Conferences

In the European Social Survey (ESS) response rates are significantly different across both countries and rounds. While between-country differences have been studied extensively elsewhere (Philippens and Billiet 2003, Philippens et al. 2003, Billiet and Philippens 2004, Billiet and Pleysier 2007), only few analyses investigate effects between the rounds (Symons et al. 2008), and none dealt with possible learning effects of survey agencies and in particular interviewers who participated during several ESS rounds.
In the current paper, we analyze possible performance gains from the use of experienced interviewers working for the same survey agency. We investigate the most important components of response (obtaining household contact, target person contact, target person cooperation) in the ESS according to whether they are produced by senior interviewers who are also used for refusal conversion calls, interviewers that have experience from previous rounds, and “usual” interviewers in their first round. We use data from the contact forms of four countries that participated during the three rounds conducted so far, used the same survey agency, and partially the same interviewers. To analyze interviewer effects, we use discrete two-level models.