Utilizing content-knowledge questionnaires to assess study eligibility and detect deceptive responding

Abstract

ABSTRACTBackground: Deceptive responding during eligibility screening presents a significant concern for assessing inclusion/exclusion criteria. This problem is compounded in settings for which biomarkers or other objective verification (e.g., urinalysis) are not feasible.Objectives: Introduce and describe content-knowledge questionnaires as an objective method for collaterally assessing study eligibility.Methods: Participants (N = 3772; 66.1% female) recruited using the crowdsourcing resource Amazon Mechanical Turk (mTurk) completed a Cannabis Knowledge Questionnaire (CKQ). The CKQ contained four-items indexing knowledge of typical cannabis costs, weights, and terminology. Self-reported cannabis use history was collected and compared to individual item and total scale scores. A separate in-laboratory assessment evaluated participants during in-person screening for cannabis, alcohol, and cocaine research protocols (N = 43).Results: Good internal consistency (α = .74) was observed. The most common correctly answered question was about dabbing (41.4%) followed by cannabis cost (37.6%), hybrid strains (36.6%), and estimated weight (29.7%). Current cannabis use was associated with large effect size increases in the rate of correct responses (RR = 3.64) as well as odds of a correct response on individual items (OR = 5.88–21.48). In the laboratory study, participants with a positive urine drug test for cannabis or those reporting lifetime regular cannabis use scored higher than those without this history (RR = 1.89–2.61).Conclusion: These findings highlight the efficiency and efficacy of including content-knowledge questionnaires for collateral assessment of study eligibility, especially when biomarkers are not possible. Future studies will be useful for extending this initial demonstration to alternative settings and substances.