Overage pupils in primary and secondary education

Pupils can be overage for their grade for two reasons: late entry and repetition. Take for example a country where children are expected to enter primary school at 6 years of age. If a child enters grade 1 at age 7, he or she is one year overage for the grade. A child who enters grade 1 at age 8 and repeats the grade will be three years overage for the grade; two of the three years are due to late entry and the third year is due to repetition.

Children who are many years overage are less likely to complete their education. If they stay in school, they graduate later than pupils who entered school at the official starting age. These overage graduates enter the labor market late and often with lower educational attainment. As a consequence, they are likely to have lower cumulative earnings over their lifetime than persons who graduated and entered the labor market at a younger age and with higher educational attainment. For the country as a whole this in turn means reduced national income and slower economic growth.

Overage school attendance is common in sub-Saharan Africa but also occurs in other regions. The figure below shows data from 36 nationally representative household surveys that were conducted between 2004 and 2009. 34 of these surveys were Demographic and Health Surveys (DHS) and the remaining two surveys, those for Bangladesh and Kyrgyzstan, were Multiple Indicator Cluster Surveys (MICS). For each country, the graph shows the share of children in primary and secondary education who are at least one or two years overage for their grade. The entrance ages and durations of primary and secondary education used in this study are those specified by the International Standard Classification of Education (ISCED).

Percentage of children in primary and secondary education who are at least 1 or 2 years overage for their grade

Source: Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS), 2004-2009.

In the sample of 36 countries, the share of children who are at least one year overage for their grade ranges from 5 percent in Armenia to 95 percent in Haiti. Other countries where at least three out of four pupils in primary or secondary education are overage include Liberia (93%), Uganda (86%), Rwanda (83%), Cambodia (78%), Mozambique (76%), and Ethiopia (75%). In addition to Armenia, the percentage of pupils who are at least one year overage is below 10 percent in Moldova and Egypt (8%).

The share of children in primary and secondary education who are at least two years overage for their grade ranges from 1 percent in Armenia to 85 percent in Haiti. In addition to Haiti, at least half of all pupils are two or more years overage in Liberia (84%), Uganda (67%), Rwanda (65%), Ethiopia (59%), Cambodia (55%), Malawi (51%), and Madagascar (50%). On average, the share of children who are at least two years overage is 19 percent less than the share of children who are at least one year overage.

However, there are exceptions. In Albania and the Ukraine, 43 and 26 percent respectively of all children in primary and secondary education are at least one year overage. By contrast, only 5 and 2 percent respectively are at least two years overage. This means that in these two countries, a relatively large number of children enter school one year late or repeat one grade, but hardly any children enter school two years late or repeat more than one grade. Late entry and repetition are therefore less likely to have negative consequences on lifetime earnings and national income in Albania and the Ukraine than in other countries.

Related articles

• Official school ages: primary, secondary, and compulsory education
• Primary school entrance age and duration
• Repetition rates in primary and secondary school
• Survival rate to the last grade of primary school
• Transition from primary to secondary education
• Regional disparities in school life expectancy
• Children of primary and secondary school age out of school
• Children out of school: Global trend 1999-2008
• Household wealth and years of education
• National wealth and years of education
• National wealth and school enrollment
• Poverty and educational attainment in the United States
• Poverty and educational attainment in the United States, part 2
• Review of the International Standard Classification of Education (ISCED)

External links

• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)
• International Standard Classification of Education (ISCED)

Friedrich Huebler, 30 June 2011, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2011/06/age.html

EFA Global Monitoring Report 2010

The Education for All Global Monitoring Report 2010 was released on 19 January 2010. The Global Monitoring Report is written annually by an independent team and published by UNESCO.

The title of this year's report is Reaching the marginalized. UNESCO estimates that 72 million children of primary school age were out of school in 2007. The report examines who these children are and why they are excluded from education. The report further argues that there is a persistent financing gap that prevents countries from reaching the goal of education for all and that, based on current trends, 56 million children of primary school age will still be out of school in 2015.

The report introduces a new database on Deprivation and Marginalization in Education that was developed by the EFA Global Monitoring Report team and the Department of Economics at the University of Göttingen. The DME database introduces a measure of "education poverty", defined as the share of the population aged 17 to 22 years with less than 4 years or less than 2 years in school. Data are presented as global snapshots and in individual country profiles. All statistics were calculated with data from Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS).

Excerpt from Nigeria country overview in DME database

Source: Deprivation and Marginalization in Education database, country overviews.

Reference

• UNESCO. 2010. EFA Global Monitoring Report 2010: Reaching the marginalized. Paris: UNESCO. (Download in PDF format, 12 MB)

External links

• EFA Global Monitoring Report 2010
• Deprivation and Marginalization in Education database
• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)

Related articles

• EFA Global Monitoring Report 2011
• EFA Global Monitoring Report 2009
• EFA Global Monitoring Report 2008
• Global Education Digest 2009
• Release of 2008 education data by UIS

Friedrich Huebler, 31 January 2010 (edited 7 March 2011), Creative Commons License
Permanent URL: http://huebler.blogspot.com/2010/01/gmr.html

MICS Compiler by UNICEF

MICS Compiler, a new website by UNICEF, provides easy access to data from Multiple Indicator Cluster Surveys (MICS), nationally representative household surveys that are carried out with support from UNICEF. The site is similar to STATcompiler, which offers data from Demographic and Health Surveys (DHS).

MICS Compiler was launched with data from 26 surveys conducted in Africa, Asia, Eastern Europe, and Latin America and the Caribbean between 2005 and 2007. Estimates are available for 39 indicators in ten areas.

1. Survey information
2. Child mortality
3. Nutrition
4. Child health
5. Environment
6. Reproductive health
7. Child development
8. Education
9. Child protection
10. HIV/AIDS, sexual behavior, and orphaned and vulnerable children

Access to the data requires two steps. In the first step, users of MICS Compiler must select one or more surveys. In the second step, the indicators are selected. The results are presented in tables or graphs. As an example, the screenshot below shows a graph with the female youth literacy rate in 21 countries.

MICS Compiler by UNICEF: Female youth literacy rate in 21 countries, 2005-2006


At present, the female youth literacy rate is the only indicator listed in the area of education but the MICS for All blog has announced plans to expand MICS Compiler with data for more indicators and more surveys. There are also plans for adding a mapping function, similar to the DHS STATmapper.

Related articles

• UNICEF releases new MICS survey data
• Multiple Indicator Cluster Survey blog
  
• Education data from household surveys

External links

• MICS Compiler
• Multiple Indicator Cluster Surveys (MICS)
• MICS for All blog
• UNICEF
• STATcompiler
• STATmapper
• Demographic and Health Surveys (DHS)

Friedrich Huebler, 30 December 2009, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2009/12/mics.html

Disparities in secondary school attendance by ethnicity, language or religion

Members of ethnic, linguistic or religious minorities face barriers to access to education in many countries. In an article on primary school attendance by ethnicity, language or religion the presence of such disparities was demonstrated with data from Multiple Indicator Cluster Surveys. The MICS are nationally representative household surveys supported by UNICEF that collect data on school attendance and other household member characteristics. In the most recent round of MICS surveys, carried out in 2005 and 2006, 17 countries collected data on school attendance by ethnicity, language or religion: Albania, Belize, Gambia, Georgia, Guinea-Bissau, Guyana, Kazakhstan, Kyrgyzstan, Lao PDR, Macedonia, Montenegro, Serbia, Sierra Leone, Thailand, Togo, Uzbekistan, and Viet Nam.

The school attendance data from the MICS surveys can be used to generate an education parity index that measures relative disparity across different groups of disaggregation, as described in the article on primary school attendance. To calculate the index, the attendance rate of the group with the lowest value is divided by the attendance rate of the group with the highest value. The result is a value between 0 and 1, where 1 means that children from different ethnic, linguistic or religious groups have the same secondary school attendance rate. Values closer to 0 indicate increasing disparity.

As an example, Thailand collected data on school attendance that can be linked to the mother tongue of the household head. The secondary school net attendance rates (NAR) for two groups of children identified in the 2005-06 MICS data are shown in Table 1.

Table 1: Secondary school attendance in Thailand

Mother tongue of household head	Secondary NAR (%)
Thai	81.2
Other language	65.8
Total	79.8

Data source: MICS 2005-06.

Among children from households whose head speaks Thai, the secondary NAR is 81.2 percent. Among children from households headed by someone with a different mother tongue, the secondary NAR is 65.8 percent. The secondary school parity index for Thailand is then calculated as follows.

Secondary school parity index	= Lowest secondary NAR / Highest secondary NAR
	= Secondary NAR of speakers of another language /    Secondary NAR of speakers of Thai
	= 65.8 / 81.2
	= 0.81

The parity index is a relative, not an absolute measure of disparity. The value 0.81 means that the secondary NAR of speakers of another language is, relatively speaking, 19 percent below the secondary NAR of Thai speakers. The absolute gap between children from the two groups is 15.4 percent, the difference between 81.2 and 65.8.

The secondary school parity index for all 17 countries with data is shown in Figure 1. The index ranges from a high of 0.98 in Viet Nam to a low of 0.17 in Serbia. The low value for Serbia is explained by extremely low secondary school attendance among the Roma ethnic group. The secondary school NAR for Roma children is 14.8 percent, compared to 85.9 percent for Serbians and 88.6 percent for children from other ethnic groups. In addition to Serbia, six other countries have index values at or below 0.5: Lao PDR, Macedonia, Guinea-Bissau, Togo, Belize, and Montenegro. In these countries, children from the most advantaged ethnic, linguistic or religious group have secondary school net attendance rates that are at least twice as high as the attendance rates of children from the most disadvantaged group. In Viet Nam, Kazakhstan, Albania, and Uzbekistan, on the other hand, disparities in access to secondary education are relatively small.

Figure 1: Secondary school parity index: School attendance by ethnicity, language or religion

Data source: MICS 2005-2006.

The attendance rates used to calculate the secondary school parity index are summarized in Table 2. The table also shows whether the national agencies that implemented a survey chose ethnicity, language or religion to identify minorities. A comparison with data on primary school attendance makes clear that disparities at the secondary level of education are much larger than disparities at the primary level, where the parity index for the same group of countries has a range from 0.59 to 0.99.

Table 2: Disparities in secondary school attendance by ethnicity, language or religion

Country	Year	Characteristic	Primary NAR (%)		Parity index
			Min.	Max.
Albania	2005	Religion	77.1	83.7	0.92
Belize	2006	Language	36.9	79.2	0.47
Gambia	2006	Ethnicity	27.5	48.2	0.57
Georgia	2005	Ethnicity	69.0	90.6	0.76
Guinea-Bissau	2006	Language	4.3	13.8	0.31
Guyana	2006	Ethnicity	56.0	81.1	0.69
Kazakhstan	2006	Language	90.8	96.0	0.95
Kyrgyzstan	2006	Language	79.3	92.4	0.86
Lao PDR	2006	Language	10.0	45.6	0.22
Macedonia	2005	Ethnicity	17.4	73.7	0.24
Montenegro	2005	Ethnicity	46.5	92.9	0.50
Serbia	2005	Ethnicity	14.8	88.6	0.17
Sierra Leone	2005	Religion	17.8	24.4	0.73
Thailand	2005-06	Language	65.8	81.2	0.81
Togo	2006	Ethnicity	22.9	53.1	0.43
Uzbekistan	2006	Language	87.1	95.4	0.91
Viet Nam	2006	Ethnicity	93.8	95.7	0.98

Data source: MICS 2005-2006.

Data source

• Multiple Indicator Cluster Surveys (MICS), 2005-2006.

Related articles

• Disparities in primary school attendance by ethnicity, language or religion
• Caste, ethnicity, and school attendance in Nepal
• Education disparity in South Asia
• Education disparity trends in South Asia

External links

• Multiple Indicator Cluster Surveys (MICS)
• UNICEF

Friedrich Huebler, 15 March 2009, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2009/03/elr2.html

Disparities in primary school attendance by ethnicity, language or religion

In many parts of the world, members of ethnic, linguistic or religious minorities face barriers to access to education. One example is Nepal, where caste and ethnicity are closely linked to primary and secondary school attendance rates. Because of the importance of this issue, "Minorities and the right to education" was the thematic focus of the first United Nations Forum on Minority Issues, which took place in Geneva on 15 and 16 December 2008.

The presence of disparities in national education systems can be demonstrated with data from Multiple Indicator Cluster Surveys (MICS), nationally representative household surveys that are carried out with the support of UNICEF. The MICS data collection process is explained in the Multiple Indicator Cluster Survey Manual 2005 (UNICEF 2006). MICS surveys conducted in 2005 and 2006 collected data on school attendance by ethnicity, language or religion in the following countries: Albania, Belize, Gambia, Georgia, Guinea-Bissau, Guyana, Kazakhstan, Kyrgyzstan, Lao PDR, Macedonia, Montenegro, Serbia, Sierra Leone, Thailand, Togo, Uzbekistan, and Viet Nam.

Minority Rights Group International (MRG) defines minorities as "non-dominant ethnic, religious and linguistic communities, who may not necessarily be numerical minorities. ... [These groups] may lack access to political power, face discrimination and human rights abuses, and have 'development' policies imposed upon them" (MRG 2009). The MICS data alone are not sufficient to identify groups that can be considered minorities as defined by MRG because the size of particular groups in relation to the entire population of a country does not indicate whether these groups are discriminated in any way. This article therefore examines differences in school attendance between all ethnic, linguistic or religious groups for which data are available. Disparities between these groups can provide insights into whether any part of a country's population faces discrimination or is otherwise disadvantaged.

With the school attendance data from the MICS surveys it is possible to generate an education parity index that measures relative disparity across different groups of disaggregation, following the methodology developed by Huebler (2008) for data on school attendance by sex, area of residence, and household wealth. The education parity index has a range of 0 to 1, where 1 indicates parity between all groups of disaggregation. This methodology can also be applied to primary school attendance rates by ethnicity, language or religion. To calculate the index, the attendance rate of the group with the lowest value is divided by the attendance rate of the group with the highest value, yielding a value between 0 and 1. The value 1 means that children from different ethnic, linguistic or religious groups have the same primary school attendance rates. Smaller values indicate increasing disparity.

The calculation of the parity index can be illustrated with data from Macedonia. A MICS survey conducted in 2005 collected data on school attendance by ethnic group of the household head. Four ethnic groups are identified in the data and their respective primary school net attendance rates (NAR) are shown in Table 1.

Table 1: Primary school attendance in Macedonia

Ethnic group of household head	Primary NAR (%)
Albanian	97.8
Macedonian	97.5
Roma	61.1
Other ethnic group	81.9
Total	94.9

Data source: MICS 2005.

Albanians in Macedonia have the highest primary NAR, 97.8 percent. By contrast, Roma have the lowest NAR, 61.1 percent. In other words, only 6 of 10 Roma children of primary school age are attending primary school. With these values, the primary school parity index for Macedonia can be calculated as follows:

Primary school parity index	= Lowest primary NAR / Highest primary NAR
	= Primary NAR of Roma / Primary NAR of Albanians
	= 61.1 / 97.8
	= 0.62

The value 0.62 means that the attendance rate of the most disadvantaged group, Roma, is 62 percent of the attendance rate of the least disadvantaged group, Albanians. In other words, the primary NAR of Roma is 38 percent below the primary NAR of ethnic Albanians. 38 percent is not the absolute but the relative difference in school attendance because the education parity index is a relative measure of disparity.

Applying the same formula to primary NAR values from other MICS surveys yields the values in Figure 1, which shows the parity index for primary school attendance by ethnicity, language or religion. In the 17 countries with data, the parity index ranges from a high of 0.99 in Guyana to a low of 0.59 in the Lao People's Democratic Republic. In Laos, speakers of the Lao language are significantly more likely to attend primary school than speakers of other languages, whose primary school NAR is 41 percent below the NAR of Lao speakers. Similar disparities exist in Togo, where members of the Para-Gourma ethnic group have a much lower primary school attendance rate than members of the Akposso-Akébou group, and in Macedonia.

Uzbekistan and Viet Nam are characterized by the near absence of disparities in primary school attendance between different ethnic, linguistic or religious groups, similar to Guyana. In these countries, the primary NAR of the group with the lowest attendance rate is only 1 or 2 percent below the primary NAR of the group with the highest attendance rate.

Figure 1: Primary school parity index: School attendance by ethnicity, language or religion

Data source: MICS 2005-2006.

The primary school net attendance rates used to calculate the parity index are listed in Table 2. The table also shows whether ethnicity, language or religion were chosen to identify minorities in a country. This choice was made by the national agencies that implemented the survey. Eight countries selected ethnicity, seven countries selected language, and two countries selected religion as the characteristic that best captures minority status.

Table 2: Disparities in primary school attendance by ethnicity, language or religion

Country	Year	Characteristic	Primary NAR (%)		Parity index
			Min.	Max.
Albania	2005	Religion	91.3	94.9	0.96
Belize	2006	Language	86.6	100	0.87
Gambia	2006	Ethnicity	53.2	72.9	0.73
Georgia	2005	Ethnicity	86.9	97.5	0.89
Guinea-Bissau	2006	Language	44.9	64.7	0.69
Guyana	2006	Ethnicity	95.7	96.8	0.99
Kazakhstan	2006	Language	95.4	98.9	0.96
Kyrgyzstan	2006	Language	86.7	95.4	0.91
Lao PDR	2006	Language	52.4	88.7	0.59
Macedonia	2005	Ethnicity	61.1	97.8	0.62
Montenegro	2005	Ethnicity	69.4	100	0.69
Serbia	2005	Ethnicity	77.9	100	0.78
Sierra Leone	2005	Religion	68.3	72.3	0.94
Thailand	2005-06	Language	94.8	98.2	0.97
Togo	2006	Ethnicity	55.2	91.1	0.61
Uzbekistan	2006	Language	94.9	96.8	0.98
Viet Nam	2006	Ethnicity	93.8	95.7	0.98

Data source: MICS 2005-2006.

References

• Huebler, Friedrich. 2008. Beyond gender: Measuring disparity in South Asia using an education parity index. Kathmandu: UNICEF.
• Minority Rights Group International (MRG). 2009. Who are minorities?
• United Nations Children's Fund (UNICEF). 2006. Multiple Indicator Cluster Survey manual 2005: Monitoring the situation of women and children. New York: UNICEF.
  

Data source

• Multiple Indicator Cluster Surveys (MICS), 2005-2006.

Related articles

• Disparities in secondary school attendance by ethnicity, language or religion
• Caste, ethnicity, and school attendance in Nepal
• Education disparity in South Asia
• Education disparity trends in South Asia

External links

• United Nations Forum on Minority Issues
• Minority Rights Group International (MRG)
• Multiple Indicator Cluster Surveys (MICS)
• UNICEF

Friedrich Huebler, 1 March 2009 (edited 15 March 2009), Creative Commons License
Permanent URL: http://huebler.blogspot.com/2009/03/elr.html

Multiple Indicator Cluster Survey blog

The Multiple Indicator Cluster Surveys (MICS) are household surveys carried out in developing countries with the support of UNICEF to collect data on the situation of children and women. The most recent round of MICS surveys was conducted between 2005 and 2007 in more than 40 countries. MICS data and documentation are available at the Childinfo website of UNICEF.

MICS surveys are among the sources of data analyzed on this site. MICS data were used in articles on trends in primary education in Nepal, children out of school in India, child labor and school attendance in Bolivia, education disparity trends in South Asia, global data on child labor and school attendance, household wealth and years of education, the link between years of schooling and literacy, and other studies.

UNICEF staff members working on the MICS have launched a new blog at globalmics.blogspot.com. The goal of the blog is "to facilitate information sharing between different organizations and individuals involved with MICS implementation around the world" and "to play the role of an unofficial, informal forum to share information on MICS activities." Articles posted since the launch have treated a variety of topics, among them acronyms and abbreviations related to MICS, members of the global MICS team, and the evaluation of the latest round of MICS.

External links

• MICS For All: A Blog Site for Multiple Indicator Cluster Surveys
• Childinfo: Monitoring the Situation of Children and Women (UNICEF)

Related articles

• Nepal: trends in primary education, 1980-2004
• India has 21 million children out of school
• Child labor and school attendance in Bolivia
• Education disparity trends in South Asia
• Child labor and school attendance
• Household wealth and years of education
• Years of schooling and literacy: Can everyone with primary education read and write?

Friedrich Huebler, 23 November 2008, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/11/mics.html

Education disparity trends in South Asia

An article on education disparity in South Asia described a newly developed Education Parity Index (EPI). This index combines data on primary school attendance, secondary school attendance and the survival rate to the last grade of primary school, disaggregated by gender, area of residence and household wealth. The value of the EPI has a theoretical range of 0 to 1, where 1 indicates absolute parity.

Through a combination of survey data from several years it is possible to analyze trends in disparity as measured by the EPI. For the trend analysis, data from the following South Asian household surveys - mainly Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) - were available.

• Afghanistan: 2003 MICS
• Bangladesh: 1999-2000 DHS, 2004 DHS, 2006 DHS
• India: 1998-99 DHS, 2000 MICS, 2005-06 DHS
• Nepal: 1996 DHS, 2000 MICS, 2001 DHS, 2006 DHS
• Pakistan: 2000-01 survey, 2006-07 DHS

The graph below plots the EPI values calculated from each survey. Due to a lack of data, no trends can be shown for Afghanistan.

Education disparity trends in South Asia, 1996-2007

Data source: Demographic and Health Surveys (DHS), Multiple Indicator Cluster Surveys (MICS), 1996-2007.

In Bangladesh, India and Nepal, the EPI has increased from the earliest to the latest year with data, indicating a decrease in disparity over the period of observation. In Bangladesh, the EPI grew from 0.79 in 2000 to 0.84 in 2006. In India, the EPI was at 0.77 in 1999 and 0.82 in 2006. In Nepal, the EPI shows the biggest increase, from 0.67 in 1996 to 0.83 in 2006, interrupted by a decrease from 2000 to 2001. Compared to the other countries, Nepal has thus made the most progress toward parity in the education system.

For Pakistan, the EPI has decreased from 2000 to 2007, indicating an increase in disparity. However, an inspection of the underlying data reveals that the earlier survey did not provide data on household wealth. Disparities related to wealth are usually greater than disparities related to gender or area of residence. If data on wealth had been available, the EPI for 2000 would most likely have been lower. The data from the 2006-07 DHS confirm this assumption. Children from the poorest quintile have much lower attendance and survival rates than children from the richest quintile, and the disparity between these two groups of children is much greater than the disparity between boys and girls and between children from urban and rural households. For example, the primary school net attendance rate (NAR) in Pakistan is 46 percent among children from the poorest household quintile but twice as high, 93 percent, among children from the richest quintile. In comparison, the primary NAR is 76 percent for boys, 67 percent for girls, 82 percent for urban children, and 67 percent for rural children according to the 2006-07 DHS.

The data gaps in the graph bring to attention one limitation of the EPI. The net enrollment rate and other data published annually by UNESCO in the Global Education Digest or the Education For All Global Monitoring Report are not disaggregated beyond gender and can therefore not be used to calculate the EPI. On the other hand, national household survey data, which permit the required level of disaggregation, are not collected every year but only every four or five years, on average.

Related articles

• Education disparity in South Asia
• Education data from household surveys
• Caste, ethnicity, and school attendance in Nepal
• Global Education Digest 2007
• EFA Global Monitoring Report 2008

External links

• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)

Friedrich Huebler, 1 November 2008 (edited 22 November 2008), Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/11/south-asia.html

Child labor and school attendance

A previous article on child labor on this site presented a definition of child labor that considers both economic activity and household chores. The inclusion of household chores leads to a more precise measure of the burden of work on children. In particular, this new child labor indicator is less biased against girls, who typically spend more time on household chores and less time on economic activity than boys.

In the graph below, the proposed child labor indicator is used to evaluate the trade-off between child labor and school attendance among children aged 7 to 14 years in 35 developing countries. This age group was selected because in all 35 countries children are expected to enter primary school by age 7. The underlying data were collected with 26 Multiple Indicator Cluster Surveys (MICS) and 9 Demographic and Health Surveys (DHS) between 1999 and 2005. 34 of the surveys are nationally representative and one, Palestinians in Syria, is a subnational sample. Surveys conducted during school vacation were excluded from the analysis. The results therefore show the trade-off between child labor and school attendance during a time of the year when children are supposed to be in school.

School attendance refers to attendance of any type of school and not only schools that are part of the formal system of education. In addition, children of secondary school age who are still in primary school are also counted as attending school for the purpose of the present analysis. In contrast, such overage children are counted as out of school when indicators like the secondary school net attendance rate (NAR) are calculated. In a further simplification, child labor is defined for all ages as at least one hour of economic activity or 28 or more hours of household chores per week.

Child labor and school attendance, children 7-14 years

Data source: Demographic and Health Surveys (DHS), Multiple Indicator Cluster Surveys (MICS), 1999-2005.

The scatter plot above demonstrates the trade-off between child labor and school attendance. Countries with low child labor rates typically have high school attendance rates and vice versa. A linear regression shows that a 10 point increase in child labor is associated with a 7.6 point decrease in school attendance at the national level.

On average across the 35 countries in the sample, 77 percent of 7- to 14-year-olds attended school at the time of they survey. In ten countries, at least 90 percent of children were in school. In seven countries - Central African Republic, Chad, Guinea-Bissau, Mali, Niger, Sierra Leone, and Somalia - less than half of all children went to school. Somalia has by far the lowest attendance rate with 19 percent.

25 percent of all children between 7 and 14 years were engaged in child labor, ranging from 4 percent among Palestinians in Syria to 78 percent in Niger and Sierra Leone. In six countries, more than half of all children in this age group were child laborers: Central African Republic, Chad, Guinea-Bissau, Niger, Sierra Leone, and Uganda.

Related articles

• Child labor: economic activity and household chores
• Child labor and school attendance in Bolivia
• Child labor and school attendance in Bolivia, part 2

External links

• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)

Friedrich Huebler, 5 October 2008, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/10/child-labor.html

Child labor: economic activity and household chores

Child labor is one of the obstacles on the way to the Millennium Development Goal of universal primary education by 2015. In a report on global child labor trends, the International Labour Organization (ILO) estimates that there are 218 million child laborers worldwide. 126 million of these children are estimated to be engaged in hazardous work (ILO 2006). The concept of child labor used by the ILO is derived from two conventions: ILO Convention 138, which sets 15 years as the general minimum age for employment, and ILO Convention 182 on the worst forms of child labor. Any work in violation of Conventions 138 and 182 is considered illegal child labor that should be eliminated.

One limitation of statistics like those published by the ILO is that they only refer to economic activity, that is work related to the production of goods and services, as defined in the United Nations System of National Accounts (UNSD 2001). This definition excludes chores undertaken in a person's own household like cooking, cleaning or caring for children.

Statistics of child labor that ignore household chores are problematic because they underestimate the burden of work on children, especially for girls. To examine the relative burden of economic activities and household chores carried out by children, data from 35 household surveys were analyzed for this article. Grouped by Millennium Development Region, these surveys are:

• Developed countries: Albania.
• Eastern Asia: Mongolia.
• South-eastern Asia: Lao PDR, Philippines.
• Southern Asia: India.
• Western Asia: Bahrain, Lebanon, Palestinians in Syria.
• Sub-Saharan Africa: Angola, Burundi, Central African Republic, Chad, Comoros, Congo, Côte d'Ivoire, Democratic Republic of the Congo, Gambia, Guinea, Guinea-Bissau, Kenya, Lesotho, Malawi, Mali, Niger, Senegal, Sierra Leone, Somalia, Swaziland, Tanzania, Uganda.
• Latin America and the Caribbean: Bolivia, Colombia, Dominican Republic, Nicaragua, Trinidad and Tobago.

The surveys were conducted between 1999 and 2005. 26 of the surveys were Multiple Indicator Cluster Surveys (MICS) and 9 were Demographic and Health Surveys (DHS). All 35 surveys collected data on work by children in the week preceding the survey. Surveys conducted during school vacation were excluded because the focus of the present analysis is work by children that should have been in school at the time of the survey.

The share of children aged 7 to 14 years in economic activity and household chores is depicted in the following graph. The graph also displays the number of hours spent per week on both types of work. All numbers are averages across the 35 surveys, weighted by each country's population between 7 and 14 years.

Economic activity and household chores, children 7-14 years

Data source: 35 DHS and MICS surveys, 1999-2005.

The results confirm that boys are more likely to be engaged in economic activity while girls are more likely to do household chores. On average across the 35 surveys, 22 percent of all boys and 19 percent of all girls between 7 and 14 years are engaged in economic activity. Boys also spend more hours on economic activity than girls, 20 compared to 19 hours. By comparison, girls are much more likely than boys to do household chores. 70 percent of all girls and 47 percent of all boys did household chores in the week preceding the survey. On average, girls spent 13 hours and boys 10 hours per week on household chores.

What are the implications of these findings for statistics of child labor, as currently defined by the ILO? Take the case of two families that need additional income to provide food for everyone in the household. In the first family, a 10-year-old boy is withdrawn from school and put to work on a farm. Because such work is considered economic activity the number of child laborers goes up. In the second family, the mother decides to start working on a farm and her 10-year-old daughter is asked to stay at home to care for her younger siblings. Because the girl is engaged in household chores the number of child laborers does not change. The consequences are the same for both children: they no longer go to school and miss out on the benefits from education.

To address the limitations of the ILO's definition of child labor, UNICEF has developed an expanded definition that covers household chores in addition to economic activity. This revised indicator is the basis for the child labor estimates that are reported in publications like Progress for Children (UNICEF 2007a) or The State of the World’s Children (UNICEF 2007b). For children 5 to 17 years of age, UNICEF defines child labor as follows:

• 5 to 11 years: any economic activity, or 28 hours or more household chores per week;
• 12 to 14 years: any economic activity (except light work for less than 14 hours per week), or 28 hours or more household chores per week;
• 15 to 17 years: any hazardous work, including any work for 43 hours or more per week.

The goal of UNICEF's child labor indicator is the measurement of work that should be eliminated because it violates international child labor conventions and interferes with school attendance. The threshold for household chores is set relatively high because it is assumed that household chores are less harmful than economic activity. Moreover, the high threshold of 28 hours household chores per week avoids a possible overestimation of the number of child laborers.

References

• International Labour Organization (ILO). 2006. Global child labour trends 2000-2004. Geneva: ILO. (Download PDF, 640 KB)
• United Nations Children's Fund (UNICEF). 2007a. Progress for children: A World Fit for Children statistical review. New York: UNICEF. (Download PDF, 3.6 MB)
• United Nations Children's Fund (UNICEF). 2007b. The state of the world's children 2008: Child survival. New York: UNICEF. (Download PDF, 4.3 MB)
• United Nations Statistics Division (UNSD). 2001. System of national accounts 1993. http://unstats.un.org/unsd/sna1993/toctop.asp.

Related articles

• Child labor and school attendance
• Child labor and school attendance in Bolivia
• Child labor and school attendance in Bolivia, part 2

External links

• International Programme on the Elimination of Child Labour (IPEC)
• ILO Convention 138
• ILO Convention 182
• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)

Friedrich Huebler, 7 September 2008 (edited 5 October 2008), Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/09/child-labor.html

Household wealth and years of education

At the national level, a country's wealth (measured by GDP per capita) and the education of its population (measured by school life expectancy) are highly correlated, as demonstrated in an article on national wealth and years of education. In developed countries with a high level of national income the population usually has more years of education than the population of low income countries.

A similar link can be observed at the level of individual households. Households whose members have a higher level of education are usually wealthier than households with less educated members. The relationship between household wealth and education can be analyzed with data from household surveys. This article looks at data from 12 nationally representative household surveys that were conducted between 2004 and 2006 in Bangladesh, Cambodia, Colombia, Egypt, Ethiopia, Haiti, India, Moldova, Nepal, Niger, Sierra Leone, and Zimbabwe. The data from Bangladesh and Sierra Leone is from Multiple Indicator Cluster Surveys (MICS) and the data from the other countries was collected with Demographic and Health Surveys (DHS).

DHS and MICS surveys collect data on assets owned by a household - for example, water supply and sanitation facilities, housing material, radio, telephone, refrigerator, bicycle, automobile, and livestock - that can be used to construct an index of household wealth (Filmer and Pritchett 2001). With this index it is possible to rank the households in a survey from poorest to richest. The households can then be divided into wealth deciles, each containing 10 percent of the sample population.

DHS and MICS surveys also collect data on the education of all household members above a certain age, usually 5 to 7 years. For the analysis in this article, the years of formal education of all household members aged 20 to 65 years were examined. For example, a person that did not complete primary school may have 3 years of education while someone with a university degree may have 16 years of education. In the next step, the average number of years of education within each wealth decile is calculated.

The data on household wealth and years of education is plotted in the graph below. Wealth deciles are plotted along the horizontal axis. The average number of years of education of persons aged 20 to 65 years in each wealth decile is plotted along the vertical axis. As an example, in Bangladesh, persons in the poorest decile have 1.3 years of education on average and persons in the richest decile have 10.1 years of education.

Household wealth and years of education

Data source: Demographic and Health Surveys (DHS), Multiple Indicator Cluster Surveys (MICS), 2004-2006.

The graph shows that an increase in the average years of education of all adult household members is correlated with an increase in household wealth. This relationship is true without exception in all 12 countries that were analyzed. Persons in higher wealth deciles always have more years of education than persons in lower deciles.

The graph also shows that the disparity between poorer and richer households in terms of education varies from country to country. In Moldova, almost everyone attends primary and secondary school and even in the poorest decile the average number of years of education is 9.3, compared to 13.6 years of education in the richest decile. In Zimbabwe, most persons attended at least primary school; persons in the poorest decile have 5.4 years of education on average and persons in the richest decile 11.4 years.

In contrast, Niger is a country where few persons between 20 and 65 years of age attended school. 80 percent of the population have less than 1 year of education. The average number of years of education is 0.3 in the poorest decile, 0.9 in the eighth decile, 1.8 in the ninth decile, and 5.3 in the richest decile. In Ethiopia, 80 percent of the adult population have fewer than 2 years of education and in Sierra Leone, 70 percent have fewer than 2 years of education. Cambodia and Nepal are also countries where a large part of the population has relatively little formal education.

In other countries, the increase in the number of years of education from poorer to richer deciles is more pronounced. In Egypt, persons in the poorest decile have 3.1 years of education on average and those in the richest decile have 13.8 years of education. In India, the average number of years of education is 1.4 in the poorest decile and 11.9 in the richest decile. In Haiti, the respective numbers are 1.2 and 10.7 years of education. In Colombia, the average number of years of education ranges from 3.6 in the poorest decile to 12.5 in the richest decile.

The positive link between wealth and years of education at the household level can be explained similarly to the link between these two variables at the national level. Persons with a higher level of education can earn more than those with less education. At the same time, members of wealthier households can afford education more easily than members of poorer households. At the extreme end, very poor families may not only lack the financial resources to send their children to school, they may also have to rely on the income from child labor to guarantee the survival of everyone in the household. This relationship between household wealth and child labor was analyzed in two articles on child labor and school attendance in Bolivia.

Reference

• Filmer, Deon, and Lant H. Pritchett. 2001. Estimating wealth effects without expenditure data - or Tears: An application to educational enrollments in states of India. Demography 38 (1), February: 115-132.

Related articles

• National wealth and years of education
• National wealth and school enrollment
• Poverty and educational attainment in the United States
• Poverty and educational attainment in the United States, part 2
  
• Child labor and school attendance in Bolivia
• Child labor and school attendance in Bolivia, part 2

External links

• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)

Friedrich Huebler, 24 August 2008, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/08/hh-wealth.html

Adult literacy in sub-Saharan Africa

Literacy data published by the UNESCO Institute for Statistics (UIS) in 2007 shows that the lowest adult literacy rates are observed in Africa and South Asia. In some countries, fewer than three out of ten adults can read and write. UIS provides national literacy data for two age groups: youths aged 15 to 24 years, and adults aged 15 years and older. A more detailed analysis of literacy is possible with data from household surveys.

Most Demographic and Health Surveys (DHS) collect data on literacy for persons between 15 and 49 years. For male household members, literacy data is sometimes collected up to an age of 54, 59, or 64 years. To assess the degree of literacy, respondents to the survey are asked to read a card with a simple sentence. If a respondent can read the whole sentence, he or she is counted as literate, in accordance with UNESCO's definition of literacy as "the ability to read and write, with understanding, a short simple sentence about one’s everyday life". Recent Multiple Indicator Cluster Surveys (MICS) by UNICEF collect data on literacy with the same method, but only for female household members between 15 and 49 years. MICS surveys are therefore not covered by the analysis that follows.

This article examines data from eight DHS surveys that were carried out in sub-Saharan Africa between 2003 and 2006. The survey data is from Benin (2006), Burkina Faso (2003), Cameroon (2004), Lesotho (2004-05), Niger (2006), Nigeria (2003), Uganda (2006), and Zimbabwe (2005-06). The data can be used to calculate overall literacy rates and also to examine trends over time by comparing literacy rates in different age groups.

The following table lists literacy rates for the male, female, and total population between 15 and 49 years of age. Zimbabwe (85%) and Lesotho (79%) are the countries with the highest literacy rates, followed by Cameroon (63%), Uganda (58%), and Nigeria (55%). In Benin (33%), Burkina Faso (18%), and Niger (13%), adult literacy rates are much lower. In seven of the eight countries there is a large difference between male and female literacy rates. In Benin, Burkina Faso, Cameroon, Niger, Nigeria, and Uganda, more men than women are literate, with a gender gap ranging from 12% to 26%. In Lesotho, the literacy rate of women is 21% greater than the literacy rate of men. In Zimbabwe, the difference between the male and female literacy rate is only 6%.

Adult literacy rate (%), population 15-49 years

Country	Male	Female	Total
Benin	45.1	21.8	32.6
Burkina Faso	24.3	11.9	17.6
Cameroon	70.6	54.8	62.5
Lesotho	68.8	90.1	79.4
Niger	20.9	7.5	13.2
Nigeria	68.4	42.8	54.9
Uganda	68.6	48.9	58.4
Zimbabwe	87.9	81.5	84.5

Source: Demographic and Health Surveys 2003-2006

For the graph below, the survey respondents were divided into five-year age groups. In all countries, literacy rates among the younger population are higher than among the older population. The literacy rate of the youngest group, 15 to 19 years, can be interpreted as a measure of the coverage and quality of the primary school system during the 1990s, when the members of this age group were of primary school age.

In Burkina Faso, Cameroon, Lesotho, Niger, and Uganda, the difference between the literacy rates of 15- to 19-year-olds and 45- to 49-year-olds ranges from 13% to 18%. These numbers indicate a relatively modest expansion of the education system between the 1960s, when the older age group was of primary school age, and the 1990s. In contrast, in Benin, Nigeria, and Zimbabwe, the difference between the literacy rates of 15- to 19-year-olds and 45- to 49-year-olds ranges from 27% to 32%. These three countries were thus more successful in their efforts to increase the number of literate citizens than the other five countries.

In Burkina Faso and Niger, less than one quarter of the population between 15 and 19 years can read and write. However, the case of Benin shows that a large increase in literacy can be achieved within only ten years. In Benin, only 24% of 25- to 29-year-olds are literate but among 15- to 19-years-olds the literacy rate has grown to 53%.

Adult literacy by five-year age group

Source: Demographic and Health Surveys 2003-2006

Related articles

• Educational attainment in sub-Saharan Africa
• Adult literacy rates
• Youth literacy rates
• Disparity between adult and youth literacy
• Disparity between male and female literacy rates
• Years of schooling and literacy: Can everyone with primary education read and write?
• Years of schooling and literacy, part 2
• Adult literacy in Nigeria
• Self-reported and tested literacy in Nigeria

External links

• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)
• UNESCO Institute for Statistics (UIS)

Friedrich Huebler, 10 May 2008, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/05/literacy.html

Reported and tested literacy in Nigeria

Household surveys like the Demographic and Health Surveys (DHS) or the Multiple Indicator Cluster Surveys (MICS) collect data on literacy with various methods. One approach is to simply ask respondents whether they can read and write. In an alternative approach, the reading ability is tested by asking respondents to read a sentence from a card. More detailed assessments of literacy, such as the National Assessment of Adult Literacy in the United States, require in-depth questionnaires that are beyond the scope of surveys like the DHS and MICS.

The Nigeria DHS of 2003 collected literacy data with two methods. First, the survey questionnaire contained the question: "Can (name) read and write in any language with understanding?" This question was asked for all household members aged 5 years and older, with two possible answers: yes or no (NPC and ORC Macro 2004: 247). The results of this literacy assessment were described in the article "Adult literacy in Nigeria", published on this site on 5 April 2008.

Second, a randomly selected subsample of women between 15 and 49 years and men between 15 and 59 years were asked to read a card with a simple sentence in their language. If the respondent could not read the whole sentence, the survey staff were asked to probe if the respondent could read any part of the sentence. The result was recorded as one of three options: (1) cannot read at all, (2) able to read only parts of sentence, or (3) able to read whole sentence. If no card in the respondents' language was available, or if the respondent was blind or visually impaired, the result was recorded in a separate category (NPC and ORC Macro 2004: 26-27).

The graph and table below compare the data from the two literacy assessment methods - self-reporting and reading test - for the population aged 15 to 49 years. The self-reported literacy rate, the blue bars in the graph, is the share of respondents that claimed to be able to read and write. The tested literacy rate, the red bars in the graph, is the share of respondents that could read a complete sentence; household members who could only read parts of a sentence were counted as illiterate.

Self-reported and tested literacy in Nigeria, population 15-49 years

Source: Nigeria Demographic and Health Survey 2003

A comparison of the self-reported and tested literacy rates in Nigeria shows that self-reporting tends to overestimate the degree of literacy in the population. 62 percent of men and women between 15 and 49 years were reported as able to "read and write in any language with understanding". However, the reading test revealed that only 55 percent could read a simple sentence. The difference of 7 percent can be explained by the fact that survey respondents are reluctant to admit that they themselves or other household members, on whose behalf they respond, are unable to read and write. This gap between self-reported and tested literacy can be observed across all groups of respondents: men, women, urban residents, and rural residents. The biggest difference exists in rural areas of Nigeria, were 54 percent of respondents claim to be literate but only 45 percent can in fact read.

Self-reported and tested literacy in Nigeria, population 15-49 years

	Self-reported literacy rate (%)	Tested literacy rate (%)	Difference (%)
Male	74.5	68.4	6.1
Female	50.9	42.8	8.1
Urban	77.8	72.2	5.6
Rural	53.6	45.2	8.4
Total	62.3	54.9	7.4

Source: Nigeria Demographic and Health Survey 2003

References

• National Population Commission (NPC) [Nigeria] and ORC Macro. 2004. Nigeria Demographic and Health Survey 2003. Calverton, Maryland: National Population Commission and ORC Macro. (Download report, PDF format, 4 MB)

Related articles

• Adult literacy in Nigeria
• Adult literacy rates
• Disparity between male and female literacy rates
• Years of schooling and literacy: Can everyone with primary education read and write?

External links

• Demographic and Health Surveys (DHS)
• Nigeria Demographic and Health Survey 2003
• Multiple Indicator Cluster Surveys (MICS)
• U.S. National Assessment of Adult Literacy

Friedrich Huebler, 13 April 2008, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/04/self-reported-and-tested-literacy-in.html

UNICEF releases new MICS survey data

UNICEF has released the first datasets from the third round of Multiple Indicator Cluster Surveys (MICS), conducted in 2005 and 2006. The first round of MICS surveys was carried out around 1995, followed by a second round of surveys around 2000.

MICS surveys are nationally representative household surveys that were developed by UNICEF in collaboration with other organizations to collect data on the situation of children and women. The most recent round of surveys collected data on household characteristics, education, child labor, water and sanitation, child mortality, maternal and newborn health, knowledge of HIV and AIDS, contraceptive use, birth registration, and other areas. The data can be used to track progress toward the UN Millennium Development Goals and other goals. The MICS program is described in detail on the Childinfo website of UNICEF.

In the area of education, the latest MICS surveys provide data for the following household member characteristics:

• highest level of education
• current school attendance
• school attendance in the previous year
• literacy

With the data it is possible to calculate primary and secondary school enrollment rates, repetition and dropout rates, survival rates, and other education statistics. The MICS surveys also allow detailed disaggregation of the data, for example by gender, area of residence, or household wealth. Examples for the kind of analysis that is possible with MICS data can be seen in previous articles on this site, for example on child labor and school attendance in Bolivia, educational attainment in India, or years of schooling and literacy.

The MICS datasets are available for download in SPSS format from the Childinfo website. The datasets are provided free of charge but interested researchers have to apply for a MICS username and password. At the time of writing, datasets from the following countries were available: Bangladesh, Belarus, Côte d'Ivoire, Cuba, Jamaica, Mongolia, Montenegro, Serbia, Sierra Leone, and Uzbekistan. In addition, survey questionnaires, survey reports, and a set of standard tabulations are provided. The questionnaires, tables, and reports can be downloaded without access restrictions.

External links

• Official MICS website of UNICEF
• List of MICS surveys
• Application form for MICS datasets
• UNICEF website

Related articles

• Education data from household surveys
• Education country profiles by UNICEF
• Primary school attendance in India, 2000
• Secondary school attendance in India, 2000
• Educational attainment in India, 1950-2000
• Primary school completion in India, 1950-2000
• India has 21 million children out of school in 2006
• Child labor and school attendance in Bolivia, 2000
• Child labor and school attendance in Bolivia, 2000, part 2
• Years of schooling and literacy: Can everyone with primary education read and write?
• Years of schooling and literacy, part 2

Friedrich Huebler, 24 February 2008, Creative Commons License
Permanent URL: http://huebler.blogspot.com/2008/02/unicef-releases-new-mics-survey-data.html

India has 21 million children out of school

India is the country with the largest number of children out of school. India has more children of school age than China and at the same time relatively low attendance rates, in spite of recent increases in primary and secondary school participation.

Newly released data from a Demographic and Health Survey (DHS) show that the primary school attendance rate has increased by more than one percentage point annually since the beginning of the decade. In 2000, 76 percent of all children of primary school age (6-10 years) were in school. By 2006, this value had increased to 83 percent (see Table 1). The attendance rate of girls increased by 9 percent over the 2000-2006 period and the attendance rate of boys by 6 percent. School attendance rates also grew in urban and rural areas, and across all household wealth quintiles. However, close to 17 percent of all children of primary school age continue to be out of school.

Table 1: Children of primary school age in school (percent), India 2000 and 2006

	2000	2006	Change 2000 to 2006
Male	79.2	85.2	5.9
Female	72.3	81.4	9.1
Urban	82.5	88.5	5.9
Rural	73.8	81.5	7.7
Poorest 20%	66.1	69.4	3.2
Second 20%	69.2	81.2	12.1
Middle 20%	78.8	87.5	8.7
Fourth 20%	82.1	92.2	10.1
Richest 20%	89.1	95.7	6.6
Total	75.9	83.3	7.5

Data sources: India Multiple Indicator Cluster Survey (MICS) 2000, India DHS 2005-06.

As a result of the increase in primary school attendance, the number of children out of school fell by almost one third from 30 million in 2000 to 21 million in 2006 (see Table 2). This pattern could be observed for boys and girls, and for residents of urban and rural areas. However, disaggregation by household wealth reveals that one group of children did not follow the nationwide trend. Among the poorest 20 percent of all households, the number of children out of school grew from 9.4 million in 2000 to 9.8 million in 2006. Although the primary school net attendance rate among children from the poorest households grew by 3 percentage points from 2000 to 2006 (see Table 1), this increase was not strong enough to keep pace with population growth in the poorest segment of the Indian population.

Table 2: Children of primary school age out of school (million), India 2000 and 2006

	2000	2006	Change 2000 to 2006
Male	13.0	9.5	-3.5
Female	16.4	11.2	-5.2
Urban	5.0	3.7	-1.3
Rural	24.5	17.0	-7.5
Poorest 20%	9.4	9.8	0.5
Second 20%	8.5	5.3	-3.2
Middle 20%	5.2	3.1	-2.1
Fourth 20%	4.3	1.7	-2.6
Richest 20%	2.0	0.8	-1.3
Total	29.5	20.7	-8.7

Data sources: India MICS 2000, India DHS 2005-06.

A comparison of the composition of the total population of primary school age and the population of children out of school shows which group of children are disproportionately more likely to miss out on education. Figure 1 shows the composition of the Indian population aged 6 to 10 years. 52 percent of all children in this age group are boys and 48 percent are girls. About one quarter of all children of primary school age live in urban areas and the remaining three quarters in rural areas.

Wealth quintiles are constructed by ranking the entire population of India, regardless of age, according to household wealth and dividing them into five equally sized groups with 20 percent each of the total population. As Figure 1 shows, households from poorer quintiles are more likely to have children than households from richer quintiles. Overall, 26 percent of all children between 6 and 10 years live in the bottom quintile and a further 23 percent in the second quintile.

Figure 1: Population of primary school age by sex, area of residence, and wealth quintile, India 2006

Data source: India Demographic and Health Survey 2005-06.

Figure 2: Children of primary school age out of school by sex, area of residence, and wealth quintile, India 2006

Data source: India Demographic and Health Survey 2005-06.

Figure 2 shows the composition of the group of children aged 6 to 10 years that are out of school. Although girls only account for 48 percent of the total number of children in this age group, they make up 54 percent of the children out of school. Rural children are disproportionately more likely to be out of school than urban children. Most strikingly, children from the poorest quintile make up almost half of all children out of school. 48 percent - 10 million of the 21 million children out of school - live in the poorest quintile. 74 percent of all children out of school live in the two poorest quintiles.

These numbers emphasize the close link between poverty and school attendance in India. School attendance rates have increased among the poorest households between 2000 and 2006 but the increase was not large enough to keep pace with population growth. Unless India places more emphasis on school attendance among the poor, the country will miss the Millennium Development Goal of universal primary education by 2015.

Related articles

• Primary school attendance by state in India
• 115 million children out of school
• Primary school attendance in India in 2006
• Educational attainment in India, 1950-2000
• Primary school completion in India, 1950-2000
• Education disparity in South Asia

External links

• Demographic and Health Surveys (DHS)
• Multiple Indicator Cluster Surveys (MICS)
• UN Millennium Development Goals

Friedrich Huebler, 13 November 2007 (edited 12 October 2008), Creative Commons License
Permanent URL: http://huebler.blogspot.com/2007/11/india-has-21-million-children-out-of.html