Stop With the Questions Already! Does Data Quality Suffer for Scales Positioned Near the End of a Lengthy Questionnaire?

Research questionnaires frequently include dozens—if not hundreds—of self-report items. Lengthy questionnaires, however, are often a necessity. In some cases, they are needed to assess the many variables found in a complex model; in other cases, they are the result of the inclusion of a single lengthy measure. This raises an important question: Do participants provide accurate responses to measures positioned at the end of a lengthy questionnaire? One possibility is that participants experience fatigue during questionnaire completion, leading them to engage in careless responding, and thus compromising the accuracy of their responses. Another possibility is that even the longest research questionnaires are generally too short to evoke participant fatigue. This latter possibility suggests that participants are largely able to maintain their attention while completing most questionnaires. Given the lack of clarity on this issue, we conducted two experiments (Study 1 N = 244; Study 2 N = 461) in which we randomly assigned each participant to complete a block of target scales at either the beginning or the end of a lengthy (> 300-item) questionnaire. Each participant also recruited an informant who provided reports of the participant's personality, attitudes, and behaviors. These informant data allowed us to examine the effects of the experimental manipulation on the target scales' convergent and criterion-related validity. The findings of both studies indicated that the target scales performed similarly across the two conditions. Given the ubiquity of lengthy questionnaires, these findings have far-reaching practical implications. [ABSTRACT FROM AUTHOR]