Download Data »
How many younger schoolchildren are in private schools? (ages 7-12)

Update, June 2011:
Data was updated in June 2011 to reflect more recent information on purchasing power parity calculations. Data on website may therefore not match identically with what is available in the current (April 2011) edition of Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. Future editions of the book will incorporate these changes. If you have any questions regarding the data, please contact comments@pooreconomics.com.

Data Sources:
Most of the surveys were developed and executed in conjunction with the World Bank Living Standards Measurement Study (LSMS). Instructions for obtaining the raw data can be found at http://iresearch.worldbank.org/lsms/lsmssurveyFinder.htm. The Family Life Surveys (Bangladesh, Indonesia and Guatemala) were conducted in collaboration with the Rand Corporation. More information is available at http://www.rand.org/labor/FLS.html. We administered the Udaipur and Hyderabad data sets with our collaborators, Rachel Glennerster and Angus Deaton. Both data sets are available through http://www.povertyactionlab.org/evaluations/data. Further information on data sources can be found in the data download package.

Methodology:
We constructed each variable of interest at the household level. Since some of the surveys asked different variants of some of the questions, we tried to define each variable in the most comparable way possible across countries. We normalized all household expenditure variables to a daily per capita level. Using the PPP conversion methodology discussed in the foreword, we converted the local currency units into 2005 world rupees. We then applied the daily poverty line cutoffs to each household's total per capita expenditures to classify households into the relevant expenditure categories. All of the tables display variable means across households within each expenditure group weighted according to the household weights provided in most surveys.