ORIGINAL ARTICLE Does Questionnaire Distribution Promote Blood Donation? An Investigation of Question–Behavior Effects Anne van Dongen, MA & Charles Abraham, PhD & Robert A. C. Ruiter, PhD & Ingrid J. T. Veldhuizen, PhD Published online: 18 November 2012 # The Society of Behavioral Medicine 2012 Abstract Purpose The purpose of this study is to assess the effective- ness of survey administration as a population-level interven- tion to increase blood donation. Methods Study 1 was a randomized controlled trial of new donors comparing 3,518 who received a questionnaire and 3,490 who did not. Study 2 compared matched, randomly selected samples of active donors; 5,789 received a ques- tionnaire, while 6,000 did not. In both studies, the depen- dent measure was the proportion of donors who attended a blood donation center to give blood within 6 months of survey posting. Study 3 compared data across five similar trials. Results No difference in volunteering to give blood was observed between those who did and did not receive a questionnaire among either new or active donors, confirm- ing the findings of two other Canadian trials. Conclusions Despite earlier optimistic findings, there is little evidence to suggest that survey administration per se will be effective in increasing blood supplies. Implications for behavior change mechanisms are discussed. Keywords Question–behavior effect . Mere measurement effect . Nonresponse bias . Blood donation . Behavior change Introduction Researchers routinely ask people what they think about particular behaviors in order to characterize respondents or predict their future behavior. Such measurement is often assumed to be neutral but may shape respondents cognitions and behavior. Answering questions may alter one’ s evalua- tion of an envisaged action, thereby changing attitudes. Recording beliefs and intentions may also modify the sa- lience of pre-existing attitudes, norms, or goals and so shift respondents’ priorities. Such “question–behavior effects” (QBEs) [1] have been reported in laboratory and field stud- ies. Thus the question arises as to whether QBEs can be used to induce population-level, behavior change. Sherman [2] observed that respondents systematically overpredicted the likelihood of engaging in socially desir- able behaviors and underpredicted the likelihood of performing socially undesirable behaviors, and, importantly, that these self-predictions shaped future behavior. For example, respondents asked about volunteering for charity were more likely to do so than those who were not ques- tioned. Sherman concluded that “asking people what they would do in a certain situation assures that they will act that way, even if they would have acted differently without the asking” (p. 215). Various process explanations of such QBEs have been considered. Studies of cognitive dissonance [3] demonstrate that writing evaluations, even in private, can shift attitudes towards the evaluated action (e.g., [4]). Similarly, becoming aware of discrepancies between what we have done and what we should do can prompt action (e.g., [5]). Thus, answering questions about our behavior and intentions may create dissonance, which questionnaire respondents are motivated to resolve by acting in accordance with their answers. Responding to questionnaires may also activate pre-existing cognitions. For example, the theory of planned A. van Dongen (*) : I. J. T. Veldhuizen Unit Donor Studies, Sanquin Research, P.O. Box 1013, 6501 BA Nijmegen, The Netherlands e-mail: a.dongen@sanquin.nl A. van Dongen : R. A. C. Ruiter Maastricht University, Maastricht, The Netherlands C. Abraham University of Exeter Medical School, University of Exeter, Exeter, UK ann. behav. med. (2013) 45:163–172 DOI 10.1007/s12160-012-9449-3