Skip to page content
Bureau of International Labor Affairs
Previous Table of Contents Next

Data Sources

Description and Limitations of Data: Statistics on Working Children

Since its adoption in 1999, ILO Convention 182, Worst Forms of Child Labor, has enjoyed the fastest pace of ratification for any Convention in the ILO's history.  The widespread ratification of this Convention clearly demonstrates the growing global awareness about exploitive child labor and the urgency to eliminate it.  This heightened attention has led to an increased need for data and research on child labor to inform policy and program design and to set local, national, and global priorities.  As a result, numerous national household-based surveys collecting data on child labor have been conducted that can be used to estimate the extent of child labor in a given country.[46] 

In the last year, data on child labor from a number of countries have been made available through national statistical offices and international organizations such as ILO-IPEC, UNICEF, and the World Bank.  Many of the statistics cited in this report are derived from these data sources.  This year’s report provides more country level estimates on the proportion of working children than in previous years’ reports; however, there are still a large number of countries for which statistics on working children are unavailable. 

Estimates of the number of working children in a country, particularly those engaged in the unconditional worst forms of child labor, can be difficult to obtain.  There is no internationally endorsed definition of working children, or a universally prescribed methodology for collecting data on children’s work.  Therefore, the lack of universal concepts and methods for collecting child labor data makes it difficult to present comparable and unambiguous estimates across countries on working children.  In addition, this lack of agreement on how to define and measure children’s work also detracts from the credibility of existing estimates.  In general, estimates on the number of working children are likely to be underestimates because the nature of household surveys do not lend themselves to collecting data on children who are working in the informal or illegal sectors of the economy, particularly children in the unconditional worst forms of child labor, such as armed conflict, commercial sexual exploitation, illicit activities, slavery, and forced or bonded labor.  In addition, the number of girls working is often underestimated because statistics often exclude girls working as unregistered domestic helpers or as full-time household helpers for their parents. 

Data collected on children’s work usually measure economic activity and may include acceptable forms of work for children of legal minimum working age.  Economic activity covers most productive activities, including market or non-market production, paid or unpaid work, and work in the formal or informal sectors.  In line with international definitions of employment, if a child worked at least 1 hour during the survey reference week, he/she is considered to be economically active.  Because surveys of children’s work most often include children ages 5 to 14, the individual country profiles in this report include an estimate of working children for this age group in the main text. Where available, estimates on the number of working children ages 15 to 17 years are included in a footnote.  In a few cases where statistics on the 5 to 14 age range are unavailable, the age ranges vary slightly.  Once again, it is important to bear in mind that the statistics on working children may not be comparable from one country to another since the sources used to produce estimates of working children use different methodologies and apply various definitions.

The three main sources of data used in this report are listed below.  Priority is given to statistics collected from national-level household surveys that were designed specifically to collect data on children’s work.  Therefore priority is given first to ILO-IPEC SIMPOC[47] surveys and then to UNICEF MICS[48] surveys.  Finally, for countries that do not have data on working children from any of these sources, estimates are drawn from ILO’s Economically Active Population Estimates and Projections: 1950-2010, which are published annually in the World Bank’s World Development Indicators.  In the few countries where no recent surveys relating to child work or child labor have been conducted, estimates of working children are not provided.

Sources of Data on Working Children

Statistics on working children in this report were obtained from the following three sources, in order of priority of use:

ILO-IPEC’s SIMPOC-sponsored Child Labor Surveys

Since its inception in 1998, SIMPOC has provided technical assistance to over 40 countries in the collection, processing, and analysis of data and information on children’s work and child labor.  SIMPOC has assisted in the production of over 250 child labor-related surveys, including 55 national surveys with a focus on children in economic activities; over 80 baseline surveys; more than 100 rapid assessments in specific areas and sectors where child labor was perceived to be acute; and numerous establishment, street, and school-based surveys.[49]  In SIMPOC surveys, the population of working children generally includes children ages 5 to 17 years[50] who are either salaried workers, unpaid workers in family enterprises, self-employed, or apprentices.  In addition, unlike traditional labor force surveys, the SIMPOC surveys collect data on some non-market work activities and work in the informal sector, including fetching water, collecting firewood, or street peddling.  Estimates of working children, however, do not include children engaged in domestic chores in their own household.  Generally, SIMPOC considers a child to be involved in domestic chores if he/she has reported to be engaged in such activities as cooking, doing the laundry, and taking care of siblings.  Since most of the SIMPOC-sponsored surveys cited in this report were conducted by national statistical offices in each country, the estimates of working children are attributed to these entities.

UNICEF-sponsored Multiple Indicator Cluster Surveys (MICS)

UNICEF began assisting countries in assessing progress for children in relation to the World Summit for Children goals through its Multiple Indicator Cluster Surveys (MICS) in 1998.[51]  The MICS questionnaire includes 75 indicators for monitoring children’s rights such as child labor, child survival and health, child nutrition, maternal health, water and sanitation, and education.  UNICEF defines child labor, not just children’s work, as: (a) children 5 to 11 years of age involved in at least 1 hour of economic activity in the preceding week or 28 hours or more of housekeeping chores in their own household;[52] or (b) children 12 to 14 years of age involved in economic activities for 14 hours or more or 42 hours of combined economic activity and housekeeping chores.[53]  More than 50 developing countries have included an indicator of child labor in their MICS questionnaire; as of November 30, 2004, 56 countries had submitted their national reports to UNICEF.[54] 

ILO Estimates and Projections of the Economically Active Population, 1950-2010

In cases where data from ILO-IPEC SIMPOC surveys or UNICEF MICS surveys were unavailable, child workforce participation rates are reported based on data from the ILO’s Estimates and Projections of the Economically Active Population (EAP): 1950-2010, which are available from the World Bank’s World Development Indicators 2004 or the ILO’s on-line database for labor statistics (LABORSTA).[55]  Statistics from the ILO’s EAP database are compiled from a variety of sources, including national population censuses and household surveys.  The EAP estimates differ from those in ILO’s SIMPOC child labor surveys in that they are based on the definition of the “economically active population” for children ages 10 to 14.[56]  Although the EAP estimates are less accurate for working children below the age of 15, they are often the only available source for comprehensive and comparable data on working children ages 10 to 14 years in many countries.[57]  Since the EAP estimates were compiled by the ILO, the estimates of working children used in this report are cited to both the ILO and the entity that maintains the actual database, such as the World Bank.

Statistics on Primary Education

In addition to estimates of working children, statistics on primary school attendance, primary school enrollment, and the percentage of children who began primary school who were likely to reach grade five are used in this report to provide complementary indicators of the number of children who work or are at-risk of working.  Where available, statistics on the percentage of children who started primary school who were likely to reach grade five are also included.  Primary attendance statistics enable the reader to infer the proportion of children in the school-age population who are not in school and may be engaged in exploitive child labor or at risk of entering hazardous work activities.  Although primary school attendance rates are more accurate than primary enrollment rates in illustrating the extent of exploitive child labor in a country, enrollment rates are more frequently collected and readily available for the majority of countries.  In addition, enrollment rates provide an indication of the availability of and interest in basic education in a country.  Therefore, primary enrollment rates are included in each country profile, and when available, recent primary attendance rates are also reported. 

There are several limitations to using primary enrollment rates as an indicator of exploitive child labor that should be kept in mind.  Primary enrollment rates reflect the number of children who are enrolled during a given school year out of the total school-age population, but do not reflect the number of children actually attending school.  Thus, a child can be enrolled in school, but never attend.  As a result, primary enrollment rates often overstate the true number of children who attend school on a regular basis, and understate the number of children who may be working.  In other cases, children who are enrolled in or attending school may also be engaged in work outside of school hours, also leading to an underestimate of children’s work.[58]  Nevertheless, to the extent that child labor and education are linked, it is beneficial to examine any data that provide a measure of children’s access to and participation in schooling, particularly in rural and impoverished areas.

Sources of Primary School Education Data

Primary school education data on gross and net primary school enrollment, repetition, and completion originate from the UNESCO Statistics Institute and were obtained from either the World Bank’s compilation of World Development Indicators 2004 or UNESCO’s Global Education Digest 2004.  Data on the percentage of primary school entrants (first graders) who reach grade five also were taken from these sources.[59]  Gross and net primary school attendance rates were mostly obtained from USAID’s Global Education Database.[60]

World Development Indicators 2004  (WDI 2004)

WDI 2004 is a World Bank publication that compiles development data from several international and government agencies, the private sector, and nongovernmental organizations around the world.  WDI 2004 includes 800 indicators on topics in six areas:  world view, people, environment, economy, states and markets, and global links.  There are 85 tables covering the six categories with basic indicators on 224 countries. 

Rates of primary enrollment, survival to grade 5, and repetition in the country profiles rely on data presented in the WDI 2004, which were compiled by the UNESCO Institute for Statistics from official responses to surveys and from reports provided by education authorities in each country.[61] 

Education statistics generally provide only a limited picture of a country’s education system.  Statistics often lag by two to three years, though an effort is being made to shorten the delay.  Moreover, coverage and data collection methods vary across countries and over time within countries, so the results of comparisons should be interpreted with caution.  For example, the U.S. Department of Labor’s 2003 Findings on the Worst Forms of Child Labor may have cited education data for a previous year compiled in the WDI 2003; however, statistics presented in this year’s report for the same year from the WDI 2004 may differ slightly because of statistical adjustments made in the school-age population or corrections to education data.  In other instances, there was no change from the WDI 2003 to the WDI 2004 because education statistics were not affected by the adjustments or corrections to the data were not needed.[62]

Global Education Database (GED)

The Global Education Database (GED) is sponsored by the Office of Education of the U.S. Agency for International Development (USAID).  The GED is a repository of international education statistics compiled from the UNESCO Institute for Statistics (UIS) and the Demographic and Health Surveys (DHS), a USAID program that has conducted full-scale nationally representative household surveys in over 60 developing countries since 1984.  There are 134 indicators compiled from the UNESCO Institute of Statistics and 71 indicators compiled from the Demographic and Health Surveys (DHS).  DHS data include rates for gross and net primary school attendance and persistence disaggregated by gender and rural/urban residence.  The DHS are also one of the only sources of comparable data across countries on primary school attendance.  The UIS data include indicators on primary school enrollment, persistence, and repetition rates, public expenditure, pupil/teacher ratios, and gender parity.  With over 200 countries represented, the database is a useful tool for cross-country comparisons of education indicators and to assess the education performance of a specific country or groups of countries over time.[63] 

Global Education Digest 2004 (GED)

The data contained in the GED 2004 have been collected from national experts in some 200 countries and then cross-checked and entered into the statistical database by the UNESCO Institute of Statistics.  Data include education indicators on the performance of a specific country in areas such as school enrollment, persistence, and repetition rates, public expenditure, pupil/teacher ratios, and percent of trained teachers.  The data currently found in the database are from the years 1970 to 2001.  Education data have been subject to the new International Standard Classification of Education (ISCED97) model since the school year starting in 1998.  Therefore, data from the pre-1998 period are not presented in the country profiles since they are not comparable with data for the 1998 to 2001 period.[64] 

[46] For a detailed discussion regarding definitions of and data sources on children’s work see Amy R. Ritualo, Charita L. Castro, and Sarah Gormly, "Measuring Child Labor: Implications for Policy and Program Design," Comparative Labor Law and Policy Journal 24 2 (2003).

[47] The International Labor Organization’s International Program on the Elimination of Child Labor maintains a Statistical Information and Monitoring Program on Child Labor (SIMPOC) office, which assists countries in generating comprehensive data on child labor.

[48] UNICEF helps countries assess progress for children in relation to the World Summit for Children goals through its Multiple Indicator Cluster Surveys (MICS).

[49] ILO-IPEC, Development of Statistical Child Labor Standards for consideration by the 18th International Conference of Labor Statisticians, project document, Geneva, August 30, 2004, 4-6.

[50] In some cases, this report uses calculations based on SIMPOC data in order to standardize across countries the age ranges for which percentages of working children are provided.

[51] UNICEF, End Decade Assessment: Multiple Indicator Cluster Survey (MICS), [online] [cited December 1, 2004], Background; available from

[52] To determine whether children are involved in housekeeping chores, the survey includes the following question: “During the past week, did (name) help with housekeeping chores such as cooking, shopping, cleaning, washing clothes, fetching water, or caring for children?”

[53] UNICEF, UNICEF - Definitions: Child protection, [online] 2004 [cited November 24, 2004], Child labour; available from

[54] UNICEF, End Decade Assessment - MICS2 - National Reports, [online] 2004 [cited December 1, 2004]; available from

[55] See below for more information on the World Bank’s World Development Indicators.

[56] World Bank, World Development Indicators 2004 [CD-ROM], Washington, D.C., 2004.

[57] ILO, Source and Methods: Labour Statistics, Vol. 10: Estimates and Projections of the Economically Active Population 1950-2010, Geneva, 2000.

[58] Despite the hazardous nature of some work activities, it is common for children to engage in child labor as a source of income in order to afford the additional costs of going to school. As a result, many children combine school and work, which often hinders a child’s performance at school.

[59] The UNESCO Statistics Institute measures survival to grade five because research suggests that 5 to 6 years of schooling is a critical threshold for the achievement of sustainable basic literacy and numeracy skills. See World Bank, World Development Indicators 2004.

[60] See the glossary of this report for definitions of gross and net primary enrollment and attendance, repetition, and completion.

[61] World Bank, World Development Indicators 2004.

[62] Ibid.

[63] USAID Development Indicators Service, Global Education Database, [online] [cited October 13, 2004]; available from

[64] UNESCO Institute for Statistics (UIS), Global Education Digest 2004: Comparing Education Statistics Across the World, [CD-Rom] 2004.