Publication type
Survey Futures Working Paper Series
Series Number
111
Series
Survey Futures Working Paper Series
Authors
Publication date
December 1, 2025
Summary:
Live video interviewing (VI) emerged as a mode for conducting large-scale social surveys during the COVID-19 pandemic but there is limited existing evidence on measurement differences between VI and in-person interviewing. This paper addresses this evidence gap by investigating measurement mode effects in a large-scale birth cohort study with large samples of respondents participating via VI and in-person, and a wide range of survey items included in the analysis of distributional differences. Due to non-random allocation to survey mode, propensity score weighting (PSW) was used as a method for causal inference, with carefully selected confounders associated with the mode of participation (exposure) and key survey items (outcomes). After PSW adjustment, the magnitude of measurement differences between VI and in-person interviewing is minimal and appear largely attributable to residual sample composition effects arising from unobserved confounders. We detected slightly more candid responses from VI participants (regarding political views and psychological stressors) and marginally higher cognition scores, likely due to the absence of a physically present interviewer. Overall, the evidence suggests that VI can be considered a suitable alternative to in-person interviewing from a measurement perspective, offering advantages such as cost-efficiency and the potential to increase overall response rates in mixed-mode surveys.
Subjects
Link
https://surveyfutures.net/working-papers/
#588983