TVSEP junior researcher Mark Brooks presented his paper ‘PAPI is gone, but errors remain – Non-sampling errors in household surveys’ in the panel session ‘Methods, data and modelling’ on the second day of the annual German Development Economics Conference, which took place on the 17-18th of June. Due to technological advances in terms of data collection tools, household surveys in developing countries have transitioned from Paper and Pencil Interviews (PAPI) to Computer Assisted Personal Interviews (CAPI). CAPI is found to essentially eliminate item missing data, whereas measurement error, while being significantly reduced, remains.
The paper analyses determinants of non-sampling errors that occur throughout the interview by applying a comprehensive approach that accounts for characteristics of the interviewer and respondent as well as the interview and survey environment. Quantitative and qualitative respondent and interviewer characteristics are identified as main drivers of non-sampling error. Notably, congruency of characteristics such as ethnicity are found to be key predictors of data quality, in particular in countries with diverse cultures.
