Recall Error in the Year of Retirement

Mitarbeiter in diesem Projekt:

Recall error is an important aspect when discussing the quality of survey data. Nevertheless, it is hard to assess recall error as this requires validation data which are often not available. This project is based on the linked dataset SHARE-RV, which combines the survey data of the Survey of Health, Ageing and Retirement in Europe (SHARE) with administrative records of the German Pension Fund. An overlap of information included in both datasets allows for learning more about recall error in single variables, such as the year of retirement. This project aims to assess the recall error in the year of retirement, which is the deviation between the self-reported year and the year provided by the administrative data. Based on Tourangeau’s “Psychological Model of the Response Process” different groups of determinates will be identified, which increase the likelihood of a misreport. The preliminary results show that most of the respondents remember the year they retired correctly. But nevertheless, there is also a considerable proportion of errors. Some determinants (as characteristics of the event and respondents cognitive abilities) had been identified which increase the likelihood of a misreport, but the error seems not to be systematic, meaning that no pattern of reporting the event too late or too early exists.

Aus diesem Projekt hervorgegangene Publikationen:

News & Events

New Social Policy and Law Shared Database

Online now!


SHARE User Conference 2019 in Budapest, Hungary

Programme online



Latest Newsletter

MEA is coordinating