Is shorter always better? Relative importance of questionnaire length and cognitive ease on response rates and data quality for two dietary questionnaires

• Published in: American journal of epidemiology Vol. 153; no. 4; pp. 404 - 409
• Main Authors: SUBAR, Amy F, ZIEGLER, Regina G, THOMPSON, Frances E, JOHNSON, Christine Cole, WEISSFELD, Joel L, REDING, Douglas, KAVOUNIS, Katherine H, HAYES, Richard B
• Format: Journal Article
• Language: English
• Published: Cary, NC Oxford University Press 15-02-2001; Oxford Publishing Limited (England)
• Subjects: Aged; Biological and medical sciences; Diet; Diet Surveys; Epidemiology; Feasibility Studies; Feeding Behavior; Female; General aspects; Humans; Male; Medical sciences; Methodology; Middle Aged; Public health. Hygiene; Public health. Hygiene-occupational medicine; Randomized Controlled Trials as Topic; Research Design - standards; Surveys and Questionnaires - standards; Surveys and Questionnaires - statistics & numerical data; Time Factors; United States; North America; America; Human; Questionnaire; Methodology; Diet; Duration; Sanitary surveillance; Epidemiology; Public health; Feeding
• ISSN: 0002-9262; 1476-6256
• DOI: 10.1093/aje/153.4.404

In this study, the authors sought to determine the effects of length and clarity on response rates and data quality for two food frequency questionnaires (FFQs): the newly developed 36-page Diet History Questionnaire (DHQ), designed to be cognitively easier for respondents, and a 16-page FFQ developed earlier for the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial. The PLCO Trial is a 23-year randomized controlled clinical trial begun in 1992. The sample for this substudy, which was conducted from January to April of 1998, consisted of 900 control and 450 screened PLCO participants aged 55-74 years. Controls received either the DHQ or the PLCO FFQ by mail. Screenees, who had previously completed the PLCO FFQ at baseline, were administered the DHQ. Among controls, the response rate for both FFQs was 82%. Average amounts of time needed by controls to complete the DHQ and the PLCO FFQ were 68 minutes and 39 minutes, respectively. Percentages of missing or uninterpretable responses were similar between instruments for questions on frequency of intake but were approximately 3 and 9 percentage points lower (p < or = 0.001) in the DHQ for questions on portion size and use of vitamin/mineral supplements, respectively. Among screenees, response rates for the DHQ and the PLCO FFQ were 84% and 89%, respectively, and analyses of questions on portion size and supplement use showed few differences. These data indicated that the shorter FFQ was not better from the perspective of response rate and data quality, and that clarity and ease of administration may compensate for questionnaire length.