Quality & Quantity 33: 411–428, 1999.
© 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Nonresponse and Recall Errors in a Study of
Absence because of Illness: An Analysis of Their
Effects on Distributions and Relationships
H. VAN GOOR and A. L. VERHAGE
Department of Sociology, University of Groningen, Grote Rozenstraat 31, 9712 TG Groningen,
Abstract. Using administrative data as validating standard, we studied the combined effects of two
sources of survey error – nonresponse and recall errors – on distributional and substantive bias in a
mail survey of absence because of illness among the employees of a Dutch road building company
(response rate 77%). No distributional bias was found in ﬁve socio-demographic variables (sex, age,
years of service, function, and district), but both nonresponse bias and recall bias occurred in our
central dependent variables: frequency and duration of absence because of illness. Nonrespondents
were on sick leave more frequently and longer than respondents. Furthermore, the self-reports of
absence because of illness of our respondents proved to be rather inaccurate. Underreporting of
frequency and duration of sick leave was more common than overreporting. Therefore, both sources
of error had a cumulative effect.
While nonresponse did not result in biased relationships, recall errors had clearly biasing con-
sequences: seven out of 30 correlation coefﬁcients analyzed were too biased to produce valid
outcomes; another six were substantially biased. Multiple regression used for predicting recent ab-
sence because of illness among our respondents also led to different outcomes depending on the
choice of data source (administration or questionnaire) for our absence variables.
Key words: nonresponse bias, recall errors, employees, absence because of illness.
Survey research is by far the data collection method used most frequently in social
sciences. Surveys are attractive because they allow us to ask the people themselves
about their attitudes and behavior. Furthermore, data on large numbers of indi-
viduals can be gathered and analyzed rather quickly and inexpensively. However,
surveys have their limitations, being susceptible to different kinds of measurement
errors. These sources of error are generally studied independently, but different
types of errors can be interrelated. For instance, it is frequently assumed that