Momentary versus retrospective reports of alcohol or cannabis use, sexual activity, and their co-occurrence

Objective: Comparing people's momentary and retrospective reports of substance use and sexual activity can illuminate discrepant recall biases across these behaviors. Extant research suggests that people tend to underreport alcohol use on retrospective surveys and overreport sexual activity. We provided an updated account of these biases, extending previous work by looking at alcohol- and cannabis-involved sexual activity as well as potential gender differences in recall biases.

Method: Using a sample of adults (n = 110; 58.2% women), we administered surveys three times a day for 28 days to measure momentary alcohol and cannabis use, sexual activity, and substance-involved sexual activity. At the end of this momentary assessment, participants completed a retrospective survey assessing how frequently they engaged in these behaviors during the 28-day period.

Results: We compared participants' momentary reports-which were scaled to account for compliance rates-and retrospective surveys. While there were no significant differences in momentary and retrospective reports of alcohol or cannabis use, participants reported higher rates of sexual activity and alcohol-involved sexual activity on the retrospective surveys than the momentary reports. Effect sizes for significant differences were medium to large (Cohen's d: 0.26-0.67).

Conclusions: Alcohol- and cannabis-involved sexual activity tend to be overreported on retrospective surveys, and preliminary findings suggest that these recall biases may vary by gender. Researchers interested in the co-occurrence of substance use and sexual activity should be aware of this potential random error and consider how to reduce recall biases based on method of data collection.