Online Program

Return to main conference page

All Times ET

Program is Subject to Change

Tuesday, June 15
Tue, Jun 15, 11:30 AM - 1:00 PM
TBD
Understanding, Assessing, and Transitioning to Mixed Mode Collections

Differences in response patterns between telephone and push-to-web administration of a cross-national survey of establishments (308186)

Eva Dimitrova, Eurofound 
*Gijs van Houten, Eurofound 

Keywords: survey mode, response patterns, data quality, cross-national, repeated cross-sectional

The European Company Survey is a voluntary survey among a random probability-based sample of private sector establishments with at least 10 employees in all European Member States. The target respondents are the manager responsible for human resources and when possible an employee representative.

The survey has been carried out four times so far. Up until the third edition in 2013 the survey was administered by telephone. The fourth edition in 2019 used a push-to-web survey design. Initial contact was made via telephone, administering a short CATI screener questionnaire in which the respondent(s) were identified, and subsequently inviting the respondent(s) to complete the substantive questionnaire online.

Part of the rationale underlying the decision to change the mode of administration was the common criticism of telephone surveys that these are more prone to desirability effects than other modes of administration. More generally, data quality in telephone surveys is expected to be lower than in other modes of administration.

This paper focuses on response patterns and response styles. To the extent that the data allow us, we identify response patterns (e.g. primacy, recency) and response styles (e.g. extreme responses, acquiescence). We subsequently analyse the differences between the 2013 CATI survey and the 2019 push-to-web survey, in terms of the incidence and magnitude of response patterns and styles, the differences in response patterns and styles between different groups of establishments (e.g. country, sector, size), and the impact of these response styles and patterns on group comparisons. To the extent that response patterns and styles can be assumed to be constant over time, this will provide use some insight in whether the choice to move from telephone to push-to-web has resulted - and more generally, whether the online administration of establishment surveys results - in higher data quality.