A Qualitative Study on the Effects of Grouped versus Interleafed Filter Questions

Abstract

Filter questions are used in many surveys to avoid asking follow-up questions that do not apply to every respondent. Often, detailed follow-up questions are asked immediately after a filter question is endorsed. This design is called interleafing. In contrast, other surveys use a ‘grouped’ approach where several yes/no filter questions are asked first, and the follow-up questions for any filter question that was endorsed are asked later in the survey. Previous research has found that when filter questions are grouped together before follow-up questions, respondents report significantly more “yes” responses than in the traditional interleafed design. Literature also suggests that a decrease in the quality of follow-up responses may be a trade-off of the grouped filter question design. Using a subset of sections from the Consumer Expenditure Survey, we conducted a cognitive study on the effects of grouping versus interleafing filter questions. We present qualitative evidence that the grouped format may increase cognitive burden on respondents and suggest that could be a reason for the observed decrease in response quality to follow-up questions in the grouped filter question design.