Data Bias towards Better-off Households
STORIES, ANALYSES, EXPERT VIEWS
Data in India has been a casualty. This impacts evidence based policy decisions. This is largely, write Jean Druze (visiting professor, Department of Economics, Ranchi University) and Anmol Somanchi (independent researcher) “because of the paucity of national household surveys. The National Sample Survey’s (NSO) latest consumption survey, ignoring the suppressed 2017-18 round, goes back to 2011-12. The District Level Household Surveys and Annual Health Surveys have not been heard of since 2013. The latest National Family Health Survey, NFHS-5…” is only partially available and the decennial census and the second Socio-Economic and Caste Census (SECC) was due this year.
In the circumstance, the Consumer Pyramids Household Survey (CPHS) of the Centre for Monitoring Indian Economy (CMIE), conducted three times a year since January 2014, “has become a veritable barometer of the Indian economy, closely watched for data on income, expenditure and employment in particular.”
It is based on ‘an all-India representative sample of over 170,000 households. There are doubts whether CPHS is a ‘robust, nationally representative and panel survey of households.’ For example, according to CPHS, adult literacy (15-49 years) was 100% in urban areas and 99% in rural areas in late 2019. “That is too good to be true. It suggests that CPHS is biased towards better-off households,” say the two authors.
Another example of the bias is on literacy rates estimates. The CPHS estimate of adult literacy (15-49 years) in 2015 is 6 percentage points higher than the NFHS-4 estimate for 2015-16, for both men and women. The bias is also evident from data on household assets. According to CPHS, for instance, 98% of households had electricity in late 2015, 93% had water within the house, 89% had a television, and 42% had a fridge. The corresponding figures from NFHS-4 are much lower: 88%, 67%, 67% and 30% respectively.
In short, conclude the two authors “far from being nationally representative, the CPHS sample is heavily biased towards better-off households, and quite likely, the bias is growing over time….”