Using crowdsourcing to examine behavioral economic measures of alcohol value and proportionate alcohol reinforcement

Online crowdsourcing websites such as Amazon's Mechanical Turk (MTurk) are increasingly being used in addictions research. However, there is a relative paucity of such research examining the validity of administering behavioral economic alcohol-related measures, via an online crowdsourcing platform. This study sought to validate an alcohol purchase task (APT) for assessing demand and a questionnaire measure of proportionate alcohol reinforcement, using an online sample of participants recruited via MTurk. Participants (N = 865, 59% female) were recruited via MTurk to complete the APT, proportionate alcohol reinforcement questionnaire, Alcohol Use Disorders Identification Test (AUDIT), and demographics. Responses on the APT were highly systematic (<3% nonsystematic data) and conformed to prototypical demand curves. Correlation analyses revealed significant associations among AUDIT total scores with a majority of the alcohol demand indices (r values .08 -53, p values < .05) as well as proportionate alcohol reinforcement, r = .43, p < .001. Regression analyses controlling for relevant covariates indicated that intensity, BP, O_max, elasticity, and reinforcement ratio predicted significant variance in AUDIT scores. This study further supports the use of online crowdsourcing websites for investigating behavioral economic determinants of alcohol misuse.