You are on the Natcen site

Click here for Scotcen

natcen map

You are on the Natcen site

Click here for Scotcen

natcen map
06 September 2018 | City, University of London

Conducting probability based mixed-mode surveys

Experimental evidence from the European Values Study


Recent developments in many countries such as falling response rates, longer field periods and rising costs challenge the position of face-to-face surveys as the “gold standard” for comparative research programs.

At the same time access to the internet is steadily rising, triggering a considerable academic interest in survey modes which avoid human intermediaries in data collection, especially web and mail modes, often combining both in a mixed-mode design. Yet, research on the differences between representative face-to-face and self-administered survey modes are still lacking for many countries, so it is unclear in how far mixed-mode surveys can be considered a viable alternative to face-to-face surveys, especially for relatively long questionnaires.

This study presents experimental evidence comparing a probability based mixed-mode survey (mail, web) with a face-to-face survey in Germany as part of the European Values Study (EVS) 2017. For this, we have randomly split a probability-based register sample into several experimental groups: The first group has received the full F2F survey, the second group has received the full survey in a mixed-mode design (mail, web) and the last group has received a reduced questionnaire in a mixed-mode matrix design, so respondents had to answer on average only about 60% of the original questionnaire. The first objective of this study is to provide experimental evidence on differences between response rates, representativeness, data quality and cost factors across the modes. Secondly, we also address the question which mixed-mode design works best in Germany. To answer this question we evaluate a series of experiments, testing the effects of incentives (5 Euro prepaid vs 10 Euro postpaid) and contact mode (sequential, simultaneous) on the aforementioned dimensions.

Where: ELGO3, Drysdale Building, City, University of London, Northampton Square, EC1V OHB

When: 5.45pm - 7.15pmThursday 6th September 2018

Speaker: Pablo Christmann, GESIS - Leibniz Institute for the Social Sciences

Series: NatCen-ESS ERIC-City methodology seminar series

Sign up here.

Back to Upcoming events