The Consumer Pyramids Household Survey (CPHS), conducted by the Centre for Monitoring Indian Economy (CMIE), is the largest longitudinal household panel in India designed to capture high-frequency socio-economic dynamics. Initiated in 2014, it fills the gap left by infrequent official surveys such as the NSS and Census, providing continuous, nationally representative data through thrice-yearly waves of household interviews. This handbook offers a guide to understanding and using CPHS data, focusing on its stratified multi-stage sampling design, weighting methodology, and operationalisation in statistical analysis. It explains how strata, primary sampling units, and ultimate sampling units are selected, and how weights are constructed to ensure representativeness despite challenges of non-response, attrition, and non-surveyed areas. The guide also demonstrates how to implement the survey design in RStudio. Beyond technical exposition, it highlights considerations for researchers working with subsets, building panels, dealing with rare events, and establishing causal inference, especially in cases where standard applications of weights may not be appropriate. Rather than prescribing fixed rules, the document underscores the need for careful judgement, adaptation, and transparency when analysing CPHS data. Its aim is to support rigorous and credible research using this uniquely rich but complex dataset.
Click here to access the handbook.
