JUNE 18, 2003

Of the impact of questionnaire length on questionnaire completion in on-line surveys

SPSS inc. and Burke inc. teamed up to study the impact of questionnaire length on questionnaire completion in on-line surveys.

Researchers have every reason to believe that premature terminations increase as questionnaires lengthen. This is the case in telephone surveys and in mail surveys. Why should it be any different in on-line surveys? SPSS and Burke set up an interesting research plan to document the impact of questionnaire length in this setting. They submitted similar, but varying-length questionnaires to three groups: a 15-minute questionnaire to individuals who had accepted a questionnaire of this length when they enrolled on an on-line panel, a 30-minute questionnaire to individuals who had accepted such a questionnaire and the same 30-minute questionnaire to individuals who had accepted 15-minute questionnaires. They observed termination rates and final sample bias.

The results indicate that premature termination rates were two to three times higher in the latter group than in the former two. They also point to significant biases in the profile of individuals who terminated prematurely — based on key indicators of product concept appeal and intention to use a service. The authors conclude that on-line surveys should be kept short, that standardized surveys could help alleviate the problem analytically, that key indicators should be placed early in the questionnaire and that "completed questionnaires" may need to be redefined in the industry.

The authors have made a PDF presentation of their findings available on the Web.

 Return to the home page
 List all news