‘‘It Is Time Computers Do Clever Things!’’: The Impact of Dependent Interviewing on Interviewer Burden

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in AbstractThis paper evaluates the impact of dependent interviewing (DI) on interviewer burden and data quality using qualitative data collected from a survey carried out in 2006 on the British Household Panel Survey (BHPS) interviewers. We find that: (i) DI has a minor effect on interviewer burden, (ii) this effect is perceived by interviewers to be positive, (iii) the mechanisms by which DI reduces interviewer burden are mainly indirect as they are mediated by respondents, and (iv) in most cases the effects of DI on interviewer burden varies in relation to the type of DI questions asked and respondent circumstances.Acknowledgement:

Propensity to consent to data linkage: experimental evidence on the role of three survey design features in a UK longitudinal panel

“…This suggests that respondents tend to be consistent with their previous answers when answering survey questions. We can speculate, in accordance with findings from other studies (Sala et al, 2011), that DI may facilitate the response process. The results from this experiment lead us to formulate the following recommendation.…”

Total Survey Error for Longitudinal Surveys

“…One specific issue associated with dependent interviewing concerns the editing of previous wave verbatim responses before feeding them forward into the script for the next interview. It is important that the respondent should recognize their responses, so language should not be changed [59]. For example, the verbatim description of a job should be fed forward, rather than the description of the occupation to which it was subsequently coded, even if the aim is to verify that the same code still applies.…”

Can I Just Check...? Effects of Edit Check Questions on Measurement Error and Survey Estimates

“…A potential cause of the findings for transitions onto and off receipt might be found with the interviewers. With DI designs, they might be more focused on reducing errors in continued receipt than on picking up transitions onto and off receipt (Sala et al 2009). Since the number of transitions onto and off receipt is small, we would however interpret these results with caution.…”