Consistency of data collected through online life history calendars

Abstract

The collection of longitudinal data is crucial in some domains such as life course studies. However, prospective studies are considerably costly, and thus retrospective data are an appealing alternative. A life history calendar is a tool specifically conceived to collect retrospective data. However, although it is designed to enhance the recall process of the respondents, the accuracy of the data collected through this approach remains unknown, particularly when data is collected online. In this study, we conducted a secondary analysis of data collected from n = 5,181 respondents through an online survey regarding their sexual health. Because we inquired about the occurrence of certain events twice during the survey, once using a life calendar and once through a traditional questionnaire, we were able to perform three types of consistency checks: (1) reporting of single events, (2) age when the events occurred and (3) correct timing between two events. The main results indicated that it is generally more difficult to remember the exact age of occurrence of an event than the event itself, that the report of related events is generally coherent, and that women are generally more accurate in their answers than men. Based on our results, it is therefore possible to identify a subset of persons whose answers are more consistent throughout the survey. This study also shows that data obtained through an online life history calendar can be of a quality similar to that obtained through a traditional online questionnaire.