Does length of questionnaire matter? A randomised trial of response rates to a mailed questionnaire

J Health Serv Res Policy. 2000 Oct;5(4):219-21. doi: 10.1177/135581960000500406.


Objective: To assess whether length of questionnaire affects response rates.

Methods: A quasi-randomised trial of women aged 70 years and over in a general practice in England. Three questionnaires of different lengths: a clinical questionnaire (four pages); the same questionnaire plus the EuroQol (five pages); the same questionnaire plus the SF-12 (seven pages). The impact of length on the proportion of returned questionnaires and item completion rates was assessed.

Results: In total, 847 questionnaires were mailed; response rates were 49%, 49% and 40% to the short, medium and long questionnaires, respectively. This difference was statistically significant when the short questionnaire was compared against the longest instrument (9% difference; 95% confidence interval (CI) of difference = 0.3% to 16.6%). Item completion rates for the clinical questionnaire did not differ. Respondents did not differ in age or self-reported health status between the three groups.

Conclusions: Increasing the length of a questionnaire from five to seven pages reduces response rates from women aged 70 years and over. However, lengthening a questionnaire does not seem to affect the quality of responses to questions near the front of the questionnaire.

Publication types

  • Clinical Trial
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • England / epidemiology
  • Female
  • Fractures, Bone / epidemiology
  • Health Surveys*
  • Humans
  • Osteoporosis / complications
  • Patient Participation*
  • Random Allocation
  • Retrospective Studies
  • Risk Factors
  • Surveys and Questionnaires / standards*