Integrating Survey Data from Different Sources: Comparability, Quality, and Applications

Combining survey data from different sources promises new research opportunities but also entails challenges related to the comparability of the data collected with different survey instruments, sampling designs, fieldwork procedures, quality assurance protocols, and characterized by different survey outcomes (e.g., response rates). Addressing these challenges is necessary to fully exploit the potential of, e.g., (1) survey data collected decades ago when current survey quality standards were not yet in place, (2) traditional probability sample surveys that struggle to achieve satisfactory response rates, and (3) the increasingly popular unrepresentative “big surveys”. In addition to sampling, the second crucial source of limitations to comparability comes from the measurement instruments (i.e., question-wording and response scale formats), which can be addressed by standardizing the coding in the data preparation phase or accounting for the differences in measurement in the modelling phase.

This session is dedicated to solutions to these challenges in data integration, including methodological innovations and applications. Welcome contributions include, but are not limited to, papers that:

1. propose methods for accounting for differences in various aspects of data quality and comparability when combining data from different sources,
2. propose methods for accounting for cross-project differences in the measurement instruments,
3. propose measures of overall survey quality or its various aspects,
4. use different survey data sources to examine a substantive problem while addressing or problematizing the potential for limited comparability of the data.

This session is intended as a regular session.