Lengthening the interview time in an annual longitudinal survey like Understanding Society may not put participants off being interviewed the following year. That’s according to new research by leading survey methodologist Professor Peter Lynn.
In a new Working Paper, Professor Lynn, based at the Institute for Social and Economic Research, home to Understanding Society, examines whether participants who are surveyed for 31 minutes compared with 26 minutes are more likely to refuse to be interviewed on the next occasion (in the case of Understanding Society approximately 12 months later).
He found no differences in the proportions of interviewees who were subsequently willing to complete an additional questionnaire or take part in further interviews. Explaining the background to the research, Professor Lynn said:
“Survey designers must decide how much content should be included in a questionnaire. There is often pressure to add additional questions, but the longer time that it takes a respondent to complete the interview or questionnaire may impose greater cognitive burden, greater discomfort or greater disruption to their other activities and this in turn may mean they decide not to participate in future interviews.”
The study made use of Wave 1 of Understanding Society’s Innovation Panel, which was set up for the purposes of methodological testing and experiments. Two equal sized groups of survey participants were given either the “shorter” or the “longer” interview.
The experiment showed no differences between the groups in completion rates for the self-completion questionnaires which interviewees were subsequently asked to fill in. Nor was there any significant difference between the groups when it came to whether or not they agreed to take part in subsequent annual waves of the survey.
Commenting on the findings, Professor Lynn said:
“Survey designers often have to balance the analytical advantages of adding a few extra minutes of questioning against the perceived disadvantages of increasing the respondent burden by making the interview longer. The evidence presented here suggests that the disadvantage in terms of subsequent participation propensity may be negligible or non-existent.”
this news has been re-posted from the Understanding Society website