but this gender difference in wage employment becomes small at higher levels of education. This suggests that for the kinds of wage jobs available in the current labor market, those held by women are more limited and education is likely to play a more important role for women's access to wage job than for men's. Next, we examine what other kinds of gender differences exist within wage employment as well as in other types of employment, by focusing on occupation and industry characteristics of women's and men's jobs. Gender Segregation in Industry and Occupation Gender segregation is prevalent in both wage employment and nonagricultural self- employment, in terms of the type of industry or occupation where men and women work. Women are particularly concentrated in retail sales, textile/garment manufacturing, education/health/cultural services, and hotels/restaurants services industries. As for occupation, women are over-represented in education professional, sales, and unskilled manual occupations. Figure 3-4 presents the industry distribution of main jobs held by men and women (ages 18-64), combining urban and rural areas. The graph demonstrates that apart from agricultural industry, men and women are distributed across very different industries. Except for half of women represented in agricultural industry, women's work is mostly concentrated in retail sales, textile/garment manufacturing, food/beverage manufacturing, hotels/ restaurants services, and education/health/cultural services industries. For men, except for agricultural industries, their work is concentrated in construction/utilities, other material production/processing,25 retail sales, transportation/communication services, and business/financial services industries. For each category of industry, the gender gap is evident indicating a considerable degree of segregation between men's and women's jobs. Table 3-6 shows more details about the industry distribution of women's and men's jobs, separately for urban and rural areas. In rural areas, other than agricultural industry, secondary industries, especially manufacturing, have emerged as an important source of work, although men are more likely than women to have jobs in these industries (23 percent of men and 14 percent of women). In urban areas, the tertiary sector--consisting of sales and services industries--is the main sector for both women's and men's work, but only a small number of industries dominate women's work. Nearly half of urban women (46 percent) are concentrated in retail sales, hotels/restaurants, and education/health/cultural services, whereas men are a little more distributed across different industries with dominant industries being transportation/communications services and construction/utilities. 24 Here we focus mostly on bivariate relationships. Multivariate analysis is required in order to parcel out interrelationships among associated variables. For example, the close association observed between education and wage employment is also related to the fact that the younger people have a higher level of education than older people and they are also more likely to have wage employment. 25 "Other material production/processing" indicated in Figure 3-4 and Table 3-6 include industries producing materials or products such crude oil, chemical products, metal or non-metal products, machinery, computer products, electronic items, medical equipments, motor vehicles/spare parts, etc. 26 Figure 3-4. Industry Distribution of Main Jobs by Gender (Age 18-64) % 60 50 Men Women 40 30 20 10 0 cu y es g s ni eal inan unic nts re on on c n & ultu ser ns se es es du g ut g r r s l s es ce g ai les en pai ua str ie O d/p me ufa inin C cti nufa ctio pr r m pro rin n ru /pro urin ltu al a l vi c ic tio , c cia atio p e r a vi c n ssi ilit f m ura R t sa Aq ore ag e er ape ts ctu ls rv oo ar an M d e/r t on er io ss com sta /f & nc et re l /re tu le na io u a ls ul m sa te ct n Ed Bu tio ote an ric le in s/ e Ag ho ma ile g st ta H ne & ta th xt era od & si n W g W e s/ n T e ev Sa , h l sa /b or th n od e sp cl Fo at hi an uc Ve Tr When we examine the industry distribution of men's and women's jobs, separately for wage employment and nonagricultural employment, it becomes clear that not only are women's and men's work separated from each other, but women's work is very confined only to a few industries within each type of employment (see Table 3-6). As for wage employment in urban areas, more than half of women are in service industries overall, and are especially concentrated in education/health/cultural services and business/financial services. Following service industries, urban women's wage jobs are concentrated in textile/garments manufacturing industry.26 Men's wage jobs in urban areas, on other hand, are more evenly divided between service and manufacturing industries, and encompass a more number of detailed industries. For wage employment in rural areas, the patterns are similar, but secondary industry jobs rather than service industry jobs are more common for both men and women. With respect to nonagricultural self-employment, we can observe further concentration of women's jobs in a few industries. Nearly 70 percent of urban women's and 60 percent of rural women's nonagricultural self-employment is concentrated either in retail sales or hotels/restaurants services. Retail sales in particular account for nearly half of women's work in nonagricultural self-employment in both urban and rural areas. For men's nonagricultural self- employment, retails sales, transportation/communication, and other production/processing manufacturing industries are major industries, accounting for about half of men's work in both urban and rural areas. 26 Compared with recent results from the Enterprise Survey conducted by the General Statistical Office, wage workers in the 2006 VHLSS, especially female workers, are under-represented in the FDI sector for some reason. Data from both the 2006 VHLSS and the 2006 Enterprise Survey indicate that nearly two-thirds of female workers (about 64 percent) in the FDI sector (urban and rural areas combined) are in textiles/garments manufacturing industry. This means that a high percentage of female wage workers in textiles/garments manufacturing industry shown in Table 3-6 is likely to be even higher, if under-represented FDI sector wage workers are taken into consideration. 27 In addition to the type of industry, the characteristics of women's and men's nonagricultural self- employment differ in other aspects as well. According to data collected at the household level on nonagricultural business activities (see Table 3-7),27 women-operated businesses are smaller in scale, with a fewer number of laborers involved: two-thirds of urban women's and nearly three- quarters of rural women's businesses are done by women themselves only. Men's nonagricultural businesses are also largely operated by themselves, but men's activities include a higher percentage of businesses employing 4 or more laborers and also a higher percentage of those with paid laborers. Men's businesses are also more likely to have a business or trading license than women's, but those enterprises in rural areas are overall substantially less likely to have a business license. Even in urban areas, only one-third of nonagricultural businesses overall have a business license. In addition, men's businesses are more likely than women's to be conducted in permanent shops or other non-permanent places, although nearly half of urban men's or more than half of rural men's nonagricultural businesses are done at home. Women's nonagricultural businesses are mostly done either at home (more than half) or in markets (more than 20 percent). Given differences in the type of industry and also in the scale, the average monthly revenue for men's businesses in urban areas (7,948,800 VND) is nearly twice as high as that for women's (4,068,200 VND). The average revenue is much lower in rural areas, but men's businesses (4,285,600 VND) again have much higher revenues than women's (1,688,350 VND). The median amounts are much lower than these average amounts, but the gender difference still remains large. As in gender differences in the type of industry, the distribution of occupations for the main job also differs considerably between men and women. As shown in Table 3-8, other than agricultural work, large numbers of men and women (age 18-64) in both urban and rural areas are either skilled or unskilled manual workers. Yet, men are more likely to hold skilled jobs whereas women are more likely to hold unskilled jobs. In urban areas, professional occupations are another important type of work for both men and women; in fact, for urban women, professional jobs are more common (21 percent) than skilled manual jobs (15 percent). Looking at men and women in wage employment only, in urban areas a higher percentage of women (44 percent) than men (33 percent) hold professional occupations. This difference in large part comes from that women are over-represented in education-related jobs (16 percent of women compared with 5 percent of men). In rural areas, professional jobs are fewer, but again a higher percentage of women than men are represented in education-related professional jobs (16 percent versus 4 percent). Importantly, although the proportion of those in administrative/managerial positions is small, men are more likely than women to hold these positions in urban as well as in rural areas. The gender composition in each occupation (Figure 3-5) demonstrates that education-related professional jobs and sales jobs are the two most predominantly-female occupations.28 Women make up more than 70 percent of the workforce in these occupations. Women also make up more than half of the workforce in unskilled manual occupations; agricultural work is more or less 27 The characteristics of nonagricultural businesses described in Table 3-7 are based on data in the 2006 VHLSS collected for all nonagricultural businesses at the household level. Since some households listed more than one nonagricultural business activity, the total sample size in Table 3-7 indicates the number of business activities (enterprises) rather than the number of individuals or households (about 20 percent of households listed more than one business activity). Female-operated and male-operated enterprises are distinguished based on the question asking about the person who manages or controls each of business activities in the household. 28 The percentage of female in each occupation is calculated for all, without restricting to any age group. 28 evenly divided between men and women. Women are least represented in administrative/ managerial occupations, accounting for just over 20 percent. In short, women's wage jobs are still very much confined either to professional occupations which usually require higher levels of schooling or to unskilled manual occupations. Men's wage jobs, on the other hand, are somewhat more widely distributed across a range of occupations including various types of skilled jobs and across different industries. The limited aspect of women's wage employment can also be seen in the level of schooling completed between men and women who hold wage jobs. Nearly half of women in wage employment have either upper secondary schooling or college education, compared with only 36 percent of men; the proportion with a college education is also higher for women than for men (data not shown). In other words, higher levels of schooling seem more crucial for women's entry into wage employment, since their wage work is still concentrated either in a few professional occupations or in unskilled jobs. Importantly, gender segregation in industry and occupation examined thus far very much mirrors gender differences in major fields of study observed among college graduates and current college attendees. It will be important for future research to explore the sources of gender segregation; in particular, whether there are certain institutional barriers for women's entry into certain occupations or industries, or whether gender segregation results more from gender-differentiated socialization or skills training. Figure 3-5. The Degree of Female Concentration by Occupation % Female in each occupation 100 % 80 60 40 20 0 es s re al l l l al al ua ua ia le on on on tu ic er Sa an an ul rv si ag si si m m ric Se es es es an Ag d d of of of ille lle m pr pr pr e/ ki Sk er n al tiv ns io ic th ra U at ed O t uc is m in Ed h/ m ec Ad i/t Sc Gender Gap in Earnings in Wage Employment For women in wage employment, their hourly wage is about 87 percent of the hourly wage received by men in urban areas, and 88 percent in rural areas. The gender gap ratio varies widely across sectors of employment, industries, and occupations. With respect to education, women and men with higher levels of education exhibit a smaller gender gap in wages, while it tends to be larger for those with lower levels of education. 29 While a more select group of women than men tends to work in wage jobs, women's earnings still remain lower than men's. Yet, the earnings gap between men and women varies widely by sector of employment, occupation, and industry. Table 3-9 presents mean hourly wages of men and women in their main job,29 by sector of wage employment, occupation, industry, and the level of schooling completed, separately for urban and rural areas. The table also includes the distribution of each of these characteristics for men and women. In urban areas, women's mean (real) hourly wage (8,206 VND for all aged 15 and older) is about 87 percent of the hourly wage received by men (9,467 VND). This gender gap represents some improvement from 2004 when the ratio was 83 percent in urban areas. In rural areas, the ratio is slightly higher at 88 percent and this represents little change from 2004 (88.5 percent). While the gender gap in earnings appears to be similar between urban and rural areas in terms of its ratio, there is a wide disparity in the level of wages between the two areas. Rural women's and men's average wages are only 63-64 percent (5,218 VND for women and 5,953 VND for men aged 15 and older) of what is received by their urban counterparts. The level of wages and the degree of the gender wage gap also vary across sectors of employment, occupations, and industries. The distribution of employment sectors for the main job shows that wage employment in rural areas mostly consists of private-sector jobs (63 for women and 76 percent for men), followed by government-sector jobs (19 percent for women and 13 percent for men). In urban areas, the share of private-sector jobs is somewhat smaller (46 for urban women and 54 percent for urban men) than in rural areas, while the shares of government sector or state-owned enterprise (SOE) jobs are larger. In both areas, women are more likely than men to have their employment in the government or foreign direct invested (FDI) sectors, while men are more likely than women to have their employment in the private sector. The share of SOE employment is similar between women and men. In urban areas, hourly wages are highest in the government and FDI sectors for both women and men, followed by the SOE sector; the wages are lowest in the private sector (see Table 3-9). The gender gap is the smallest in the private sector (88 percent) where wage levels are relatively low, whereas the gender gap tends to be larger in the other sectors (80-82 percent). In rural areas, patterns of the gender wage gap tend to diverge: the government sector shows the highest level of wages with women's wages exceeding men's, while the private sector shows the lowest level of wages with one of the larger gender gap (the gender gap is the largest in the FDI sector in rural areas). The larger gender gap in the rural private sector comes from that rural women in this sector have very low wages, especially those women working in smaller agricultural enterprises. Across occupations, we observe more complicated patterns of wage levels and gender differentials, particularly between urban and rural areas. In urban areas, a relatively large number of men and women in professional occupations--those in science/technology/ medical and education-related jobs--enjoy higher levels of wages and the gender gap also tends to be smaller in these occupations. In rural areas, male professionals in these jobs also have higher wages, but females' wages are much lower, leading to a relatively larger gender gap in wages. Part of the reason lies in that female professionals in rural areas are more likely to be concentrated in mid- 29 The hourly wage here includes the main salary and other benefits. Hourly wages are obtained by dividing total wages by the number of hours worked on the main job. 30 level health or education jobs. Yet, it is also the case even in urban areas: female professionals are more likely to be found in mid-level positions while male professionals are found more in high-level positions. A fuller explanation for the diverging pattern of gender wage differentials between urban and rural areas will require a close examination of various factors that determine wage levels in each area, which is beyond the scope of the current descriptive study. As for skilled manual occupations where large numbers of men and women in both urban and rural areas work, the gender wage gap is relatively larger than the average gap, probably due to differences in the type of industry that men and women work. The gender wage gap among unskilled workers in urban areas tends to be smaller, but mainly because of very low levels of wages for both men and women in this occupation. As seen in Table 3-9, the level of wages is also closely associated with the level of schooling completed. In particular, those with some college education or more enjoy much higher levels of wages in both areas, compared with those with lower levels of schooling. Importantly, women and men with higher levels of education exhibit a smaller gender gap in wages, especially in urban areas. The lower the level of schooling completed, the gender gap tends to be larger. This suggests that a relatively high proportion of well-educated women in wage employment and their high representation in a few professional occupations is one important factor leading to a relatively smaller gender gap in the current labor market. As more and more wage jobs are created in different sectors that require people with different levels of schooling, it will be important to observe how the gender gap changes in the future. 4. Health Status and Access to Health Care Improving general health of the population should be an integral part of any country's development efforts, and Vietnam's development process has also reflected this effort. Along with rapid economic development in Vietnam, total fertility rates have fallen, children's mortality rates have declined, awareness of reproductive health has increased, and life expectancy has improved. Yet, much more improvement is desired in various areas of health in Vietnam. In particular, great policy attention is called for in providing equitable access to and utilization of health care services, as well as improving the provision of quality and affordable health services. Improving general health and access to health services is essential for improving living standards of individuals and the family, since health problems of one's own or of family members can seriously constrain one's productive capacities and economic opportunities, as well as generate financial burdens on the family. Women usually experience more acute or chronic health problems over the life cycle than men, but as mothers and caretakers of the family, women often face disadvantages in their access to or utilization of health care services. This section provides a general overview of health status reported by respondents in the 2006 VHLSS and their access to health care services. In particular, we focus on whether there are significant gender differences in reported illness and in access to health care services among people reporting illness. In addition to gender differences, we also examine how health status and access to health care vary by other characteristics such as age, ethnicity, region, family living standards, and health insurance status. Examining health status and utilization of health care services from a gender perspective is a big task of its own, requiring a comprehensive analysis of various issues involved in children's and adults' health, heath care seeking behaviors, public and private expenditures on health care, availability of health care services at local and national levels, provision of health insurance, and so on. This report provides only a limited examination of health status and access to health care using selected measures from the 2006 VHLSS. A fuller 31 examination of health data available in the VHLSS will be necessary for a better understanding of health issues from a gender perspective. Gender and Indicators of Health Except for children under age 15, women are more likely than men to report illness across age groups. Yet, among people reporting serious illness, the number of days missing normal activities or being bedridden due to illness is higher for men than for women. Table 4-1 presents four indicators of illness reported by respondents of all ages, and how each indicator of illness varies by gender, age, area, ethnicity, and socioeconomic characteristics of respondents.30 The four indicators of illness presented in this table are (a) illness (or injury) in the past 4 weeks; (b) illness in the past 12 months; (c) whether respondents were unable to participate in normal activities due to illness (e.g., being absent from school or work); and (d) whether respondents were bedridden due to illness and needed care assistance. Here, we consider (a) and (b) as general indicators of illness, and (c) and (d) as indicators of more serious illness. In all four indicators presented, females are overall more likely than males to report illness (see also Figure 4-1). As for illness in the past 12 months, for instance, more than half of females (56 percent) reported illness compared with 49 percent of males. With limited information available in the VHLSS regarding specific causes or types of illness, we cannot analyze more detailed aspects of illness reported by respondents. It is important to note that self-reported health status or illness tend to involve subjective evaluation of one's own health and that people who are better informed with health issues or have better access to health care facilities may be more perceptive of one's health conditions or illness. Therefore, responses regarding having been absent from normal activities due to illness or having been bedridden are often regarded as better indicators of health status, and they are also likely to capture incidences of serious illness. For these two indicators, females are still somewhat more likely than males to report illness. As for being absent from normal activities due to illness, 33 percent of females reported illness compared with 29 percent of males (the difference is statistically significant). The proportion reporting illness varies by age, consistent with a generally U-shaped pattern between age and health status (see Table 4-1; see also Figure 4-2 for the proportion of males and female reporting absence from normal activities due to illness). Since children are often ill, the proportion reporting illness is quite high for young children. It then decreases gradually up to ages 15-19 or 20-29, and starts increasing again for older groups with the highest proportion of being ill shown for those aged 60 and older. Except for children under age 15, females at all ages are more likely than males to report illness in all four indicators of illness examined. The gender difference, however, is relatively smaller for those aged 60 and older, particularly in the two indicators of serious illness (being absent from normal activities and being bedridden due illness). 30 Health questions were asked to all household members, while parents responded for children under age 15. In some households, the respondent answering questions for all household members may have reported regarding health as well. 32 Figure 4-1. Indicators of Health Status by Gender % 70 60 Male Female 50 40 30 20 10 0 Ill in past 4 weeks Ill in past 12 months Ill & unable to do Ill & bedridden with normal activites assistance Figure 4-2. Percent Reporting Absence from Normal Activities Due to Illness by Gender and Age (in Past 12 Months) Male Female % 70 60 50 40 30 20 10 0 0-4 5-9 10-14 15-19 20-29 30-39 40-49 50-59 60+ Age group Reported health status also varies by area and region, but the pattern is somewhat different depending on which illness indicator we consider (see Table 4-1). For questions regarding general illness or injury during the past 4 weeks or 12 months, the proportion reporting illness is higher in urban than rural areas, similarly for males and females. It also tends to be higher in richer regions like the South East. When we consider more serious illness with indicators of being absence from normal activities or being bedridden due to illness, however, the proportion reporting illness is higher in rural areas than in urban areas, again similarly for males and females.31 By region, it is highest in the Central Highlands.32 Across all regions, females are more likely than males to report general or serious illness. 31 The results on rural-urban differences for these two indicators are in accordance with results from other health studies in Vietnam such as the 2001-2002 Vietnam National Health Survey (VNHS) which includes more detailed assessments of health status among Vietnamese. Yet, results on regional differences are not so consistent between the 2001-2002 VNHS and the 2006 VHLSS. 33 With respect to ethnicity, the Kinh/Chinese are more likely than ethnic minorities to report general illness, but ethnic minorities are slightly more likely to report serious illness when both indicators of being absent from normal activities or being bedridden due to illness are considered (the difference is not statistically significant for the latter two indicators though; see Table 4-1). Similar to the Kinh/Chinese, ethnic minority females are in general more likely than males to report serious illness. Yet, there are large differences across different ethnic groups: the Central ethnic groups are most likely to report serious illness or general illness (except for being ill in past 4 weeks among females), and the gender difference in the Central ethnic group is quite small. On the other hand, the Khmer and Cham are least likely to report serious illness and their proportions reporting serious illness are even smaller than those of the Kinh/Chinese. The Northern Mountain ethnic groups also show lower proportions reporting serious illness, compared with the Kinh/Chinese as well as with the other two ethnic groups. As for expenditure quintiles, since age is closely associated with reported illness and age structure may vary across quintiles, Table 4-1 shows the proportion reporting illness in each quintile which is standardized by age structure in order to eliminate the influence of age. Across different indicators of illness, we observe somewhat varying patterns of association with expenditure quintiles. The proportion reporting general illness increases for the higher expenditure quintiles, similarly for males and females. For the indicators of serious illness, however, the association is not clear-cut. Both males and females from the richest quintile are less likely than others to report being absent from normal activities due to illness, but there is not much difference across other quintiles. Within each quintile, however, gender differences are apparent, especially for illness in the past 12 months and being absent from normal activities due to illness. With respect to education, our analysis is limited only to those aged 15-49 partly in order to minimize the effect of age on illness as well as its effect on education. 33 When we look at two indicators of serious illness, we observe that those with higher levels of education are less likely to report serious illness than those with lower levels of education (see Table 4-1). At every level of education, females are more likely to report serious illness than males. As a result, the proportion reporting serious illness (being absent from normal activities or being bedridden) is highest for females (age 15-49) who had no schooling at all. This suggests that women's education can play a significant role in protecting their health, such as raising awareness of reproductive health care and importance of safe birth. While females are in general more likely than males to report general illness or more serious illness, the average number of days missing normal activities or being bedridden over the 12- month period is higher for males than for females. Figures 4-3 shows that among people reporting serious illness, the average number of days missing normal activities like school or work is lower for females at almost all ages except for ages 20-29. Relatively larger gender differences are observed particularly for the age 30-39 and 40-49 groups where men report a higher number of days missing normal activities due to illness. Similar patterns are observed for the average number of days being bedridden due to illness (results not shown). Since there is no specific 32 We need to interpret regional differences reported in this study with caution. Since age is very closely associated with reported illness, differences in age structure across regions, rather than differences in living standards or living environments across regions, may underlie regional differences in reported illness shown here. For a more accurate assessment, age structures across regions need to be standardized, which is not undertaken in the current study. This study reports the results of age standardized distribution only with respect to expenditure quintiles. 33 For each level of schooling completed, age structure is very different since older people are significantly less likely to have higher levels of education, and schooling has increased progressively for the younger generation. 34 information regarding the nature or the type of illness in the VHLSS for the 12-month reference period, it is difficult to make clear inferences from this pattern. It could be that men's illness (or injury) tends to be more serious when it occurs, incapacitating them for more days from their normal activities. In addition to possible gender differences in the nature of illness, it could also be the case that even when faced with serious illness, women are less able to afford to miss their normal activities (including their household responsibilities) or to stay bedridden for a longer duration. Figure 4-3. Average Number of Days Missing Normal Activities among People Reporting Illness (in Past 12 Months) Days Male Female 40 30 20 10 0 0-4 5-9 10-14 15-19 20-29 30-39 40-49 50-59 60+ Age group Access to Health Care Services Among people reporting serious illness for the 12-month period, more than three-quarters had access to some type of health care services, with no apparent gender differences overall. Yet, gender differences are found in access to health care for certain age groups and for people with certain types of health insurance. Based on reported illness over the 12-month period, we further explore whether there are gender differences in access to health care among people reporting illness, and how access to health care varies by other factors such as area, age, ethnicity, living standards, and the type of health insurance. Table 4-2 presents the percentage of males and females who used any health care services over the 12-month period (including visits by health care workers or other medical practitioners), restricting only to those who reported being absent from normal activities due to illness.34 According to Table 4-2, more than three-quarters of males and females who reported illness accessed health care during the past year, and there is no significant gender difference in the 34 People who are visiting health care centers or visited by health care workers for the reasons of reproductive health care are also included, as long as they reported some serious illness over the 12-month period. Reasons for health care visit are examined in Table 4-4. 35 overall percentage (77 percent of males and 78 percent of females).35 Between urban and rural areas, there is also no significant difference, either for males or for females (the difference by area among males is not statistically significant). Yet, some gender differences are observed across different age groups. As illustrated in Figure 4-4, the rate of using health services is lower for females than for males up to age 15-19, while it is higher for females in the ages of 20-59. At older ages (60 and above), the difference by gender is relatively small. The higher rate of health care visit by females in the ages of 20-49 is likely to reflect their needs for reproductive health care. The gender difference among children aged 5-14, especially in rural areas, is relatively larger than other age groups, suggesting that illness among rural boys and girls tend to be handled somewhat differently by their parents. Figure 4-4. Percent People Using Health Care Services (Among people reporting illness in past 12 months) % Male Female 100 80 60 40 20 0 0-4 5-9 10-14 15-19 20-29 30-39 40-49 50-59 60+ Age group By ethnicity, across all ethnic groups, females are somewhat more likely than males or similarly to have used health care services when they were ill (see Table 4-2). Ethnic minorities overall appear to have used health care services less than the Kinh/Chinese, but this is true only for the Tay and others, and the Northern Mountain ethnic groups. Males from these two ethnic groups in particular were considerably less likely to have used health services than males from other ethnic groups (including the Kinh/Chinese), with only two-thirds seeking health care compared with more than three-quarters in the other groups. The Central ethnic groups and the Khmer/Cham were either more likely than the Kinh/Chinese or similarly to have used health services, both for females and males alike. By expenditure quintiles, we see somewhat different patterns of association between urban and rural areas (see Table 4-2). In urban areas, there is no particular pattern of association between expenditure quintiles and their access to health care services, except that males in the poorest quintile were least likely to use health care services when they were ill. In rural areas, there is a general pattern of increasing use of health care services at the higher quintiles, similarly for males 35 For all respondents (of all ages) including those who did not report any illness, the rate of visit to health services was 42 percent for females and 34 percent for males over the 12-month period. 36 and females. At the richest quintile, 83-84 percent of males and females have used health services when they were ill, compared with 73-74 percent of males and females at the poorest quintile. Across all expenditure quintiles, gender differences are relatively small. A relatively high use of health services among people who had illness, especially in rural areas or among certain ethnic minority groups, reflect the importance of health coverage programs that are targeted for the poor, young children, ethnic minorities, or people living in particularly disadvantaged areas. According to Table 4-2, there are some varying rates of health care use by the type of health coverage, however. People with health insurance as policy beneficiaries or with other voluntary health insurance--who tend to be older than others--are most likely than other groups to have used health services for the past year.36 Expectedly, those who have no health insurance are less likely than others to have used health care services, although still more than 70 percent of them have used health care services when they were ill. Importantly, those with students' health insurance, which is a part of voluntary health insurance programs, are less likely than others to have used health care services; in particular, females with students' health insurance are least likely, among all groups, to have used health care services both in urban and rural areas (69 percent in urban areas and 65 percent in rural areas). In fact, the gender difference for those with student health insurance in rural areas is large (65 percent for rural females and 75 percent for rural males), compared with those with other types of health insurance where gender differences are either negligible or females tend to have a higher rate of using health services. This is consistent with the pattern of health care use by age group examined above, where we observe substantially lesser use of health care among young rural females aged 5-14. From the current analysis, it is difficult to infer why this is the case; further analyses on access to health care will be required to assess whether this pattern indeed reflects differential treatment by rural parents for their sons' versus daughters' illness, or whether this simply reflects different types of illness between boys and girls in rural areas. Types of Health Care Services Used and Reasons for Their Use Among people who have used health care services for the 12-month period, types of health facilities used vary largely between urban and rural areas, but there is no apparent gender difference within each area. Reasons for visiting health care facilities tend to differ somewhat by gender, mainly due to women's visit for reproductive health care. The above examination shows that the rate of using health services by males and females who report illness is in general similar between urban and rural areas. Yet, the type of health care services utilized tends to differ greatly between urban and rural areas, reflecting important differences in the kinds of health facilities available in the two areas. Table 4-3 presents the distribution of health care services used among all people who had any access to health care, regardless of their illness.37 In urban areas, about 44 percent of males and females used district or provincial hospitals at least once in the past 12 months, and another 26 percent used private clinics. And, there is no gender difference in the use of these health facilities. In rural areas, on 36 People with these two types of health insurance also tend to be older; the average age for people with health insurance as policy beneficiaries is 51 and the average age for people with voluntary health insurance is 47. The average age for all respondents is 31. 37 The data on types of health care services examined in this section present only general information. Some respondents in the VHLSS had more than one visit to health care services over the 12-month period, and some have visited different types of facilities. The majority of respondents (about 80 percent); however, listed only one type of health services for the 12-month period. Although there are multiple types of services or facilities used by some respondents, we present the distribution of only the first type of health services listed by respondents. 37 the other hand, commune health centers were the main access to health care for both males and females, with one third using commune health centers; compared to males, females were somewhat more likely to rely on commune health centers as well (36 percent of females versus 31percent of males). About 20 percent of rural males and females also used district hospitals, and another 22 percent used private clinics. While there are large differences between urban and rural areas in the type of health facilities utilized, there are no apparent gender differences within each area. As for reasons of using health care services, Table 4-4 presents separately for three broad age groups by gender and area.38 For the vast majority of males and females who used health services, the main reason was to receive treatment. As expected, however, other reasons tend to vary by age group. For the young (age 0-19), vaccination is also a reason for visiting a health care facility, in addition to treatment or consultation. For women aged 20-49, pregnancy and other reproductive health care are also important reasons. There are no apparent gender differences within each age group except that women aged 20-49 are more likely to use health services for reproductive health reasons. In this section, the examination on the use of health care services has mainly focused on general access to health care by gender and a few selected characteristics. The study also briefly looked at types of health services used for the 12-month period and reasons for their use. It will be important in future research to examine more detailed aspects of the utilization of health care services from a gender perspective; for instance, how the use among out-patient services, in- patient services, and self treatment varies by gender. Given various types of health insurance coverage, it will also be important to analyze how the utilization of health services and out-of- pocket spending on health care differ across different types of health insurance, and whether benefits from each type of health insurance--in terms of reducing out-of-pocket payments--differ by gender. 5. Disability and Impact on Work Participation Given Vietnam's recent history with war and poverty, it is important for policy makers to have a good understanding of the extent and the nature of disabilities in the population. Disability among children or adults can hinder their developmental potential and productive capacities, posing a barrier in their effort to achieving better living standards. Having a family member with a disability also creates care and economic burden on other family members, and it may restrict economic participation of family members. Poor socioeconomic circumstances, poor health, lack of nutrition, and lack of healthy environment among children can all be causes of disability that limit opportunities and economic capacities as adults, which in turn likely lead them back to poverty. This section examines the prevalence, types, and causes of disability in the population by gender, using data from a special module on disability included in the 2006 VHLSS. 39 The VHLSS 38 The reasons of visiting health care services examined in this section are also based on the data on health care use for the 12-month period, as in the case of types of health care facilities used. In the VHLSS, for each type of health services used, its reason for visit was asked. Therefore, some respondents (about 20 percent) had multiple reasons for visiting health care services for the 12-month period. In this section, we again focus on the reason which is associated with the first visit listed by respondents. For about 80 percent of respondents, only one health service was used and hence only one reason was provided. 39 Since the VHLSS is a household survey collecting information on disability for people living in households, it excludes by definition those living in centers for people with disabilities. 38 measures disability in terms of bodily functional limitations and restrictions in activities.40 The questions in the VHLSS focus on disabilities in six categories: vision, hearing, mobility, cognition, communication, and self care (see Appendix A-2 for details on specific questions). The questions were fielded only to those aged 5 and older. For each category of disability, respondents were asked whether they have some difficulty, much difficulty, or impossible. 41 If a person has difficulty in any of the six categories, we consider the person as having a disability. We consider a person having a severe disability if the person has much difficulty or more in any of the six categories. Prevalence and Types of Disability Having a disability of some type increases with age, and gender differences in the disability rate vary by age. For people aged 30-49, men are more likely than women to have a disability, whereas for people aged 50 and older, women are more likely to have a disability. Among people aged 60 and older in particular, women are more likely than men to have a severe disability. Table 5-1 presents the percentage of people aged 5 and older who reported having a disability in each of the six categories, along with the overall rate of disability (i.e., the proportion having a disability in any of the six categories). These measures of disability are presented by gender, age, region, ethnicity, family living standards, and the level of education completed. The overall rate of disability shows that females are somewhat more likely to have a disability than males (17 percent of females compared with 15 percent of males). This overall rate of disability has largely to do with a relatively high percentage of people reporting a problem with their vision, compared with other categories of disability. The percentage of respondents reporting any vision problem is over 10 percent for both males and females, whereas the percentage reporting a disability in other categories is much lower, mostly below 5 percent (except for the category of mobility among females). Since people develop many health problems and functional difficulties as they age, the disability rate varies by age and the gender difference also varies by age. As shown in Figure 5-1, the overall disability rate increases steeply for people aged 40 and older: compared to the rate for people under age 40 (3-6 percent), the rate nearly triples for those aged 40-49 (16-17 percent). For people aged 60 and older, nearly two-thirds report having a disability of some type. There is no significant gender difference in the overall rate of disability among people age under 30. For those aged 30-39 who are in the prime working age, the overall disability rate is higher for men (6.4 percent) than for women (3.7 percent), although both rates are substantially lower than the rates observed among older people. As will be discussed shortly, this is associated with the 40 We need to keep in mind that disability is a difficult concept to measure and surveys within a country or across countries often use varying concepts and varying questions to measure the prevalence and the severity of disability in a population. Hence, estimates for the population with disabilities resulting from different surveys are usually not comparable. Recently, there has been much effort to use internationally standardized concept and questions to measure disability in national surveys, and the questions included in the 2006 VHLSS represent this effort. It is based on the International Classification of Functioning, Disability, and Health (often called the ICF approach), which defines disability with bodily functional limitations together with restriction in activities and social participation. The questions included in the 2006 VHLSS, however, covers only six categories of disability out of the nine categories recommended by the ICF approach. 41 For disability in each category, the VHLSS asks a set of questions--a main general question followed by a few more detailed questions. In this study, we estimate the rate of disability by using the main questions asked for disability in each category (see also Appendix A-2). 39 greater likelihood of accidents among males aged 30-39. For those aged 50 and older, the overall disability rate is higher for women than for men. Figure 5-1. Prevalence of Disability by Gender and Age % 70 Male Female 60 50 40 30 20 10 0 5-9 10-14 15-19 20-29 30-39 40-49 50-59 60+ Age group To examine the prevalent types of disability more easily, Table 5-2 re-presents the percentage reporting different types of disability by gender, for three broad age groups: age 5-19, age 20-49, and age 50 and older. For children aged 5-19, the disability rate is very low for most categories.42 For people aged 20-49, problems with vision are most common, followed by disabilities in mobility and cognition, although their percentages are all quite small (5 percent or less). For this age group, the overall disability rate is slightly higher for men than for women (9.2 percent versus 8.1 percent). For people aged 50 and older, again, the most common problem is with vision and the second common problem is with mobility, followed by disability in cognition and hearing, for males and female alike. Except for hearing, the disability rate in each of these categories is higher for women than for men in this age group; in particular, there is a relatively large gender difference in the percentage reporting difficulty with mobility (27 percent of women versus 18 percent of men). Overall, nearly half--52 percent of women and 46 percent of men--in this age group report having some type of disability. Looking at the disability rate by some other characteristics (see Table 5-1), we observe a higher overall rate of disability in urban areas than in rural areas for males and females alike, mainly due to a difference in the proportion reporting vision problems. For other types of disability, the rate of disability is similar between rural and urban areas, except for disability in communication where the rate is higher in rural than urban areas.43 Within each area, the overall disability rate is higher for women than for men. 42 Compared to other surveys, the 2006 VHLSS did not include separate questions on disabilities among children; the same set of questions was asked for everyone aged 5 and older. For questions regarding self care and communication, some responses seem to indicate problems associated with mere young age rather than those associated with some type of developmental disabilities. Responses regarding the causes of disability indicate that this may have been a confounding factor; some respondents list the cause of disability as age itself for children. 43 Unlike questions in other categories of disability, the questions concerning disability in communication are more specific by referring to pre-existing health conditions: "Due to his/her physical or emotional conditions, 40 By ethnicity, the overall rate of disability is higher for the Kinh/Chinese than ethnic minorities, again in part due to a large difference in vision problems between the two groups (see Table 5-1). Among different groups of ethnic minorities, the Northern Mountain ethnic groups show the lowest overall rate of disability for both males and females. Among ethnic minority females, the Khmer and Cham have the highest overall rate of disability, partly due to that this group consists of a higher proportion of older people than other groups.44 The difference among ethnic minority males is relatively smaller; the Tay and others, and the Central ethnic groups show a somewhat higher rate of disability than the other two groups. Within each ethnic minority group, females have a slightly higher overall rate of disability than males. With respect to family living standards, we observe somewhat higher overall rates of disability at the richer quintiles, similarly for males and females, mostly because of a higher rate of vision disability (see Table 5-1). For disability categories other than vision, we generally see the highest rate of disability among males and females at the poorest quintile (except for mobility among females), although the difference across quintiles is quite small. In fact, when we consider only severe level of disability, we observe a more apparent pattern of increasing rates of disability at the poorer quintiles, as discussed below. This suggests that poor economic circumstances and its associated living environments are closely related to disabilities. For each quintile, females still show a higher rate of disability than males. With respect to education, we focus only on men and women aged 20-49, in order to minimize the confounding effect of age on education and also the effect of aging itself on disability. The overall rate of disability is much higher for those with no schooling or less than primary education, similarly for men and women (see Table 5-1). Except for vision disability where those with higher education have a higher rate of disability, all other types of disability indicate the highest rate for those with no schooling. For this 20-49 age group, the overall disability rate is higher for men than for women within each level of schooling completed. Similar results are found when we consider the percentage of people who have a severe level of disability in each category of disabilities (see Table 5-3). For the overall rate of severe disability, it is slightly higher for females (4.0 percent) than for males (3.2 percent). Unlike the overall rate of disability that includes people with a moderate level of disability, the rate of severe disability is slightly higher in rural areas than in urban areas among females; among males, there is no difference between urban and rural areas for severe disability. With respect to age, as expected, the rate of severe disability is highest among people aged 60 and older, and the gender difference is large as well: 21 percent of females compared with 15 percent of males have a severe disability at these older ages. The large gender difference for this age group is observed particularly in the category of mobility: 13 percent of women compared with 8 percent of men have a severe disability in mobility (results not shown). For people aged 50-59, only about 4 percent of men and women have a severe disability and there is no gender difference. does the person have difficulty in communicating. e.g., understanding others, and making himself/herself understood?" The questions in other categories of disability, including those on cognition, do not specify "due to physical or emotional conditions" (see Appendix A-2). It is not clear whether the inclusion of this phrase resulted in more visible rural-urban differences in the category of communication, compared with other categories of disability. 44 Both the Kinh/Chinese and the Khmer/Cham females consist of a similarly high proportion of those aged 50 and older, compared with other ethnic minority females. About 12 percent of females from the two groups are aged 60 and older and another 10 percent are aged 50-59. For the other ethnic groups, only 13-15 percent of females are aged 50 and older overall. 41 From the household perspective, 12 percent of households overall have any family member with a severe disability, and the percentage is higher for rural households (13 percent) than urban households (11 percent; see Table 5-3). This, however, is likely to be an underestimate, since the VHLSS does not collect information on disability among children under age 5. Importantly, the proportion of households which have any family member with a severe disability increases for the lower quintiles; households at the poorest quintile are 50 percent more likely than those at the richest quintile to have a family member with a severe disability (15 percent compared with 10 percent). Causes of Disability Causes of disability vary considerably by age. For older people aged 50 and older, aging itself is a major cause of their disability, whereas for children (age 5-19), birth defects and illness are two notable causes. For adults (age 20-49), along with illness, accidents are an important cause of disability especially among men. Although gender differences are small, males are more likely to be affected by birth defects, accidents, or war/agent orange. To examine causes of disability, using data on multiple reasons listed for each type of disabilities, six categories were created: birth defects, war/agent orange, accidents, illness, aging, and other.45 As shown in Figure 5-2 (see also Table 5-4), causes of disability differ considerably by age group. While older people mostly report aging itself as a cause of disability, for young people aged 5-19 birth defects, along with illness, are important causes of disability. For people aged 20-49, especially men, accidents is a notable cause of disability, along with other causes. More detailed aspects for each age group are as follows. Figure 5-2. Causes of Disability by Gender and Age Group Birth defects War/Agent orange Accidents Illness Age Other % 100 80 60 40 20 0 Male Female Male Female Male Female Age 5-19 Age 20-49 Age 50+ 45 The 2006 VHLSS questions on causes of disability ask for more than one reason for each type of disabilities. While there can be multiple causes for a person's disability, analysis here provides summary statistics based on the primary reason reported for each type of disability. Also, while some people have more than one type of disability and hence different causes for different types of disabilities, mutually exclusive categories are created in this study by ordering the list of causes. The priority is given in the order of birth defects, war/agent orange, accidents, illness, age, and others. 42 For children aged 5-19, birth defects, along with illness, are main causes of disability. Males are more likely to be affected by birth defects (26 percent of males and 20 percent of females), while females are more affected by illness (30 percent of females and 27 percent of males).46 For people aged 20-49, illness is reported as the most common cause--nearly half responded illness as the major cause. Importantly, females with a disability are more likely to be affected by illness than males (52 percent of females compared with 43 percent of males). Birth defects are also reported as a cause of disability for approximately 10 percent of males and females alike. For this age group, accidents are a noticeable cause of disability, especially for males: males are four times more likely than females to report accidents as a cause of their disability (13 percent versus 3 percent). Although the percentage of people listing war/agent orange as a cause is small, it is higher for males than for females. For older people aged 50 and above, aging itself is a dominant cause of disability with more than 70 percent reporting aging as a cause, followed by illness. Females are somewhat more likely to report aging as a main cause than males (75 percent versus 70 percent). While there are no large gender differences in other causes of disability, war/agent orange is more likely to be a cause of disability for older males than older females (8 percent of males and 2 percent of females). (As for the association between causes of disability and other characteristics such as ethnicity and living standards, we cannot examine in detail because of small sample sizes.) Disability, School Attendance, and Work Participation Having a disability, especially a severe disability, hinders children's school attendance and adults' work participation. For both males and females, those with a severe disability are only about half as likely as those without any disability to attend school or participate in the labor force. The extent to which disability can be a barrier to children's school attendance or adults' work participation can be seen in Table 5-5. The table shows the percentage of children (age 6-17) attending school and the percentage of adults (age 25-54) who were employed in the past 12 months, by the level of one's own disability and also by the level of disability among other family members. Among children (age 6-17), having a severe disability hampers their school attendance for both males and females. Children with a severe disability are only half as likely as those without a disability to attend school. Only 44 percent of male children with a severe disability attend school, and a somewhat lower percentage of female children with a severe disability, 41 percent, do so. Male children with a moderate level of disability also seem less likely to attend school than those without a disability, but the same pattern is not observed for female children. Although there could be other associated factors, this certainly suggests that many children who have a severe disability are likely to face serious disadvantages for their future economic opportunities due to lack of schooling. As for the potential impact of having a family member with a disability, both male and female children who have any family member with a severe disability are also somewhat less likely to attend school, compared with those children who do not. It is important to keep in mind that this 46 Age itself is listed as a cause of disability for about 20 percent of young children, probably because some questions on disability (e.g., cognition and self care) are ambiguous and interpreted incorrectly by respondents. 43 relationship does not necessarily implicate a causal impact, since the likelihood of having a family member with a severe disability and children's school attendance tend to have other correlates that influence the relationship, including poor economic circumstances and living environments. Among adults (age 25-54), we also find a similar relationship between disability and their work participation. Both males and females with a moderate disability are somewhat less likely to have worked than those without a disability; still, more than 90 percent of males and 89 percent of females who reported a moderate disability have worked in the past 12 months. On the other hand, those with a severe disability are substantially less likely to have worked; a little more than half of males (53 percent) and only 42 percent of females who have a severe disability have worked. Compared with men, having a severe disability among women of working age seems to have a greater limiting impact on their work participation. A similar pattern is found when we consider urban and rural areas separately (results not shown). As in the case of children's school attendance, both males and female of working age who have any family member with a severe disability are somewhat less likely to have worked than those who do not have any family member with a disability. While this study is confined to examining a simple bivariate association between family member's disability and work participation, it will be important in future research to investigate this issue more closely. Having a severe disability of one's own can certainly restrict various activities including school attendance and income generating activities. On the other hand, it is not so obvious from the outset how having a family member with a disability would affect economic activities of other family members. There can be two offsetting effects. When a family member, for instance, a husband, is restricted in economic activities due to disability, this can create increased need for a wife to work to support the family. At the same time, however, having a family member with a disability can create additional care burden for other family members, especially women, by hindering their work participation outside the household. The direction of these two offsetting effects can be influenced by who has a disability (e.g., adults or children), the type of disability, the availability of other family members who can provide either economic or caregiving support, and the availability of local support systems for the disabled. In this context, it will be especially important to explore how the impact of having a family member with disability varies between women and men. Suggestions for Future Research Using recent data from the 2006 VHLSS, this report assessed the current gender situation in Vietnam by focusing on several key social and economic characteristics of households and household members. In specific, the report examined how females fare in education, employment, and health, by looking at indicators such as school attendance, attendance at extra classes, labor force participation, types of employment, earnings in wage employment, health status, access to health care, and the prevalence of disability. Since the current study analyzed these indicators mainly based on descriptive statistics, it will be important for future studies to closely look into their causal determinants and their effects in multivariate frameworks. Also, given the limited scope of the current report, future studies should further explore other data available in the VHLSS from a gender perspective. In view of the recently enacted law on Gender Equality, it will be essential to fully utilize available data and make efforts to improve existing data, in order to aid the formulation of specific gender policies and to provide timely assessment of the effectiveness of the law. Based on findings of the current 44 study, this section outlines several key issues that require further research attention, along with the areas that need improved data, in order to enhance our understanding of the gender situation in Vietnam. As demonstrated in the section on education, most visible progress has been made in reducing the gender gap in school attendance, with girls now exceeding boys in attending upper secondary schools. Increased school attendance by rural females and ethnic females is especially noticeable in recent VHLSS data. Progress in girls' education is also in accordance with their higher rate of attendance at extra classes. At both upper secondary and higher levels of schooling, however, the rate of school attendance still varies significantly by area, ethnicity, and family's socioeconomic characteristics. · It will be important for future studies to identify the kinds of barriers specific to different groups, particularly to rural, ethnic, and poor households. The kinds of barriers can be categorized into personal, family, institutional, and environmental factors, which include children's health, language barriers (for ethnic minorities), household/family characteristics, provision of schools at local levels, types of infrastructures hindering access to schools, kinds of expenditures associated with school attendance, and economic opportunities associated with educational attainment. · Even though much progress has been made in closing the gender gap in school attendance, it is crucial in future research to investigate what it means in terms of quality of education. Do higher levels of schooling provide advanced knowledge and skills base that can meet the needs of rapidly developing society? While this report focused mainly on the indicator of school attendance, it is imperative to examine the contents of education and identify measures of schooling quality, with close attention to gender differences. · Another important future task is to investigate potential gender differences in the benefits or outcomes of educational attainment. In particular, given large regional differences in the level and pace of economic development, it will be essential for future studies to incorporate different regional industrial structures that underlie labor demand for young people. Considering the structure of labor demand across regions is crucial for understanding incentives for and barriers to educational attainment. For instance, what are local economic opportunities available for young people completing different levels of schooling? How do those opportunities differ between females and males, and also by region or by ethnicity? In addition, how do different levels of schooling between females and males meet skills needs in newly emerging industrial jobs? · In learning about labor market paths taken by young females and males with different levels of schooling, it will also be important to examine gender differences in migration patterns, especially among rural youth. Using the panel component available in the VHLSS, migration patterns can be explored: for instance, whether there are similarities or differences in leaving the household between females and males, and in their reasons for leaving the household (e.g., school, work, or marriage). In relation to migration patterns by gender, it will also be important to explore the extent to which migrant female and male workers contribute to their family incomes. The VHLSS, unfortunately, does not specify the sources of remittance from non-household members; having specific information on the source would be valuable in examining economic contributions made by migrant daughters and sons. 45 · Given the growing prevalence of attendance at extra classes and its associated expenditures, more research is needed to examine its actual impact on school advancement and how its effect varies between male and female students. Since there appears to a close relationship between extra class attendance and family living standards, it will be crucial to assess the independent effect of extra classes on school advancement in a multivariate framework: that is, whether attendance at extra classes itself contributes to one's academic performance and advancement to higher levels of schooling irrespective of one's socioeconomic backgrounds, or whether extra class attendance merely reflects privileges of higher economic backgrounds. If advancement to higher levels of education becomes more advantageous to those who can afford high expenditures associated with extra class attendance, the meritocratic function of education may lessen while contributing to increasing income inequality. In the area of employment, although women's employment still largely consists of agricultural work for the country as a whole, wage employment and nonagricultural self-employment have been increasing, especially in urban areas. While women in Vietnam exhibit a very high rate of labor force participation similar to men, there is a considerable degree of gender segregation in industry and occupation categories, both in wage employment and nonagricultural self- employment. · In wage employment, there still exists the persistent gender gap in earnings, in spite of women's higher levels of educational attainment relative to male wage workers. We need a better understanding of the determinants of wages between women and men in the current labor market, using recent data in a multivariate framework. Past research indicates that returns to education were relatively low in Vietnam during the early 1990s but they have been increasing through the late 1990s, particularly in the private sector and for females with higher education (Gallup 2004; Liu 2005; Nguyen 2004). Using recent data, it will be important to analyze how the determinants of wages vary between women and men. Since wages tend to differ widely by sector of employment and occupation, special attention needs to be paid to the extent to which wages are determined by worker characteristics (e.g., education or work experience) versus the extent to which wages are influenced by sector of employment or other job characteristics. In addition, it will be important to examine how benefits other than the main salary differ by gender, as well as by sector of employment, occupation, and industry. In view of the Gender Equality Law, it will be crucial to identify precise sources of gender differences in earnings among wage workers. · One important finding of this report is that despite remarkable increases in women's school attendance, their major fields of study in higher education are still concentrated in a few traditionally female fields and these are closely mirrored in gender segregation in occupations. It will be important for future research to examine whether women's differing choice of major fields in higher education reflects diverging opportunities in the labor market, or whether it represents more of sex-differentiated preferences resulting from earlier socialization. Does women's concentration in a few predominantly-female occupations reflect their preference and lack of relevant training to enter other occupations? Or, are women more likely to be pushed into these few occupations due to institutional barriers in entering other occupations? What are the kinds of institutional factors that may channel women and men into diverging career paths and earnings? Are women and men recruited and hired with the same procedures and standards, or are there gender-differentiated recruitment processes in certain occupations? Answers to these questions will provide critical information in monitoring the impact of the recent Gender Equality Law. 46 · Given a large share of women engaged in nonagricultural self-employment, particularly in urban areas, but given wide differences in the type of work engaged between women and men, a closer examination is needed for gender differences in more detailed aspects of nonagricultural self-employment. Women's self-employment in this area mostly consists of household enterprises which are small in scale, largely in the informal sector, and heavily concentrated in retail sales or foods/lodging services. We need a better understanding of the stability of women's work in this area and the impact of different kinds of work on their living standards, not only compared with their male counterparts, but also compared with other female wage workers. In addition, it will be important to explore potential differences among women themselves in nonagricultural self-employment: differences in skills base, access to credit resources, contribution of their earnings to household income, and so on. · In examining the above issues, the existing VHLSS data can provide information for certain aspects, but some modifications or additions in the VHLSS data collection would be useful for better assessing the gender situation in nonagricultural self-employment. Since nonagricultural self-employment operates mostly as household enterprises, the VHLSS collects information on these activities largely at the household level, rather than at the individual level (individual-level information is collected for working hours, industry, and occupation categories). With current VHLSS data, therefore, it is difficult to assess whether skills training is involved at all for women's specific activities, whether women have obtained access to credits for these activities, and which other household members are involved in these activities. In view of the Gender Equality Law, more individual-level information is needed to evaluate the current status of financial or skills support systems for female entrepreneurs or female-operated nonagricultural self-employment vis-à-vis male- operated ones. With respect to health, women usually experience more chronic health problems than men over the life cycle. The current report shows that health problems and disabilities are prevalent at older ages, especially among women. This may result from women's relative lack of access to preventive care or to quality reproductive care from earlier in their life cycle. One major cause of disability being illness, it is crucial for women (as well as for men) to have access to quality health care which can prevent their illness from becoming permanent disabilities. Women's health is also vital for children's and family's well-being, since women's health influences greatly on children's nutritional status and their healthy development, which in turn affects their school attendance and future productive capacities. · Given a brief analysis of access to health care in this study, more detailed studies are needed for understanding gender differences in other aspects of access to health care. In particular, further research attention is required for the type of health care services used, that is, whether females and males differ in their utilization among out-patient services, in-patient services, and self treatment. It is also important to analyze how out-of-pocket expenditures on health vary between females and males in the household. All these should be examined with close attention to types of health insurance, which vary by age, family living standards, and area or region. · As for the impact of disability, it will be important to further explore how the impact of having a disability on school attendance or work participation varies by the type of disability, household and other local characteristics. Given a sizable proportion of households having a family member with a severe disability (12 percent), it is also important to examine how having a family member with a disability affects household economics--including care 47 burden, impact on income generating activities, impact on health care costs, and so on. The identification of various factors associated with the impact of disability is essential for developing policies that can accommodate people with disabilities into the workforce, and for designing the kinds of support systems necessary for people with a disability as well as for families with disabled family members. · Currently, there is a lack of data with respect to care work going on within the household. Some additional information in the VHLSS can be valuable in examining the gender situation in the kinds of unpaid work being done in the household. For instance, the existing data on housework in the VHLSS only ask about overall housework being done; the breakdown of information for major kinds of housework--cleaning, cooking, taking care of children, taking care of elders, etc--would provide more useful data on the division of household labor by gender, as well as the extent of care work being carried out within the household. In light of the recent enactment of Gender Equality Law, various gender indicators available in the VHLSS can play an important role in providing empirical bases for formulating specific policy strategies and in monitoring the effectiveness of the law. Certainly, more empirical data are needed in order to evaluate the impact of Gender Equality Law, especially in such areas as family relations and domestic violence. Yet, given the availability of the VHLSS which has already been providing valuable empirical data on gender indicators in the areas of education, employment, health, and household economics to some extent, it is important for researchers and policy makers to fully utilize the available data source and make efforts to improve the existing one. Analysis of gender indicators over time with a consistent data source is essential for assessing progress and challenges in achieving gender equality in Vietnam. 48 References Baulch, Bob, Hung T. Pham, and Barry Reilly. 2008a. "Ethnicity and Household Welfare in Rural Vietnam, 1993-2004." Working paper (Third draft). Accessed at http://www.ids.ac.uk/UserFiles/File/poverty_team/ethnicity_and_household_welfare_in_rural_Vi etnam.pdf Baulch, Bob, Hung T. Pham, and Thi Thu Phuong Nguyen. 2008b. "The Economic Development of Ethnic Minorities in Vietnam." Policy Brief. Institute of Development Studies at the University of Sussex. Accessed at http://www.ids.ac.uk/UserFiles/ File/poverty_team/POLICYBRIEF-final-Eng.pdf Betson, David M. 2004. "Poverty Equivalence Scales: Adjustment for Demographic Differences Across Families." Paper presented at the National Research Council Workshop on Experimental Poverty Measures, June 15-16, Washington, DC. Deaton, Angus. 1997. The Analysis of Household Surveys. Baltimore, MD: The Johns Hopkins University Press (published for the World Bank). Desai, Jaikishan. 2000. Vietnam Through the Lens of Gender: Five Years Later, Results from the Second Vietnam Living Standards Survey A report prepared for the Food and Agriculture Organization of the United Nations, Regional Office for Asia and the Pacific. Gallup, John Luke. 2004. "The Wage Labor Market and Inequality in Vietnam" Pp. 53-94 in Economic Growth, Poverty, and Household Welfare in Vietnam, edited by Paul Glewwe, Nisha Agrawal, and David Dollar. Washington, DC: The World Bank. Lee, Sunhwa. 2006. "Gender Analysis of the 2004 Vietnam Household Living Standard Survey." A paper prepared for the World Bank, Vietnam. Liu, Amy Y.C. 2005. "Changing Wage Structure and Education in Vietnam 1993-1998: The Roles of Demand." Working papers 05-4 in International and Development Economics, Australian National University. Ministry of Health and General Statistical Office, Vietam. Results of the Vietnam National Health Survey 2001-2002. Nelson, Julie. 1993. "Household Equivalence Scales: Theory vs. Policy." Journal of Labor Economics 11: 471-493. Nguyen, Nga Nguyet. 2004. "Trends in the Education Sector." Pp. 425-466 in Economic Growth, Poverty, and Household Welfare in Vietnam, edited by Paul Glewwe, Nisha Agrawal, and David Dollar. Washington, DC: The World Bank. Swinkles, Rob, and Carrie Turk. 2006. "Explaining Ethnic Minority Poverty in Vietnam: A Summary of Recent Trends and Current Challenges." Background paper for CEM/MPI meeting on ethnic minority poverty, Hanoi, September 2006. 49 White, Howard, and Masset Edoardo. 2002. "Child Poverty in Vietnam: Using Adult Equivalence Scales to Estimate Income-Poverty for Different Age Groups." MPRA Paper No. 777, posted November 2007. Munich Personal RePEc Archive. World Bank Institute. 2005. Introduction to Poverty Analysis. Washington, DC: The World Bank Institute. World Health Organization. 2003. Health and Ethnic Minorities in Viet Nam. Vietnam: World Health Organization. 50 Tables Table 1-1. Distribution of Households in 1998 (VLSS), 2004 and 2006 VHLSS 1998 VLSS 2004 VHLSS 2006 VHLSS (No. of households) (N = 5,999) (N = 9,188) (N = 9,189) Region (%) Red River Delta 22.8 24.0 23.7 Northern Uplands 17.1 North East 11.3 11.3 North West 2.6 2.8 North Central Coast 14.0 12.6 13.0 South Central Coast 10.3 8.7 8.3 Central Highlands 3.1 4.9 5.2 South East 12.5 15.5 15.5 Mekong River Delta 20.3 20.6 20.2 Rural vs. Urban (%) Rural 76.0 73.5 72.5 Urban 24.0 26.5 27.5 Ethnicity (%) Kinh 85.9 88.6 87.8 Chinese 1.8 0.9 0.9 Ethnic minorities 12.3 10.5 11.3 Household type (%) Nuclear 71.2 74.2 73.0 Vertical 23.7 18.6 18.7 Others 5.1 7.2 8.2 Mean Household size 4.7 4.3 4.2 Gender of household head (%) Male 73.7 74.5 74.5 Female 26.4 25.5 25.6 Marital status of household head (%) Married 81.2 80.7 81.1 Widowed 13.6 14.6 14.1 Divorced 1.5 1.8 1.7 Separated 1.5 0.9 0.9 Unmarried 2.2 2.0 2.3 Age of household head (%) <=24 0.9 0.5 0.6 25-34 16.7 12.7 10.8 35-44 30.8 29.0 28.3 45-54 20.0 26.5 28.4 55-64 16.5 14.4 15.7 65 and older 15.1 16.9 16.2 Mean age of household head 47.8 49.0 49.8 Poverty and food poverty incidence of all household members (No. of individuals) (N = 28,509) (N = 40,438) (N = 39,071) % in poverty 37.4 19.5 16.0 % in food poverty 15.0 7.4 6.7 Note: The statistics for 1998 are taken from Desai (2000), Tables 2.1, 2.2, and 3.2. The statistics for 2004 and 2006 are author's own calculations. Percentages are weighted; sample sizes are unweighted totals. 51 Table 1-2. Poverty Rates by Gender and Selected Characteristics (%) Total Male Female (No. of individuals) (N = 39,071) (N = 19,157) (N = 19,914) Total 16.0 15.6 16.3 Rural vs. Urban Rural 20.4 19.9 20.8 Urban 3.9 3.7 4.0 Region Red River Delta 8.8 8.3 9.3 North East 25.0 23.5 26.5 North West 49.0 48.0 50.1 North Central Coast 29.1 27.6 30.6 South Central Coast 12.6 12.6 12.5 Central Highlands 28.6 28.3 29.0 South East 5.8 5.8 5.9 Mekong River Delta 10.3 10.5 10.1 Ethnicity Kinh/Chinese 10.3 9.9 10.7 Ethnic minorities* 52.3 51.6 53.0 Tay/Thai/Muong/Nung 42.6 41.2 44.0 Other Northern Mountain ethnic groups 70.9 69.5 72.3 Central ethnic groups 71.1 73.2 69.0 Khmer/Cham 33.6 32.9 34.4 Age group 0-14 23.8 23.7 23.8 15-24 13.4 12.5 14.4 25-34 17.3 16.2 18.4 35-44 13.8 13.9 13.7 45-54 9.4 9.4 9.4 55-64 10.9 11.1 10.7 65+ 15.8 13.6 17.4 Household structure Nuclear 14.2 13.9 14.5 Vertical 17.9 17.7 18.1 Others 22.1 21.7 22.4 Gender of household head Male 17.2 16.7 17.7 Female 11.8 11.0 12.3 Note: Percentages are weighted; sample sizes are unweighted totals. * For details of ethnic minority groups, see Appendix A-1. 52 Table 1-3. Characteristics of Male-Headed and Female-Headed Households Male-Headed Female-Headed Total Married Widowed Other* Total Married Widowed Other* (No. of households) (6,933) (6,656) (184) (93) (2,256) (859) (1,068) (329) Region (%) Red River Delta 24.1 24.4 22.0 13.9 22.4 23.8 21.8 20.9 North East 12.0 12.1 9.4 9.5 9.4 12.9 6.3 10.2 North West 3.2 3.2 0.4 2.7 1.7 2.2 1.3 1.9 North Central Coast 13.7 13.9 12.3 2.8 10.8 8.1 13.8 8.3 South Central Coast 8.3 8.3 8.2 8.5 8.4 8.3 8.4 8.4 Central Highlands 5.6 5.8 2.7 1.2 3.9 3.7 4.3 3.6 South East 13.1 12.5 20.8 39.7 22.5 25.2 18.6 28.1 Mekong River Delta 20.0 19.8 24.3 21.7 20.8 15.8 25.5 18.8 Rural vs. Urban (%) Rural 77.3 77.9 68.4 58.5 58.4 45.2 70.4 54.5 Urban 22.7 22.1 31.6 41.5 41.6 54.8 29.6 45.5 Ethnicity (%) Kinh/Chinese 86.8 86.7 90.4 92.4 94.2 94.7 93.3 95.5 Ethnic minorities 13.2 13.3 9.6 7.7 5.8 5.3 6.7 4.5 Marital status (%) Married 95.8 100.0 0.0 0.0 38.1 100.0 0.0 0.0 Widowed 2.8 0.0 100.0 0.0 47.0 0.0 100.0 0.0 Divorced/separated 0.6 0.0 0.0 40.5 8.4 0.0 0.0 56.4 Unmarried 0.9 0.0 0.0 59.6 6.5 0.0 0.0 43.6 Age of household head (%) <=24 0.7 0.6 0.0 11.3 0.2 0.5 0.0 0.3 25-34 12.4 12.6 0.5 28.5 6.1 12.1 1.0 6.7 35-44 31.8 32.5 4.6 34.4 18.1 29.7 7.5 21.8 45-54 27.8 28.4 11.0 15.7 30.3 37.2 20.6 43.4 55-64 13.9 13.9 17.4 7.5 21.3 15.5 25.3 23.3 65 and older 13.5 12.1 66.6 2.7 24.0 5.0 45.6 4.5 Mean age of household head 48.2 47.8 68.2 38.2 54.4 46.6 62.4 49.3 Household structure (%) Nuclear 75.9 76.9 54.6 54.4 64.6 77.7 54.1 64.2 Vertical 17.2 16.9 28.7 18.0 23.2 15.2 31.7 16.7 Others 6.9 6.3 16.7 27.6 12.3 7.1 14.3 19.1 Mean household size 4.4 4.5 3.4 2.2 3.6 4.1 3.5 2.8 % with adult male(s) age 18+ 100.0 100.0 100.0 98.1 69.9 91.7 60.7 43.0 Presence of children % with children age <6 26.5 26.9 21.3 6.9 21.5 21.0 23.3 16.9 % with any children <18 73.6 75.0 49.8 24.3 60.3 69.4 57.1 47.5 Presence of elders % with elders age 60+ 28.0 26.5 76.5 34.6 41.6 25.1 60.9 22.7 (continued) 53 Table 1-3. Characteristics of Male-Headed and Female-Headed Households (continued) Male-Headed Female-Headed Total Married Widowed Other* Total Married Widowed Other* (No. of households) (6,933) (6,656) (184) (93) (2,256) (859) (1,068) (329) Education (%) No diploma 4.4 4.1 13.6 4.6 15.3 3.3 27.1 8.7 Less than primary 16.9 16.7 29.2 10.0 26.0 16.1 34.8 23.4 Primary school 27.1 27.1 20.8 40.9 19.0 20.8 16.2 23.2 Lower secondary 32.8 33.3 22.2 19.4 22.3 29.4 15.6 25.1 Upper secondary 14.1 14.3 9.3 11.7 12.3 20.6 5.2 13.5 College/university or more 4.6 4.5 5.0 13.6 5.2 9.8 1.2 6.2 Labor force status (%) Employed 90.0 91.1 50.7 94.5 72.8 85.3 59.7 82.2 Unemployed 0.2 0.2 0.5 2.1 0.1 0.1 0.0 0.2 Inactive 9.8 8.7 48.8 3.4 27.1 14.6 40.3 17.6 Employment sector (%) Wage employment 31.2 31.2 15.8 45.3 25.0 30.3 17.0 29.7 Agriculture self-employment 46.5 46.4 66.8 29.2 40.4 32.4 50.9 37.7 Non-agriculture self- employment 22.4 22.4 17.4 25.5 34.6 37.3 32.2 32.6 (For all household members) Expenditure quintiles (%) 1 poorest 21.7 21.8 19.5 19.6 13.9 8.4 18.6 16.1 2 21.2 21.5 10.8 17.6 15.7 9.8 20.5 18.9 3 20.7 20.7 21.0 10.3 17.7 14.7 20.3 18.5 4 19.3 19.3 19.8 20.9 22.5 25.8 21.2 15.4 5 wealthiest 17.1 16.7 29.0 31.5 30.2 41.4 19.4 31.1 Poverty Incidence (%) 17.2 17.2 15.0 16.9 11.8 6.8 15.5 15.7 Food poverty incidence (%) 7.3 7.3 5.3 6.5 4.6 2.8 6.2 4.8 Note: Percentages are weighted; sample sizes are unweighted totals. * Other category includes those who are divorced, separated, or never married. 54 Table 1-4. Household Characteristics by Ethnicity Ethnic Minorities Tay, Thai, Northern Kinh & Muong & Mountain Central Khmer & Chinese Total Nung Ethnic Ethnic Cham (No. of households) (7,805) (1,384) (753) (257) (238) (136) Region (%) Red River Delta 26.6 1.2 1.8 1.3 0.6 0.0 North East 8.0 37.4 51.7 64.2 0.0 0.0 North West 0.8 18.6 26.9 27.7 0.0 0.0 North Central Coast 13.3 10.2 14.2 3.2 10.9 0.0 South Central Coast 8.9 3.8 0.0 0.0 19.1 0.6 Central Highlands 4.2 13.0 3.5 2.7 55.9 0.0 South East 16.9 4.8 1.4 1.0 13.6 10.5 Mekong River Delta 21.3 11.0 0.6 0.0 0.0 88.9 Rural vs. urban (%) Rural 70.0 92.1 92.3 95.4 90.3 90.3 Urban 30.0 7.9 7.7 4.7 9.7 9.7 Household structure (%) Nuclear 73.8 67.0 66.4 62.2 72.7 66.7 Vertical 18.5 20.5 22.6 16.3 16.9 22.1 Others 7.7 12.5 11.1 21.5 10.5 11.2 Mean household size 4.1 5.0 4.8 5.6 5.5 4.6 Age of household head (%) <=24 0.5 1.8 1.7 3.2 1.9 0.0 25-34 9.5 21.5 23.0 26.4 22.0 8.0 35-44 27.9 31.5 34.1 32.8 26.0 27.2 45-54 29.0 23.7 24.3 18.6 25.1 25.6 55-64 16.2 12.3 10.9 13.3 13.3 15.7 65+ 17.1 9.2 6.0 5.8 11.6 23.6 Mean age of household head 50.4 44.8 43.3 42.7 46.0 52.1 % with adult male(s) age 18+ 91.7 96.6 97.1 95.2 95.9 95.3 Presence of children % with any child age <6 23.5 38.3 33.4 47.1 51.3 27.9 % with any child age 6-10 26.1 38.6 32.6 44.3 58.2 26.8 % with any child <18 68.2 85.9 85.2 90.2 90.8 75.5 Presence of elders % with any elder age 60+ 32.1 26.3 26.8 22.8 24.6 31.6 % with any male elder age 60+ 18.0 14.7 14.8 10.8 16.6 16.2 % with any female elder age 60+ 25.0 20.9 20.7 18.5 18.5 28.3 Gender of household head (%) Male 72.9 86.8 88.5 91.6 85.6 75.3 Female 27.1 13.2 11.5 8.4 14.4 24.7 Marital status of household head (%) Married 80.1 88.9 91.5 90.4 87.7 77.3 Widowed 14.7 8.8 6.6 7.6 9.3 19.7 Divorced/separated 2.7 1.3 1.0 1.1 1.5 2.9 Never married 2.5 0.9 0.9 1.0 1.5 0.0 (continued) 55 Table 1-4. Household Characteristics by Ethnicity (continued) Ethnic Minorities Tay, Thai, Northern Kinh & Muong & Mountain Central Khmer & Chinese Total Nung Ethnic Ethnic Cham (No. of households) (7,805) (1,384) (753) (257) (238) (136) Completed schooling of household head (%) No schooling 5.6 19.7 5.9 35.1 39.5 29.8 Less than primary 18.4 26.2 19.8 29.6 32.9 39.7 Primary 24.6 28.3 35.5 20.8 18.5 21.0 Lower secondary 31.4 19.8 28.8 12.5 8.5 6.7 Upper secondary 14.8 4.7 7.6 1.0 0.6 2.9 Junior college/university 5.2 1.4 2.3 1.0 0.0 0.0 Labor force status of household head (%) Employed 84.6 93.6 95.8 94.6 92.8 83.5 Unemployed 0.2 0.1 0.0 0.0 0.0 0.7 Inactive 15.2 6.4 4.2 5.5 7.2 15.8 Employment sector of the employed (%) Wage employment 31.8 15.8 13.6 8.4 16.9 35.5 Agriculture self-employment 40.9 75.8 79.5 87.6 75.8 40.5 Nonagriculture self-employment 27.4 8.4 6.9 4.0 7.3 24.1 (All household members) Expenditure quintiles (%) 1 poorest 13.9 59.1 48.9 75.4 78.1 45.5 2 19.7 22.2 26.1 16.3 14.0 28.4 3 21.5 10.7 14.5 3.4 5.4 13.6 4 22.2 6.1 7.4 3.7 2.5 10.7 5 wealthiest 22.8 2.0 3.1 1.3 0.1 1.8 Poverty incidence (%) 10.3 52.3 42.6 70.9 71.1 33.6 Food poverty incidence (%) 3.2 29.2 18.6 49.8 48.3 11.3 Note: Percentages are weighted; sample sizes are unweighted totals. 56 Table 2-1. Levels of Schooling Completed among Men and Women Aged 18-64 (%) MEN Age 18-21 22-24 25-34 35-44 45-54 55-64 No schooling 2.4 3.0 4.8 3.6 2.7 3.5 Less than primary 4.7 7.3 13.2 12.6 11.6 19.2 Primary 19.0 20.5 31.3 28.7 22.0 23.8 Lower secondary 30.3 22.8 23.9 33.0 39.4 31.4 Uppper seconary 43.0 40.6 18.1 17.5 17.8 15.0 Junior College 0.3 2.5 2.1 0.9 1.0 0.8 University or more 0.4 3.5 6.5 3.7 5.5 6.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 % currently attending school* 40.9 20.7 3.8 1.0 0.6 0.0 (N) (1,774) (1,072) (2,428) (2,804) (2,268) (1,104) WOMEN Age 18-21 22-24 25-34 35-44 45-54 55-64 No schooling 3.7 4.4 6.7 5.0 6.7 11.8 Less than primary 5.8 7.7 12.1 15.3 19.6 35.4 Primary 16.5 22.9 34.6 27.6 23.1 22.7 Lower secondary 26.2 19.4 23.7 32.1 34.9 19.5 Uppper seconary 46.8 35.6 15.1 16.3 11.1 6.8 Junior College 0.8 4.6 2.6 1.4 1.6 1.2 University or more 0.4 5.5 5.1 2.4 3.0 2.8 Total 100.0 100.0 100.0 100.0 100.0 100.0 % currently attending school* 41.7 15.3 1.9 0.8 0.1 0.0 (N) (1,581) (1,000) (2,531) (2,925) (2,599) (1,322) Note: Percentages are weighted; sample sizes are unweighted. * Those who are attending vocational schools are also included in the rate of current attendance. 57 Table 2-2. Levels of Schooling Currently Attended and Completed in the School-Age Population (Age 6-24; %) 2006 VHLSS Male Female 6-10 11-14 15-17 18-21 22-24 6-10 11-14 15-17 18-21 22-24 Currently attending Preschool 4.2 5.3 Primary 88.8 11.7 0.4 88.2 12.1 0.1 Lower secondary 2.9 77.4 18.0 0.6 3.3 78.3 16.9 0.3 Upper secondary 2.5 51.5 14.6 0.6 2.7 57.1 13.0 0.1 Junior college/university 0.2 15.4 14.4 0.5 17.2 11.3 Vocational training 1.5 10.1 5.7 0.9 11.1 4.0 Not attending - highest level completed Primary or less 4.0 8.2 18.4 25.5 30.7 3.3 6.8 15.8 25.4 35.0 Lower secondary 0.1 9.9 20.1 21.5 0.2 8.5 18.5 19.2 Upper secondary 0.3 13.4 22.8 0.1 13.9 23.4 Junior college/university 0.2 4.4 0.6 7.1 (N) (1,645) (1,770) (1,617) (1,774) (1,072) (1,501) (1,886) (1,463) (1,581) (1,000) 2004 VHLSS* Male Female 6-10 11-14 15-17 18-21 22-24 6-10 11-14 15-17 18-21 22-24 Currently attending Preschool 4.2 3.4 Primary 88.8 17.3 0.8 89.7 16.4 0.4 Lower secondary 2.7 73.4 24.1 1.3 0.1 2.1 73.2 18.1 0.8 0.1 Upper secondary 1.8 46.2 18.3 0.9 2.4 49.7 12.6 0.6 Junior college/university 0.1 12.1 13.8 0.2 15.3 8.4 Vocational training 1.3 10.8 6.2 0.7 7.5 4.2 Not attending - highest level completed Primary or less 4.3 7.5 18.5 29.4 37.9 4.8 7.7 21.2 32.0 42.1 Lower secondary 0.1 8.7 16.9 20.0 0.3 9.0 18.4 19.1 Upper secondary 0.3 10.5 16.9 0.7 13.1 19.7 Junior college/university 0.6 4.1 0.2 5.9 (N) (1,967) (2,154) (1,636) (1,785) (999) (1,895) (2,036) (1,454) (1,670) (848) Note: Percentages are weighted; sample sizes are unweighted totals. *The results from 2004 VHLSS are author's own calculations. 58 Table 2-3. Major Fields of Study in Higher Education (%) (Junior College, University, MA, and Ph.D. levels all combined) Current Attendees (All ages) Male Female Total Education/teaching 9.0 24.6 16.1 Humanities/social science 6.5 8.4 7.3 Business 13.1 29.8 20.8 Law 2.3 1.2 1.8 Natural science 4.0 3.1 3.6 Math/statistics 2.2 3.7 2.9 Computer 8.1 4.9 6.7 Technology/engineering* 38.6 7.1 24.1 Agriculture/animal science 4.5 2.4 3.5 Other** 11.8 14.8 13.1 Total 100.0 100.0 100.0 (Sample N) (482) (419) (901) Graduates (Age 30-44) Male Female Total Education/teaching 22.3 50.6 35.5 Humanities/social science 8.9 8.0 8.5 Business 13.2 22.5 17.5 Law 4.5 0.0 2.4 Natural science 3.1 1.3 2.3 Math/statistics 3.1 6.0 4.4 Computer 2.5 1.4 2.0 Technology/engineering* 23.5 4.2 14.5 Agriculture/animal science 2.6 1.3 2.0 Other** 16.5 4.7 11.0 Total 100.0 100.0 100.0 (Sample N) (188) (155) (343) Note: Percentages are weighted; sample sizes are unweighted totals. * The category of technology/engineering includes mining technology, processing, construction, and architecture. **Other category includes the fields of health, social service, tourism, transportation, environment, national security/ defense, and unspecified areas. 59 Table 2-4. Current School Enrollment among Children Aged 15-17 (Percent in school of any level)* Male Female All Gender gap (%) % in school N % in school N % in school N Total 71.5 1,617 75.5 1,463 73.4 3,080 -4.0 Area Urban 82.0 331 84.6 319 83.3 650 -2.6 Rural 68.5 1,286 72.7 1,144 70.4 2,430 -4.2 Region Red River Delta 84.0 313 84.6 250 84.3 563 -0.7 North East 76.1 245 81.9 198 78.7 443 -5.8 North West 75.0 80 56.9 96 65.4 176 18.0 North Central Coast 75.8 199 81.5 176 78.5 375 -5.7 South Central Coast 73.4 151 80.6 149 76.9 300 -7.2 Central Highlands 65.9 133 76.0 125 70.8 258 -10.1 South East 67.1 189 77.0 198 72.1 387 -9.9 Mekong River Delta 55.0 307 57.6 271 56.2 578 -2.6 Ethnicity Kinh/Chinese 73.5 1,322 77.5 1,182 75.4 2,504 -4.0 Ethnic minorities 59.0 295 63.5 281 61.2 576 -4.5 Tay/Thai/Muong/Nung 71.5 154 77.0 140 74.1 294 -5.5 Northern Mountain ethnic 58.0 60 43.4 69 50.6 129 14.6 Central ethnic 50.9 46 63.4 49 57.7 95 -12.5 Khmer/Cham 27.1 35 37.9 23 31.4 58 -10.8 Expenditure quintiles 1 poorest 48.8 337 54.4 325 51.5 662 -5.6 2 66.6 382 71.6 340 68.9 722 -5.1 3 74.9 358 81.5 306 77.9 664 -6.6 4 79.9 306 81.1 298 80.5 604 -1.1 5 wealthiest 88.8 234 92.2 194 90.3 428 -3.4 Mother's schooling** No schooling 51.9 155 44.6 165 48.0 320 7.4 1-4 years 51.1 333 61.2 285 55.6 618 -10.2 5-8 years 67.0 424 73.6 373 70.1 797 -6.5 9-11 years 82.0 460 87.4 444 84.7 904 -5.4 12+ years 96.5 209 93.4 176 95.1 385 3.1 Father's schooling** No schooling 37.4 70 46.7 71 42.2 141 -9.3 1-4 years 49.5 216 59.3 191 54.0 407 -9.8 5-8 years 57.2 407 67.4 354 61.9 761 -10.2 9-11 years 85.1 504 82.6 450 83.9 954 2.4 12+ years 91.9 229 95.0 240 93.4 469 -3.1 Note: Percentages are weighted; sample sizes are unweighted totals for each group. * Percent in school includes those attending vocational schools. ** For mother's and father's schooling, only those children with mothers or/and fathers in the household are included. 60 Table 2-5. Current School Enrollment among 18-21 Year Olds (Percent in school of any level)* Male Female All Gender gap (%) % in school N % in school N % in school N Total 40.9 1,774 41.7 1,581 41.3 3,355 -0.9 Area Urban 50.9 1,388 57.2 1,202 54.0 2,590 -6.4 Rural 37.8 386 36.4 379 37.1 765 1.4 Region Red River Delta 54.5 339 49.4 289 52.1 628 5.0 North East 43.4 250 45.3 279 44.4 529 -2.0 North West 40.5 112 26.9 103 33.9 215 13.6 North Central Coast 46.1 206 55.7 158 50.2 364 -9.6 South Central Coast 52.4 156 44.9 148 48.8 304 7.5 Central Highlands 31.9 126 41.5 112 36.5 238 -9.6 South East 34.7 227 41.2 178 37.6 405 -6.5 Mekong River Delta 24.3 358 25.0 314 24.6 672 -0.7 Ethnicity Kinh/Chinese 42.3 1,448 43.7 1,253 43.0 2,701 -1.4 Ethnic minorities 31.9 326 31.4 328 31.7 654 0.5 Tay/Thai/Muong/Nung 42.2 160 42.7 182 42.4 342 -0.5 Northern Mountain ethnic 27.7 69 12.5 83 18.8 152 15.2 Central ethnic 21.9 64 30.4 43 25.5 107 -8.5 Khmer/Cham 12.3 33 4.7 20 9.5 53 7.6 Expenditure quintiles 1 poorest 17.0 331 16.2 317 16.6 648 0.7 2 27.0 343 27.8 300 27.4 643 -0.8 3 40.6 408 36.2 301 38.7 709 4.3 4 51.8 400 55.9 359 53.7 759 -4.0 5 wealthiest 62.9 292 63.9 304 63.4 596 -1.0 Mother's schooling** No schooling 19.2 198 15.4 157 17.5 355 3.8 1-4 years 28.8 387 18.9 354 24.1 741 9.9 5-8 years 29.7 437 37.6 386 33.4 823 -7.9 9-11 years 51.7 538 54.1 448 52.7 986 -2.4 12+ years 74.8 188 77.2 197 76.0 385 -2.4 Father's schooling** No schooling 17.8 73 7.0 78 13.8 151 10.9 1-4 years 15.1 238 17.4 224 17.5 462 -2.3 5-8 years 23.3 456 30.6 376 30.1 832 -7.3 9-11 years 40.2 533 50.9 449 51.5 982 -10.7 12+ years 60.2 256 69.8 253 70.7 509 -9.7 Note: Percentages are weighted; sample sizes are unweighted totals for each group. * Percent in school includes those attending vocational schools. ** For mother's and father's schooling, only those children with mothers or/and fathers in the household are included. 61 Table 2-6. Attendance at Extra Classes (Only among students who currently attend from primary up to upper secondary schools) Male Female All Gender gap % in class N % in class N % in class N (%) Total 49.1 4,556 53.0 4,427 51.0 8,983 -3.9 Level of current schooling Primary 33.9 1,712 37.9 1,581 35.8 3,293 -4.1 Lower secondary 52.0 1,719 54.9 1,778 53.5 3,497 -2.9 Upper secondary 66.9 1,125 71.4 1,068 69.1 2,193 -4.5 Area Urban 67.2 979 71.8 909 69.4 1,888 -4.7 Rural 43.6 3,577 47.6 3,518 45.6 7,095 -4.0 Region Red River Delta 79.7 785 79.5 802 79.6 1,587 0.2 North East 32.7 700 39.0 668 35.8 1,368 -6.3 North West 12.1 286 15.5 229 13.6 515 -3.4 North Central Coast 53.6 609 56.7 585 55.1 1,194 -3.1 South Central Coast 52.2 461 58.5 435 55.2 896 -6.3 Central Highlands 36.7 436 45.2 421 40.9 857 -8.5 South East 59.1 534 57.9 575 58.5 1,109 1.2 Mekong River Delta 25.6 745 29.3 712 27.4 1,457 -3.7 Ethnicity Kinh/Chinese 56.1 3,596 59.4 3,549 57.7 7,145 -3.4 Ethnic minorities 13.0 960 16.5 878 14.6 1,838 -3.5 Tay/Thai/Muong/Nung 15.4 494 22.2 489 18.7 983 -6.8 N. Mountain ethnic 15.4 205 6.9 157 11.7 362 8.5 Central ethnic 7.1 196 12.3 177 9.6 373 -5.2 Khmer/Cham 9.5 65 6.1 55 7.9 120 3.4 Expenditure quintiles 1 poorest 23.6 1,131 30.6 1,142 27.2 2,273 -7.0 2 42.6 1,016 48.7 1,067 45.7 2,083 -6.0 3 55.3 964 55.7 896 55.5 1,860 -0.5 4 59.7 850 65.2 767 62.3 1,617 -5.5 5 wealthiest 73.5 595 77.1 555 75.3 1,150 -3.7 Mother's schooling* No schooling 17.8 508 21.8 406 19.6 914 -4.0 1-4 years 29.3 805 36.6 828 33.0 1,633 -7.3 5-8 years 44.4 1,276 49.1 1,267 46.8 2,543 -4.6 9-11 years 64.5 1,245 64.0 1,281 64.2 2,526 0.4 12+ years 71.1 612 75.8 557 73.3 1,169 -4.7 Father's schooling* No schooling 12.0 230 16.2 204 14.0 434 -4.2 1-4 years 23.7 609 28.0 580 25.8 1,189 -4.3 5-8 years 36.1 1,140 46.2 1,152 41.2 2,292 -10.1 9-11 years 61.2 1,370 61.9 1,345 61.6 2,715 -0.7 12+ years 70.8 709 72.1 695 71.4 1,404 -1.3 Note: Percentages are weighted; sample sizes are unweighted totals for each group. * For mother's and father's schooling, only those children with mothers or/and fathers in the household are included 62 Table 2-7. Average Expenditure on Extra Classes for Past 12 Months (Per student in VND in thousand) Male Female All Gender gap VND N VND N VND N Total 417.91 1,742 425.05 1,824 421.56 3,566 -7.1 Level of schooling Primary school 301.30 454 291.64 453 296.52 907 9.7 Lower secondary school 373.12 663 342.61 748 357.01 1,411 30.5 Upper secondary school 554.07 625 621.63 623 588.00 1,248 -67.6 Area Urban 803.04 515 790.56 502 796.95 1,017 12.5 Rural 243.59 1,227 279.08 1,322 262.04 2,549 -35.5 Expenditure quintiles 1 poorest 133.13 186 130.26 252 131.46 438 2.9 2 184.98 361 193.11 421 189.32 782 -8.1 3 251.64 432 263.54 422 257.49 854 -11.9 4 350.24 420 393.21 386 370.84 806 -43.0 5 wealthiest 1,030.10 343 1,079.54 343 1,054.75 686 -49.4 Note: Average expenditures are calculated only among students who attended extra classes, excluding zero values; sample weights are used for calculation. (Sample sizes are unweighted totals.) 63 Table 3-1. Labor Force Participation Rates in Past 12 Months (% worked)* 2006 VHLSS All Areas Urban Rural Ages Male Female Male Female Male Female 6-10 1.5 1.1 0.3 0.0 1.8 1.4 11-14 11.6 11.0 3.3 3.7 14.0 13.0 15-17 33.2 28.6 14.6 14.4 38.4 33.0 18-24 68.1 65.7 55.9 52.0 72.3 71.0 25-34 95.9 92.3 93.5 86.2 96.9 94.6 35-44 97.8 94.0 95.8 88.8 98.6 96.0 45-54 95.0 90.8 91.4 84.8 96.7 93.5 55-64 80.9 68.4 67.3 47.5 86.8 79.1 65+ 44.3 30.6 28.0 17.8 50.7 35.0 All ages 6+ 64.8 61.0 60.7 55.4 66.3 63.1 (Sample N) (17,664) (18,511) (4,311) (4,551) (13,353) (13,960) Ages 15-64 only 81.3 78.2 76.0 69.3 83.3 81.8 Ages 25-64 only 94.3 88.9 90.0 80.2 96.1 92.6 2004 VHLSS** All Areas Urban Rural Ages Male Female Male Female Male Female 6-10 3.8 3.3 0.3 0.2 4.6 4.0 11-14 17.0 18.4 7.1 3.9 19.7 22.3 15-17 41.6 41.5 22.3 18.4 46.6 48.3 18-24 68.2 66.8 48.2 50.9 74.8 73.2 25-34 96.8 92.2 94.4 86.4 97.7 94.4 35-44 98.2 94.0 96.6 88.6 98.8 96.1 45-54 95.5 89.1 93.6 81.1 96.3 92.6 55-64 79.4 65.9 62.9 46.1 86.8 75.1 65+ 42.6 32.1 27.8 20.0 47.7 36.2 All ages 6+ 64.5 61.7 60.2 54.8 66.0 64.1 (Sample N) (18,379) (18,989) (4,268) (4,660) (14,111) (14,389) Ages 15-64 only 82.6 79.5 75.9 69.2 85.1 83.6 Ages 25-64 only 94.9 88.5 90.8 79.9 96.5 92.1 Note: Percentages are weighted; sample sizes are unweighted totals. *The labor force participation rate here refers to the rate of being employed at some point during the past 12 months; the rate does not include people who were looking for a job or unemployed. ** Results from 2004 VHLSS are author's own calculations. 64 Table 3-2. Number of Weeks Worked on Income-Generating Activities Per Year* ALL AREAS Male Female Mean weeks Mean weeks Age % No work worked % No work worked 6-10 98.5 10.1 98.9 12.8 11-14 88.4 17.4 89.0 15.8 15-17 66.9 26.2 71.4 26.6 18-24 31.9 36.4 34.3 36.7 25-34 4.1 43.1 7.7 40.7 35-44 2.2 42.7 6.0 41.6 45-54 5.0 40.6 9.2 39.5 55-64 19.1 32.5 31.6 30.8 65+ 55.7 24.1 69.4 24.0 Total 35.2 38.3 39.0 37.4 (Sample N) (17,664) (11,487) (18,511) (11,408) URBAN AREAS Male Female Mean weeks Mean weeks Age % No work worked % No work worked 6-10 99.7 n/a 100.0 n/a 11-14 96.7 17.4 96.3 17.3 15-17 85.4 32.4 85.6 33.2 18-24 44.1 42.6 48.0 42.5 25-34 6.5 49.3 13.8 47.2 35-44 4.3 50.1 11.2 49.6 45-54 8.6 47.4 15.3 47.4 55-64 32.7 40.7 52.5 41.4 65+ 72.0 32.6 82.2 34.9 Total 39.3 46.4 44.6 46.1 (Sample N) (4,311) (2,622) (4,551) (2,553) 2004 Total 39.8 45.3 45.2 45.5 RURAL AREAS Male Female Mean weeks Mean weeks Age % No work worked % No work worked 6-10 98.2 9.0 98.6 12.8 11-14 86.0 17.4 87.0 15.7 15-17 61.6 25.6 67.0 25.7 18-24 27.7 34.7 29.0 35.1 25-34 3.1 40.6 5.4 38.4 35-44 1.4 40.0 4.0 38.7 45-54 3.3 37.5 6.5 36.2 55-64 13.2 29.7 21.0 27.6 65+ 49.3 22.2 65.0 22.0 Total 33.7 35.6 36.9 34.6 (Sample N) (13,353) (8,865) (13,960) (8,855) 2004 Total 34.1 34.2 35.9 33.5 Note: Mean weeks are weighted; sample sizes are unweighted totals. *The number of weeks is calculated only among those who did any work, excluding zero values. Data on the number of weeks worked come from information on the two main jobs worked during the past year. The survey includes questions on jobs other than the two main jobs, but they do not specify the number of weeks worked. Weeks are calculated based on 8 hours of work per day, and 6 days of work per week. 65 Table 3-3. Average Number of Hours Spent on Housework (per day) ALL AREAS Male Female Age % None Mean hrs % None Mean hrs (per day) (per day) 6-10 84.6 1.3 76.7 1.3 11-14 57.2 1.4 41.0 1.5 15-17 41.2 1.4 24.9 1.7 18-24 51.7 1.4 25.0 2.0 25-34 38.6 1.4 6.6 2.4 35-44 29.9 1.5 4.5 2.4 45-54 28.4 1.6 4.0 2.4 55-64 27.6 1.6 6.9 2.5 65+ 42.4 1.7 33.6 2.2 Total 43.7 1.5 21.3 2.2 (Sample N) (17,664) (10,037) (18,511) (14,594) URBAN AREAS Male Female Age % None Mean hrs % None Mean hrs (per day) (per day) 6-10 92.6 1.3 82.4 1.2 11-14 69.6 1.3 47.9 1.5 15-17 46.8 1.3 25.7 1.6 18-24 61.2 1.3 24.6 2.0 25-34 47.4 1.4 7.6 2.5 35-44 38.2 1.6 5.1 2.7 45-54 34.8 1.7 5.1 2.7 55-64 30.1 1.8 8.4 2.9 65+ 48.9 1.8 39.6 2.5 Total 50.1 1.5 21.0 2.4 (Sample N) (4,311) (2,153) (4,551) (3,586) 2004 Total 52.4 1.6 23.0 2.4 RURAL AREAS Male Female Age % None Mean hrs % None Mean hrs (per day) (per day) 6-10 82.3 1.3 75.2 1.3 11-14 53.6 1.4 39.1 1.5 15-17 39.7 1.4 24.6 1.7 18-24 48.4 1.4 25.1 2.0 25-34 35.1 1.5 6.2 2.3 35-44 26.9 1.5 4.3 2.3 45-54 25.3 1.5 3.5 2.2 55-64 26.5 1.6 6.1 2.3 65+ 39.8 1.7 31.5 2.1 Total 41.4 1.5 21.4 2.1 (Sample N) (13,353) (7,884) (13,960) (11,008) 2004 Total 43.0 1.5 21.5 2.1 Note: Mean number of hours spent on housework is calculated only among those who did any housework, excluding zero values. Mean hours presented are weighted. 66 Table 3-4. Types of Employment for All Jobs of Past 12 Months among Men and Women (Age 18-64; %) 2006 VHLSS All areas Urban Rural Men Women Men Women Men Women Wage employment only 27.2 20.0 54.1 42.4 17.4 12.2 Self-employment only 45.3 64.7 36.0 49.8 48.7 69.9 Both 27.5 15.3 9.9 7.8 34.0 17.9 Total 100.0 100.0 100.0 100.0 100.0 100.0 (Sample N) (10,122) (10,147) (2,464) (2,416) (7,658) (7,731) 2004 VHLSS All areas Urban Rural Men Women Men Women Men Women Wage employment only 25.5 17.9 53.7 40.6 15.8 10.1 Self-employment only 45.5 65.8 35.7 50.2 48.8 71.2 Both 29.0 16.3 10.6 9.2 35.3 18.7 Total 100.0 100.0 100.0 100.0 100.0 100.0 (Sample N) (10,212) (10,146) (2,381) (2,384) (7,831) (7,762) Note: Percentages are weighted; sample sizes are unweighted totals. 67 Table 3-5. Type of Employment for Main Job Held in Past 12 Months among Men and Women (Age 18-64; %) Men Women Non- Non- Wage Agricultural agricultural Wage Agricultural agricultural employment self-emp self-emp Total (Sample N) employment self-emp self-emp Total (Sample N) TOTAL 42.1 38.3 19.7 100.0 (10,122) 27.4 47.1 25.5 100.0 (10,147) Area Urban 60.8 9.3 30.0 100.0 (2,464) 47.6 11.5 41.0 100.0 (2,416) Rural 35.3 48.9 15.9 100.0 (7,658) 20.3 59.6 20.0 100.0 (7,731) Region Red River Delta 50.6 26.2 23.2 100.0 (1,908) 29.2 42.6 28.3 100.0 (2,087) North East 30.4 57.4 12.2 100.0 (1,533) 16.7 70.2 13.1 100.0 (1,601) North West 19.4 72.5 8.1 100.0 (552) 12.9 80.1 7.0 100.0 (565) North Central Coast 33.6 53.3 13.1 100.0 (1,034) 13.7 68.3 18.0 100.0 (1,060) South Central Coast 47.7 32.1 20.2 100.0 (911) 30.0 42.0 28.0 100.0 (891) Central Highlands 28.7 50.7 20.6 100.0 (665) 20.5 58.9 20.6 100.0 (664) South East 55.5 20.4 24.1 100.0 (1,346) 48.0 18.9 33.1 100.0 (1,232) Mekong River Delta 39.7 38.3 22.0 100.0 (2,173) 28.4 39.4 32.1 100.0 (2,047) Ethnicity Kinh/Chinese 45.5 32.9 21.6 100.0 (8,346) 29.8 41.8 28.5 100.0 (8,353) Ethnic minorities 19.8 73.3 6.9 100.0 (1,776) 11.8 82.6 5.6 100.0 (1,794) Tay and others 16.4 77.6 6.1 100.0 (981) 9.9 84.6 5.5 100.0 (995) N. Mountain ethnic 8.0 89.0 3.0 100.0 (336) 2.7 95.0 2.3 100.0 (362) Central ethnic 23.2 70.7 6.1 100.0 (294) 11.3 87.5 1.2 100.0 (291) Khmer/Cham 45.5 37.2 17.3 100.0 (165) 38.5 42.0 19.4 100.0 (146) (continued) 68 Table 3-5. Type of Employment for Main Job Held in Past 12 Months among Men and Women (Age 18-64; %) (continued) Men Women Non- Non- Wage Agricultural agricultural Wage Agricultural agricultural employment self-emp self-emp Total (Sample N) employment self-emp self-emp Total (Sample N) URBAN AREAS Age 18-24 73.0 8.7 18.3 100.0 (375) 72.3 8.0 19.8 100.0 (338) 25-34 71.7 6.1 22.2 100.0 (576) 61.7 9.3 29.0 100.0 (541) 35-44 55.6 8.1 36.3 100.0 (669) 37.5 10.6 52.0 100.0 (691) 45-54 54.7 10.3 35.0 100.0 (633) 42.1 11.9 45.9 100.0 (645) 55-64 42.6 19.9 37.6 100.0 (211) 19.0 24.8 56.2 100.0 (201) Marital Status Married 55.7 10.3 34.0 100.0 (1,816) 41.7 13.3 45.0 100.0 (1,738) Widowed x x x x (12) 32.9 11.1 56.0 100.0 (129) Divorced/separated 78.4 1.5 20.1 100.0 (41) 51.0 8.0 41.0 100.0 (83) Unmarried 74.2 6.2 19.6 100.0 (595) 71.7 5.8 22.6 100.0 (466) Education No schooling x x x x (28) 16.8 37.5 45.7 100.0 (61) Less than primary 42.3 21.7 36.0 100.0 (173) 29.8 20.0 50.2 100.0 (237) Primary 50.1 13.6 36.4 100.0 (577) 30.1 15.5 54.4 100.0 (550) Lower secondary 53.0 11.4 35.6 100.0 (659) 35.2 15.1 49.7 100.0 (611) Upper secondary 63.5 5.1 31.4 100.0 (646) 60.7 5.5 33.8 100.0 (650) Junior college/university 92.2 0.4 7.4 100.0 (381) 89.2 0.2 10.6 100.0 (307) (continued) 69 Table 3-5. Type of Employment for Main Job Held in Past 12 Months among Men and Women (Age 18-64; %) (continued) Men Women Non- Non- Wage Agricultural agricultural Wage Agricultural agricultural employment self-emp self-emp Total (Sample N) employment self-emp self-emp Total (Sample N) RURAL AREAS Age 18-24 46.2 45.7 8.1 100.0 (1,591) 39.7 47.4 12.9 100.0 (1,384) 25-34 43.1 40.8 16.1 100.0 (1,756) 24.0 54.8 21.3 100.0 (1,804) 35-44 33.8 47.1 19.1 100.0 (2,081) 18.3 60.2 21.6 100.0 (2,077) 45-54 27.4 54.2 18.4 100.0 (1,536) 10.7 66.2 23.1 100.0 (1,737) 55-64 13.6 68.9 17.6 100.0 (694) 5.6 76.3 18.1 100.0 (729) Marital Status Married 31.2 50.9 17.9 100.0 (5,842) 31.2 50.9 17.9 100.0 (5,889) Widowed 12.9 68.2 18.9 100.0 (51) 12.9 68.2 18.9 100.0 (444) Divorced/separated 46.4 44.0 9.6 100.0 (45) 46.4 44.0 9.6 100.0 (166) Unmarried 49.3 41.6 9.1 100.0 (1,720) 49.3 41.6 9.1 100.0 (1,232) Education No schooling 29.7 63.1 7.3 100.0 (404) 18.3 73.6 8.2 100.0 (729) Less than primary 31.1 55.3 13.6 100.0 (1,155) 15.7 62.3 22.1 100.0 (1,452) Primary 30.5 54.4 15.1 100.0 (2,260) 15.2 62.5 22.3 100.0 (2,266) Lower secondary 33.1 48.3 18.5 100.0 (2,539) 14.9 63.7 21.4 100.0 (2,289) Upper secondary 47.2 35.6 17.2 100.0 (1,150) 43.4 38.8 17.8 100.0 (831) Junior college/university 86.4 7.1 6.5 100.0 (150) 89.5 4.8 5.8 100.0 (164) Note: Percentages are weighted; sample sizes are unweighted totals. 70 Table 3-6. Type of Industry for Main Job among Men and Women (age 18-64; %) Urban Rural Men Women Men Women All Adults 18-64 Primary 13.5 13.7 66.0 59.2 Agriculture/forestry 11.1 12.8 53.9 64.5 Aquaculture 2.7 0.7 5.3 1.6 Secondary 33.3 23.7 22.7 13.5 Mining 1.3 0.4 1.1 0.3 Food/beverage manufacturing 3.4 4.4 2.3 2.9 Textiles/garments production 3.7 10.5 1.1 5.0 Wood/paper manufacturing 2.3 1.9 1.4 2.2 Other production/processing 10.5 4.4 5.9 2.2 Construction/utilities 12.1 2.2 10.9 0.8 Tertiary Trades 15.8 25.6 6.8 10.9 Vehicle sales/repairs 2.7 0.6 0.9 0.1 Wholesale & agenet sales 3.3 3.5 1.0 0.9 Retail sales 9.8 21.4 5.0 9.9 Services 37.2 37.3 11.3 9.7 Hotels/restaurants 4.6 12.2 0.8 3.0 Transportation & communications 13.1 1.9 3.6 0.3 Business & financial services 9.7 5.8 3.3 0.8 Education, health & cultural services 7.4 12.8 2.9 4.5 Sanitation & personal services 2.4 4.5 0.7 1.0 Total 100.0 100.0 100.0 100.0 (Sample N) (3,188) (2,976) (8,770) (8,474) (continued) 71 Table 3-6. Types of Industry for Main Job among Men and Women (age 18-64; %) (continued) Urban Rural Men Women Men Women Adults in Wage Employment Primary 4.5 3.2 19.8 21.5 Secondary 43.4 35.7 53.2 41.0 Mining 2.0 0.6 2.4 0.8 Food/beverage manufacturing 4.0 5.8 3.0 6.0 Textiles/garments production 4.2 14.9 2.6 18.5 Wood/paper manufacturing 2.7 2.7 2.8 4.0 Other production/processing 12.3 7.5 12.6 8.7 Construction/utilities 18.3 4.1 29.9 3.0 Tertiary Trades 10.2 10.1 4.9 4.9 Vehicle sales/repairs 1.8 0.6 1.2 0.3 Wholesale & agent sales 2.9 3.3 1.0 0.9 Retail sales 5.5 6.3 2.7 3.8 Services 42.0 51.0 22.2 32.6 Hotels/restaurants 2.7 5.7 0.8 2.5 Transportation & communications 11.9 3.3 4.3 1.2 Business & financial services 14.4 11.2 8.7 3.7 Education, health & cultural services 11.1 25.6 7.6 21.7 Sanitation & personal services 1.9 5.2 0.9 3.5 Total 100.0 100.0 100.0 100.0 (sample N) (1,475) (1,106) (2,573) (1,483) Adults in Nonagricultural Self-Employment Primary 6.3 1.7 21.6 10.8 Secondary 22.8 16.2 24.8 25.3 Mining 0.4 0.3 1.5 0.9 Food/beverage manufacturing 3.0 3.9 7.6 8.3 Textiles/garments production 3.7 8.1 1.4 5.9 Wood/paper manufacturing 2.2 1.6 2.6 7.1 Other production/processing 10.2 1.9 9.4 2.3 Construction/utilities 3.3 0.5 2.2 0.7 Tertiary Trades 32.0 50.4 31.8 48.8 Vehicle sales/repairs 5.3 0.8 3.0 0.1 Wholesale and agent sales 5.3 4.8 3.8 3.5 Retail sales 21.3 44.8 25.0 45.2 Services 39.0 31.8 21.8 15.2 Hotels/restaurants 9.8 23.2 3.4 12.5 Transportation & communications 19.7 0.8 13.0 0.5 Business & financial services 3.1 1.3 1.4 0.4 Education, health & cultural services 2.3 1.5 1.3 0.2 Sanitation & personal services 4.0 5.0 2.8 1.6 Total 100.0 100.0 100.0 100.0 (sample N) (727) (981) (1,149) (1,460) Note: Percentages are weighted; sample sizes are unweighted totals. 72 Table 3-7. Characteristics of Male-Operated and Female-Operated Nonagricultural Household Businesses Urban Areas Rural Areas Male- Female- Male- Female- operated operated Total operated operated Total (No. of businesses) (626) (930) (1,556) (1,205) (1,617) (2,822) % with business license 35.4 31.6 33.1 22.4 14.3 17.8 Number of laborers (%) 1 only 54.1 66.4 61.4 57.4 73.3 66.4 2-3 33.2 29.5 31.0 34.0 24.7 28.7 4-5 6.3 2.5 4.0 5.0 1.9 3.2 6-10 4.7 1.1 2.6 2.5 0.2 1.2 11-36 1.8 0.5 1.0 1.1 0.0 0.5 Total 100.0 100.0 100.0 100.0 100.0 100.0 Average number of laborers 2.3 1.6 1.8 1.9 1.4 1.6 % with paid laborer 24.3 13.2 17.7 16.5 6.0 10.5 Place of business activities (%) Home 47.3 52.0 50.1 57.9 59.7 58.9 Industrial zone/trade center 0.8 1.0 0.9 1.0 0.4 0.7 Markets 5.2 22.8 15.6 8.7 22.7 16.7 Other shops/permanent places 22.2 15.2 18.0 11.1 6.8 8.6 Non-permanent place 24.6 9.0 15.4 21.2 10.4 15.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 Average number of months in operation 11.1 11.1 11.1 9.6 10.0 9.8 Monthly revenue (VND in thousand) Mean 7,948.80 4,068.24 5,641.44 4,285.61 1,688.35 2,817.38 Median 2,560.00 1,800.00 2,000.00 1,500.00 900.00 1,100.00 Note: Sample sizes indicate unweighted totals of nonagricultural businesses operated at the household level. Since some households have more than one business activity, the sample size indicates the total number of household businesses, not the number of individuals. Statistics presented here are calculated using sample weights. 73 Table 3-8. Categories of Occupation for Main Job among Men and Women (Age 18-64; %) Urban Rural Men Women Men Women Total Administrative/managerial 3.5 1.0 2.3 0.7 Professional - science/health/technical 13.4 9.9 1.7 1.0 Professional - education related 3.1 7.5 1.3 3.2 Other professional/armed forces 4.1 4.0 1.3 0.8 Services - skilled 3.0 2.2 1.0 0.5 Sales - skilled 2.3 6.9 1.1 2.7 Agriculture/forestry/fishery 13.6 13.7 58.6 65.9 Skilled manual workers 28.7 14.9 16.4 7.6 Unskilled manual workers 28.4 39.9 16.5 17.6 Total 100.0 100.0 100.0 100.0 (sample N) (2,464) (2,416) (7,658) (7,731) Wage Employment Only Administrative/managerial 5.2 1.9 6.2 3.3 Professional - science/health/technical 21.0 20.1 4.4 4.7 Professional - education related 5.0 15.8 3.6 15.7 Other professional/armed forces 6.5 8.3 3.6 3.7 Services - skilled 3.6 1.8 2.3 1.4 Sales - skilled 1.1 3.2 0.4 1.9 Agriculture/forestry/fishery 4.2 3.5 18.6 21.6 Skilled manual workers 31.2 20.8 33.5 21.9 Unskilled manual workers 22.1 24.7 27.5 25.8 Total 100.0 100.0 100.0 100.0 (sample N) (1,475) (1,106) (2,573) (1,483) Nonagricultural Self-Employment Only Administrative/managerial 1.2 0.2 0.4 0.0 Professional - science/health/technical 2.1 0.8 0.8 0.3 Professional - education related 0.0 0.1 0.2 0.1 Other professional/armed forces 0.5 0.2 0.0 0.1 Services - skilled 2.5 3.3 1.0 1.1 Sales - skilled 5.4 13.1 5.9 11.7 Agriculture/forestry/fishery 5.8 1.3 20.4 9.6 Skilled manual workers 32.4 12.2 28.6 15.5 Unskilled manual workers 50.1 68.8 42.7 61.6 Total 100.0 100.0 100.0 100.0 (sample N) (727) (981) (1,149) (1,460) Note: Percentages are weighted; sample sizes are unweighted totals. 74 Table 3-9. Mean Hourly Wages of the Main Job in Wage Employment (VND in thousand) Urban Areas Rural Areas Men Women Gender Men Women Gender wage wage Wage (N) % Wage (N) % gap Wage (N) % Wage (N) % gap All ages 6+ 9.453 (1,523) 8.198 (1,144) 86.7 5.925 (2,776) 5.164 (1,648) 87.2 All ages 15+ 9.467 (1,519) 8.206 (1,140) 86.7 5.953 (2,743) 5.218 (1,619) 87.7 Adults ages 18-64 9.593 (1,475) 8.314 (1,106) 86.7 6.076 (2,573) 5.413 (1,483) 89.1 Sector of employment (All ages 15+) Government 12.853 (397) 24.3 10.283 (369) 29.6 80.0 8.707 (372) 12.8 9.306 (305) 18.5 106.9 State owned enterprises (SOE) 10.638 (258) 17.6 8.568 (183) 16.5 80.5 5.785 (238) 8.4 5.387 (144) 8.9 93.1 Private enterprises 7.261 (803) 53.5 6.360 (507) 46.3 87.6 5.328 (2,059) 75.6 3.949 (1,019) 62.6 74.1 Foreign invested enterprises (FDI) 12.825 (61) 4.5 10.573 (81) 7.7 82.4 10.109 (74) 3.2 5.468 (151) 10.0 54.1 Total (1,519) 100.0 (1,140) 100.0 (2,743) 100.0 (1,619) 100.0 Occupation (All ages 15+) Administrative/managerial 12.705 (98) 5.5 9.135 (27) 1.9 71.9 5.323 (165) 6.1 5.291 (52) 3.0 99.4 Science/technology/medical professional 13.987 (284) 20.5 12.623 (216) 19.5 90.3 10.618 (114) 4.1 6.387 (68) 4.3 60.2 Education professionals 13.944 (75) 5.0 11.450 (186) 15.5 82.1 12.014 (98) 3.3 9.427 (232) 14.4 78.5 Other professionals 12.754 (93) 6.3 9.274 (85) 8.0 72.7 7.131 (97) 3.4 7.147 (54) 3.4 100.2 Services 7.879 (65) 3.8 5.507 (24) 1.8 69.9 4.051 (60) 2.2 4.044 (23) 1.4 99.8 Sales 6.064 (17) 1.2 7.634 (35) 3.1 125.9 6.632 (12) 0.5 4.375 (29) 1.9 66.0 Agriculture/forestry/fishery 6.942 (78) 4.2 4.464 (50) 3.8 64.3 4.909 (574) 19.5 4.215 (383) 22.1 85.9 Skilled manual workers 7.836 (465) 30.9 6.086 (217) 20.7 77.7 6.085 (875) 32.7 4.557 (355) 22.6 74.9 Unskilled workers (excluding agriculture) 5.857 (344) 22.7 5.006 (300) 25.6 85.5 5.252 (748) 28.2 4.022 (423) 27.0 76.6 Total (1,519) 100.0 (1,140) 100.0 (2,743) 100.0 (1,619) 100.0 Industry (All ages 15+) Primary 7.192 (83) 4.4 5.169 (45) 3.4 71.9 5.003 (606) 20.6 4.223 (381) 22.1 84.4 Secondary 8.533 (644) 43.6 6.972 (387) 35.7 81.7 5.889 (1,400) 52.8 4.426 (653) 41.2 75.2 Tertiary Trades 7.685 (150) 10.3 6.954 (110) 10.2 90.5 5.549 (134) 5.0 4.450 (76) 5.0 80.2 Services 11.127 (642) 41.7 9.531 (598) 50.7 85.7 7.110 (603) 21.6 7.058 (509) 31.7 99.3 Total (1,519) 100.0 (1,140) 100.0 (2,743) 100.0 (1,619) 100.0 Education completed (All ages 15+) No schooling 4.705 (17) 0.9 3.556 (16) 1.1 75.6 4.357 (111) 3.9 3.460 (117) 7.0 79.4 Less than primary 5.585 (80) 4.9 4.189 (71) 6.0 75.0 4.938 (368) 13.1 4.022 (245) 14.7 81.4 Primary 6.089 (309) 20.1 5.026 (169) 15.1 82.5 5.360 (708) 25.8 4.323 (381) 23.2 80.6 Lower secondary 7.409 (355) 23.0 5.896 (222) 18.4 79.6 5.416 (883) 32.4 4.578 (370) 23.4 84.5 Upper secondary 9.873 (404) 26.9 8.113 (386) 34.4 82.2 6.847 (540) 20.0 5.917 (359) 22.4 86.4 Junior college/university 14.758 (354) 24.2 13.104 (276) 25.1 88.8 12.983 (133) 4.9 10.579 (147) 9.3 81.5 Total (1,519) 100.0 (1,140) 100.0 (2,743) 100.0 (1,619) 100.0 75 Table 4-1. Gender and Health Status All Respondents All Respondents (in past 12 months) % Absent from % Ill in past 4 % Ill in past 12 % Bedridden and school/work due to weeks months needed assistance illness Male Female Male Female Male Female Male Female (Sample N) (19,157) (19,914) (19,157) (19,914) (19,157) (19,914) (19,157) (19,914) Total 17.3 20.5 49.0 56.2 28.5 33.3 9.8 11.6 Age 0-4 33.4 32.0 63.9 65.5 35.0 34.2 21.2 20.7 5-9 20.5 18.7 56.0 58.3 32.6 32.1 11.2 9.4 10-14 14.4 13.3 47.3 47.4 27.8 26.4 7.6 6.8 15-19 10.0 10.9 38.6 41.4 21.1 22.5 5.6 6.0 20-29 9.4 11.8 35.9 42.6 19.1 24.4 5.3 8.8 30-39 13.7 16.5 44.3 54.0 24.1 31.8 6.3 9.2 40-49 17.7 21.7 50.3 60.1 28.9 35.9 9.0 10.0 50-59 20.8 30.0 57.6 69.3 33.9 42.2 11.4 14.1 60+ 34.7 38.6 72.1 76.9 50.7 52.6 23.5 24.8 Area Urban 18.6 22.5 55.5 62.7 24.1 28.4 9.4 11.1 Rural 16.9 19.8 46.8 53.9 30.0 35.1 10.0 11.7 Region Red River Delta 14.7 17.8 44.2 52.3 29.7 35.5 10.7 12.2 North East 15.7 18.0 40.9 47.1 30.6 34.8 9.4 11.3 North West 12.8 16.9 34.0 41.5 26.7 33.6 10.7 12.9 North Central Coast 14.4 17.7 39.6 48.0 30.7 35.5 11.3 12.5 South Central Coast 12.7 17.5 40.1 50.2 23.2 29.5 9.6 12.3 Central Highlands 25.1 25.4 58.6 62.7 34.6 38.7 11.6 14.2 South East 21.6 25.0 62.9 68.0 20.6 24.3 7.2 8.8 Mekong River Delta 20.2 23.5 57.9 64.3 31.5 35.6 9.6 11.5 Ethnicity Kinh/Chinese 17.5 20.8 50.3 57.7 28.3 33.3 9.6 11.5 Ethnic minorities 16.2 18.8 41.3 46.3 29.3 33.6 11.0 12.1 Tay/Thai/Muong/Nung 16.5 21.1 39.2 45.9 29.1 34.7 11.2 12.3 N. Mountain ethnic 11.5 11.9 30.6 36.9 24.6 30.1 7.7 11.0 Central ethnic 20.3 19.5 53.1 54.5 38.0 39.2 15.2 15.1 Khmer/Cham 13.5 18.2 43.9 47.9 19.7 22.8 6.7 6.9 (continued) 76 Table 4-1. Gender and Health Status (continued) All Respondents All Respondents (in past 12 months) % Absent from % Ill in past 4 % Ill in past 12 % Bedridden and school/work due to weeks months needed assistance illness Male Female Male Female Male Female Male Female (Sample N) (19,157) (19,914) (19,157) (19,914) (19,157) (19,914) (19,157) (19,914) Expenditure quintile 1 Lowest 15.9 18.1 42.0 47.5 28.3 32.1 9.9 12.2 2 17.1 19.3 46.2 52.3 31.0 34.6 10.2 11.1 3 16.1 20.4 48.2 56.0 29.8 35.7 9.3 11.8 4 18.4 21.6 51.9 61.1 28.6 33.7 9.8 10.9 5 Highest 19.2 23.4 56.9 64.4 24.8 30.4 10.0 11.9 Expenditure quintile (Standardized) 1 Lowest 17.7 19.7 45.6 51.4 31.7 35.3 11.2 13.0 2 18.9 20.7 48.8 55.1 33.1 36.2 11.5 12.5 3 19.1 21.9 53.5 59.4 32.8 36.6 12.0 12.7 4 23.2 24.0 60.1 66.5 33.4 36.0 12.9 12.5 5 Highest 25.4 25.5 68.8 73.9 28.9 33.6 12.0 13.7 Schooling completed (Age 15-49) No schooling 14.7 16.3 44.2 50.5 29.5 35.3 10.0 12.4 Less than primary 16.5 19.2 48.8 57.6 28.3 34.5 8.2 10.4 primary 14.1 17.1 43.3 51.7 23.5 28.8 6.7 9.1 Lower second 12.2 15.6 41.8 48.9 24.6 30.6 6.6 8.8 Upper second 9.8 12.0 37.3 45.9 18.8 23.6 5.0 6.1 JC/university 10.9 12.5 45.2 49.0 15.7 22.7 6.5 7.7 (Total) 12.6 15.6 42.1 50.1 23.2 29.1 6.5 8.6 Note: Percentages are weighted; sample sizes are unweighted totals. 77 Table 4-2. Access to Health Care among People Reporting Illness in Past 12 Months % Visiting any Health Care Worker/Center in Past 12 months among People Reporting Absence from Normal Activities due to Illness Total Urban Rural Male Female Male Female Male Female (Sample N) (5,546) (6,719) (1,173) (1,441) (4,373) (5,278) Total 77.2 78.0 74.8 77.8 77.8 78.1 Age 0-4 82.9 83.1 82.8 84.4 82.9 82.7 5-9 77.2 69.9 74.2 77.9 78.0 68.0 10-14 72.0 64.6 65.9 63.4 73.6 64.9 15-19 69.9 69.2 74.4 66.4 69.0 70.0 20-29 70.4 77.4 62.7 76.6 73.0 77.7 30-39 73.4 77.7 67.4 77.4 75.1 77.8 40-49 77.6 78.0 75.3 74.0 78.3 79.2 50-59 81.2 84.6 80.9 81.5 81.3 85.7 60+ 86.9 85.5 85.6 88.3 87.3 84.7 Ethnicity Kinh/Chinese 78.0 78.3 74.9 77.7 79.0 78.5 Ethnic minorities 72.1 76.1 73.6 80.5 72.0 75.7 Tay/Thai/Muong/Nung 67.2 72.3 * * 66.5 71.9 Northern Mountain ethnic 66.7 75.7 * * 67.7 75.3 Central ethnic 82.4 82.9 * * 83.7 83.9 Khmer/Cham 77.8 81.4 * * 77.7 79.9 Expenditure quintile 1 Lowest 72.1 74.0 67.5 75.5 72.5 73.9 2 76.9 77.4 84.9 75.2 76.2 77.6 3 77.9 79.2 74.2 80.3 78.5 79.1 4 81.4 79.1 75.2 77.1 83.7 79.9 5 Highest 77.4 80.5 74.1 78.1 82.8 84.3 Type of health insurance (HI) None 72.4 75.8 67.3 73.3 73.6 76.3 HI for children under 6 81.0 83.5 77.5 92.1 82.0 80.7 HI for the poor 79.2 80.2 78.9 88.4 79.3 79.1 HI for policy beneficiaries 90.0 88.2 91.0 91.3 89.7 87.5 Required state HI 81.9 82.3 80.2 81.6 83.2 83.0 Required non-state HI 85.6 84.2 * 82.2 * 85.9 Student HI 73.7 66.4 70.4 69.4 75.1 65.0 Other voluntary HI 87.7 87.6 82.9 85.5 90.2 88.8 Health card 81.8 82.2 82.9 79.4 81.6 82.6 Note: Percentages are weighted; sample sizes are unweighted totals. * Sample size is too small to report. 78 Table 4-3. Types of Health Care Services Utilized in Past 12 Months (%)* Urban Areas Type of health service Total Male Female Village health center 0.4 0.5 0.4 Commune health center 8.1 8.0 8.2 Regional general clinics 4.7 4.4 5.0 District hospital 20.3 19.8 20.6 Provincial hospital 23.7 24.3 23.2 Center hospital 6.0 5.6 6.3 Other state-owned hospital 2.7 3.4 2.2 Private hospital 3.0 2.4 3.5 Other hospital 0.2 0.3 0.1 Private clinic 26.0 26.7 25.5 Traditional herbal doctor 1.1 1.0 1.2 Other health center 3.9 3.7 4.1 Total 100.0 100.0 100.0 (Sample N) (3,760) (1,585) (2,175) Rural Areas Type of health service Total Male Female Village health center 3.6 3.6 3.6 Commune health center 33.5 31.1 35.5 Regional general clinics 3.5 3.6 3.4 District hospital 18.8 19.9 18.0 Provincial hospital 9.9 10.4 9.6 Center hospital 2.3 2.5 2.2 Other state-owned hospital 0.8 0.8 0.9 Private hospital 1.7 1.6 1.7 Other hospital 0.1 0.1 0.1 Private clinic 22.1 22.7 21.6 Traditional herbal doctor 1.6 1.4 1.7 Other health center 2.1 2.4 1.9 Total 100.0 100.0 100.0 (Sample N) (11,183) (4,954) (6,229) Note: Percentages are weighted; sample sizes are unweighted totals. * For the question on the type of health care services used during past 12 months, some respondents listed more than one. In this report, we focus on the distribution of only the first health care facility listed by respondents. Only 20 percent of respondents answered with more than one type of health care facilities. 79 Table 4-4. Reasons for Visiting Health Care Facilities in Past 12 months (%)* Urban Areas All Ages Age 0-19 Age 20-49 Age 50+ Reasons for visit Male Female Male Female Male Female Male Female Vaccination 1.8 2.7 5.0 6.3 0.2 2.6 0.0 0.2 Pregnancy & other GYN reason 0.0 5.6 0.0 0.3 0.0 13.4 --- --- Check-up & consulting 13.6 15.8 7.0 13.1 18.5 17.8 15.7 15.3 Treatment 84.6 75.9 88.0 80.4 81.4 66.3 84.3 84.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 (Sample N) (1,585) (2,175) (559) (550) (555) (919) (471) (706) Rural Areas All Ages Age 0-19 Age 20-49 Age 50+ Reasons for visit Male Female Male Female Male Female Male Female Vaccination 1.7 1.6 3.7 3.5 0.5 1.2 0.4 0.3 Pregnancy & other GYN reason 0.0 4.8 0.0 1.3 0.0 10.8 0.0 0.0 Check-up & consulting 11.7 13.4 10.1 10.8 11.1 14.3 14.9 14.7 Treatment 86.6 80.2 86.3 84.4 88.4 73.7 84.7 85.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 (Sample N) (4,954) (6,229) (1,923) (1,838) (1,792) (2,569) (1,239) (1,822) Note: Percentages are weighted; sample sizes are unweighted totals. * Information on reasons for visiting health care services was collected along with each type of health care facilities visited during the 12- month period. While some respondents listed more than one reason for visiting health care facilities, our analysis here focuses only on the first reason listed by respondents. About 20 percent of respondents listed more than one reason for visiting health care facilities. 80 Table 5-1. Prevalence of Disabilities (Age 5 and older only; percent people reporting some difficulty or more in each category of disabilities) MALE (N = 17,939) FEMALE (N = 18,764) Com- Com- Cog- Self- munica- Cog- Self- munica- Total* Vision Hearing nition Mobility care tion Total* Vision Hearing nition Mobility care tion Total 14.5 10.2 3.1 4.1 4.7 1.8 2.4 16.9 12.5 3.5 5.3 7.3 2.0 3.0 Area Urban 16.6 12.7 3.0 4.0 4.7 1.9 1.9 19.9 15.9 3.4 5.3 7.9 1.8 2.3 Rural 13.7 9.2 3.1 4.2 4.7 1.8 2.6 15.9 11.2 3.6 5.3 7.1 2.1 3.3 Age 5-9 5.5 0.7 0.5 1.8 0.9 3.2 2.2 5.3 0.8 0.2 1.7 0.7 2.9 1.6 10-14 3.8 2.0 0.4 1.0 0.7 0.8 1.1 3.4 2.2 0.6 0.6 0.4 0.2 0.5 15-19 3.5 1.6 0.7 1.2 0.7 0.7 1.2 5.4 3.3 0.4 1.1 0.9 0.6 1.1 20-29 5.0 1.9 0.8 1.6 1.1 0.6 1.1 4.6 2.0 0.6 1.4 0.9 0.5 1.2 30-39 6.4 2.3 0.8 2.3 1.8 0.7 1.8 3.7 1.4 0.6 1.4 1.1 0.5 0.8 40-49 16.6 12.0 1.2 3.0 3.6 1.2 1.7 15.6 11.0 1.4 3.6 3.9 0.5 1.4 50-59 30.7 25.1 4.7 5.1 6.7 2.2 2.0 35.0 28.5 4.1 6.8 11.7 1.9 1.8 60+ 62.8 51.2 21.4 23.0 30.0 8.4 11.1 65.5 52.8 20.7 26.6 40.2 10.4 16.0 Region Red River Delta 14.3 9.5 3.7 3.9 5.1 2.1 2.4 16.7 12.2 3.7 4.9 6.0 2.4 3.1 North East 15.1 10.3 2.8 4.4 4.7 1.3 2.2 17.8 12.8 4.0 5.8 7.4 2.0 3.1 North West 10.6 7.1 2.3 2.3 3.1 1.0 2.1 11.4 7.5 2.4 3.2 3.6 1.0 1.6 North Central Coast 13.1 8.6 2.8 4.1 3.9 2.0 3.1 14.8 10.2 3.7 4.9 6.8 2.6 4.0 South Central Coast 13.0 9.0 2.5 2.4 4.1 2.0 1.8 16.6 11.8 3.4 4.6 6.8 2.3 2.6 Central Highlands 13.3 9.3 2.9 4.5 4.8 1.8 2.6 15.2 10.8 3.3 5.8 6.5 1.9 3.4 South East 17.3 12.6 2.9 5.9 5.5 2.4 3.2 19.2 14.7 3.6 7.0 8.7 1.7 3.1 Mekong River Delta 14.6 11.1 3.2 3.7 4.5 1.3 1.7 17.7 14.0 3.1 4.9 6.9 1.7 2.5 (continued) 81 Table 5-1. Prevalence of Disabilities (continued) MALE (N = 17,939) FEMALE (N = 18,764) Com- Com- Cog- Self- munica Cog- Self- munica Total* Vision Hearing nition Mobility care -tion Total* Vision Hearing nition Mobility care -tion Ethnicity Kinh/Chinese 14.9 10.6 3.2 4.2 4.7 1.9 2.4 17.5 13.1 3.5 5.4 7.6 2.1 2.9 Ethnic minorities 11.8 7.4 2.5 3.5 4.3 1.5 2.4 13.4 8.6 3.5 5.0 5.8 1.8 3.4 Tay/Thai/Muong/Nung 12.4 8.1 2.6 3.3 4.3 1.3 1.9 13.5 9.0 3.8 4.9 6.6 1.6 3.3 N. Mountain ethnic 9.9 4.7 1.8 3.6 2.8 1.3 2.4 11.4 5.1 2.8 3.6 3.8 2.3 3.9 Central ethnic 12.4 7.5 2.3 4.1 5.9 2.3 3.8 13.7 9.1 3.4 6.5 5.4 2.0 3.9 Khmer/Cham 10.5 8.0 2.9 3.0 3.5 1.4 1.8 15.4 11.3 3.5 4.7 6.0 2.0 2.4 Income quintile 1 Lowest 14.0 8.2 3.6 5.1 5.3 2.2 3.7 15.4 9.8 4.0 5.9 6.9 2.4 4.0 2 13.9 9.4 3.3 4.3 4.5 1.9 2.7 16.7 12.1 4.3 5.9 7.7 2.3 3.8 3 13.3 8.9 3.3 4.1 4.7 1.8 2.4 16.3 12.2 3.3 5.1 7.6 2.1 2.9 4 14.9 11.4 2.4 3.4 4.2 1.4 1.5 17.9 13.6 3.3 5.2 7.2 1.7 2.4 5 Highest 16.2 12.7 2.8 3.7 4.7 1.9 1.8 18.4 14.7 2.6 4.5 7.3 1.6 2.0 Schooling completed (Age 20-49 only) No schooling 24.6 5.9 5.3 15.1 6.5 4.2 14.0 20.5 4.3 3.2 13.6 5.9 4.8 11.5 Less than primary 12.1 7.0 1.8 3.4 2.2 1.4 2.5 11.2 7.0 1.1 3.4 2.6 0.3 1.4 Primary 9.2 5.1 0.9 2.5 2.2 0.6 1.1 7.2 4.5 0.8 1.6 1.7 0.3 0.6 Lower second 8.2 4.7 0.6 1.7 2.3 0.6 0.9 7.2 4.5 0.9 1.5 2.4 0.4 0.6 Upper second 6.4 4.6 0.6 0.8 1.4 0.5 0.7 5.3 4.2 0.2 0.5 1.1 0.0 0.1 Junior co/university 10.6 9.0 0.0 0.0 1.3 0.8 0.0 7.4 7.2 0.0 0.0 0.2 0.0 0.0 Note: Percentages are weighted; sample sizes are unweighted totals. * Total indicates the overall rate of disability and refers to the percentage of people reporting some difficulty or more in any of the six categories of disability. For each type of disability, three levels of response categories were given in the questionnaire to assess functional limitations: some difficulty, a lot of difficulty, and impossible. (For a detailed list of questions, see Appendix A-2.) 82 Table 5-2. Types of Disability by Age Group (Percent reporting disability in each category) Male Female Total Age 5-19 Vision 1.5 2.3 1.9 Hearing 0.6 0.4 0.5 Mobility 0.7 0.7 0.7 Self-care 1.3 0.9 1.2 Cognition 1.3 1.1 1.2 Communication 1.4 1.0 1.2 Total* 4.1 4.6 4.3 (Sample N) (6,261) (5,926) (12,187) Age 20-49 Vision 5.3 4.9 5.1 Hearing 0.9 0.8 0.9 Mobility 2.1 2.0 2.1 Self-care 0.8 0.5 0.7 Cognition 2.3 2.2 2.2 Communication 1.5 1.1 1.3 Total* 9.2 8.1 8.6 (Sample N) (8,389) (8,694) (17,083) Age 50 and Older Vision 37.8 41.7 40.0 Hearing 12.9 13.1 13.0 Mobility 18.1 27.1 23.1 Self-care 5.2 6.6 5.9 Cognition 13.8 17.5 15.9 Communication 6.4 9.5 8.1 Total* 46.3 51.6 49.3 (Sample N) (3,287) (4,144) (7,431) Note: Percentages are weighted; sample sizes are unweighted totals. * Total indicates the percentage of people reporting disability in any of the six categories; therefore, the total percentage does not match with the sum of percentages listed for each category. 83 Table 5-3. Rates of Disability for Different Levels of Severity (Age 5 and Older Only) (Percent reporting disability for each level of severity)* Male Female All Households (Sample N) (N = 17,937) (N = 18,764) (N = 9,189) % HH with anyone with severe Moderate Severe Total Moderate Severe Total disability Total 11.3 3.2 14.5 12.9 4.0 16.9 12.0 Area Urban 13.4 3.3 16.6 16.3 3.5 19.8 11.1 Rural 10.5 3.2 13.7 11.7 4.2 15.9 12.9 Age 5-9 3.9 1.6 5.5 3.9 1.4 5.3 10-14 2.8 1.0 3.8 2.8 0.6 3.4 15-19 2.4 1.2 3.6 3.9 1.5 5.4 20-29 3.5 1.5 5.0 3.1 1.5 4.6 30-39 4.5 2.0 6.4 2.6 1.1 3.7 40-49 14.0 2.6 16.6 14.1 1.4 15.5 50-59 26.1 4.5 30.7 30.8 4.3 35.1 60+ 47.6 15.2 62.8 44.5 21.0 65.5 Region Red River Delta 11.0 3.4 14.3 12.6 4.2 16.7 12.1 North East 12.9 2.3 15.1 14.2 3.6 17.8 10.4 North West 8.7 2.0 10.6 9.2 2.2 11.4 8.4 North Central Coast 9.8 3.3 13.1 10.4 4.3 14.8 12.9 South Central Coast 9.9 3.1 13.0 12.1 4.5 16.6 12.8 Central Highlands 10.3 3.0 13.3 11.4 3.8 15.2 13.3 South East 13.0 4.2 17.3 15.3 3.9 19.2 13.9 Mekong River Delta 11.7 3.0 14.6 13.7 4.0 17.7 12.7 Ethnicity Kinh/Chinese 11.6 3.3 14.9 13.4 4.1 17.5 12.5 Ethnic minorities 9.4 2.4 11.8 10.0 3.4 13.4 11.5 Tay/Thai/Muong/Nung 10.2 2.3 12.4 10.1 3.3 13.5 11.2 N. Mountain ethnic 8.2 1.7 10.0 9.7 1.7 11.4 7.8 Central ethnic 9.4 2.9 12.4 8.7 5.0 13.7 14.5 Khmer/Cham 7.5 3.0 10.5 12.3 3.1 15.4 13.2 Income quintile 1 Lowest 10.0 4.0 14.0 11.0 4.4 15.4 15.1 2 10.6 3.3 13.9 11.8 4.9 16.7 14.9 3 10.3 3.1 13.3 12.3 4.0 16.3 12.3 4 12.2 2.7 14.9 14.2 3.7 17.9 11.1 5 Highest 13.3 2.8 16.2 15.4 3.1 18.4 9.5 Schooling completed (Age 20-49) No schooling 9.4 15.2 24.6 9.7 10.9 20.5 Less than primary 8.8 3.3 12.2 9.8 1.4 11.2 Primary 7.7 1.4 9.2 6.5 0.7 7.2 Lower secondary 6.9 1.3 8.2 6.3 0.9 7.2 Upper secondary 5.3 1.1 6.4 4.9 0.4 5.3 Junior co/university 9.8 0.8 10.6 6.8 0.6 7.4 Note: Percentages are weighted; sample sizes are unweighted totals. *People with moderate disability include those who reported "some difficulty" for any of six categories of disabilities; People with severe disability include those who reported "a lot of difficulty" or "impossible" for any of six categories of disabilities. 84 Table 5-4. Causes of Disability by Gender and Age Group (%) Age 5-19 Age 20-49 Age 50+ Male Female Total Male Female Total Male Female Total Birth defects 25.9 19.9 22.9 10.8 11.7 11.2 1.6 1.5 1.5 War/agent orange 4.2 4.1 4.2 5.6 2.7 4.3 7.6 1.9 4.2 Accidents 2.9 1.7 2.3 12.5 3.4 8.1 2.3 1.3 1.7 Illness 27.3 29.7 28.5 42.7 51.6 46.9 16.3 18.8 17.8 Age (Aging) 17.8 19.4 18.6 8.9 10.9 9.8 69.8 74.7 72.7 Other 21.8 25.3 23.6 19.5 19.8 19.6 2.5 1.8 2.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 (Sample N) (256) (277) (533) (755) (709) (1,464) (1,544) (2,127) (3,671) Note: Percentages are weighted; sample sizes are unweighted totals. Table 5-5. Disability, School Attendance, and Work Participation Male (N) Female (N) % Children attending school (Age 6-17) Own disability No disability 87.0 (4,844) 89.2 (4,641) Moderate disability 84.4 (139) 92.4 (167) Severe disability 43.5 (49) 41.1 (42) Disability in family None 87.4 (3,310) 89.8 (3,190) Moderate disability 87.3 (1,182) 90.0 (1,157) Severe disability 79.5 (540) 80.7 (503) % Adults employed in past 12 Months (Age 25-54) Own disability No disability 97.8 (6,540) 93.7 (7,093) Moderate disability 94.1 (794) 88.7 (835) Severe disability 52.7 (166) 42.0 (127) Disability in family None 98.0 (4,679) 93.9 (4,943) Moderate disability 96.1 (1,984) 92.0 (2,187) Severe disability 87.9 (837) 85.7 (925) Note: Percentages are weighted; sample sizes are unweighted totals for each group. 85 Appendix A-1. Classification of Ethnic Minority Groups Ethnic minorities are divided into four groups in this study in part based on the guidelines used in the 1999 Population Census and in part based on previous research showing varying socioeconomic characteristics and outcomes among different groups of ethnic minorities. The guidelines in the 1999 Census classify ethnic groups according to the region where the group predominates: the Northern mountain, Central, and Southern ethnic groups. This classification was used in the analysis of the 2001-2002 Vietnam National Health Survey. Research examining differences among ethnic minorities shows that among the Northern mountain ethnic groups, the Tay, Thai, Muong, and Nung differ considerably from the other Northern ethnic groups in their geography and associated economic activities, with subsequent differences in living standards (Baulch et al. 2008b). We therefore separate these four ethnic groups from the Northern mountain ethnic groups. Similarly, the experiences of the Khmer and Cham are regarded divergent from the other Southern ethnic groups such as Xtieng, Cho Ro; hence, the former two groups are categorized into one separate group. We include the two latter Southern ethnic groups (the sample size of which is quite small) into the Central ethnic group, as done by Baulch et al.(2008b). In the current report, the 54 ethnic minority groups listed in the 2006 VHLSS are classified into the four broad ethnic groups as follows, along with the Kinh and Chinese. The Chinese are combined with the Kinh as usually done in most studies and statistical classifications. % distribution in Main Groups Ethnic Groups the 2006 VHLSS Kinh and Chinese Kinh, Chinese 86.5% Tay, Thai, Muong, and Tay, Thai, Muong, and Nung 7.0% Nung Northern mountain H'mong, Dao, Ngai, San Chay, San Diu, Giay, Kho 2.3% ethnic groups Mu, Khang, Xinh Mun, Ha Nhi, Lao, La Ha, La Chi, Phu La, La Hu, Lu, Lo Lo, Mang, Pa Then, Co Lao, Cong, Bo Y, Si La, Pu Peo Central ethnic groups Hre, Raglai, Bru-Van Kieu, Tho, Co Tu, Co, Ta Oi, 2.8% Chut, O Du, Jarai, E-De, Ba Na, Xo Dang, Co Ho, Mnong, Gie Trieng, Ma, Chu Ru, Brau, Ro Mam, Xtieng, Cho Ro Khmer and Cham Khmer, Cham 1.5% 86 Appendix A-2. The 2006 VHLSS Questions on Disability Used for Analysis In this report, the following questions from the 2006 VHLSS are used to measure disability in six categories: vision, hearing, mobility, cognition, self care, and communication. The same set of questions was asked to all respondents aged 5 and older. Specific questions and response categories are as follows. Response categories: Not difficult; A little difficult; Very difficult; or Impossible. Vision: Does [Name] have visual difficulty even when wearing glasses? Hearing: Does [Name] have any aural difficulty even when wearing a hearing aid? Mobility: Does [Name] have difficulty going on foot or walking upstairs/downstairs? Cognition: Does [Name] have any difficulty remembering or concentrating? Self care: Does [Name] have difficulty taking care of himself/herself, e.g., bathing, or getting dressed? Communication: Due to his/her physical or emotional conditions, does [Name] have difficulty communicating, e.g., understanding others and making himself/herself understood? When a respondent indicated "a little difficult," "very difficulty," or "impossible," the person is coded as having a disability in each of the above six categories. When responses are "very difficult" or "impossible," the person is also coded as having a severe disability. For the overall measure of disability, we code the person having a disability if he/she has a disability in any of the six categories. The person is also coded as having a severe disability if he/she has a severe disability in any of the six categories. The VHLSS questionnaire includes other sub-questions for each of the six categories following the main questions listed above, in order to measure more specific aspects of disability in each category. For instance, in the case of mobility, following the general question listed above, three detailed questions are asked to respondents: Does [Name] have difficulty doing the following by himself/herself without using tools to help? a. Walking about 400 meters long? b. Walking up to 10 steps without stopping? c. Bowing and bending his/her body, or standing on his/her knees? More detailed questions regarding disability in each category are not analyzed in this study.