(PS7-A11) A Pilot Test of Reddit Data Collection Among Under-sampled LGBTQ+ Community Members

As research inclusive of the LGBTQ+ community continues to expand in conjunction with increased public representation, the field of psychology has an increased interest in collecting data from a wide range of identities within the LGBTQ+ community. However, many identities remain difficult to sample from traditional methods due to low trust in research and a lack of representation within sampled populations, such as college students. The lack of access to these communities necessitates novel sampling methods adapted to existing limitations. Many LGBTQ+ community members with under-sampled identities utilize online platforms in order to find community with others who share their identities, making targeted online recruitment an ideal method for recruiting underrepresented individuals.  
In this study, we tested an online recruitment strategy focused on sampling under-sampled identities within the LGBTQ+ community via Reddit. We collected a list of LGBTQ+ subreddit forums emphasizing under-sampled identities and designed a recruitment protocol to promote community engagement. After identifying subreddits, we sent a research request to the subreddits’ moderators accompanied by the survey itself. After receiving permission from moderators, we began posting the recruitment materials directly into posts on each subreddit. We allocated a $100 gift card per every 250 respondents.

In five days we collected 135 responses from LGBTQ+ participants on their alcohol use over the past month. The data cost approximately two dollars a person. We collected data from an almost equal proportion of males and females (43.7% versus 39.3%) with 5 participants indicating being assigned intersex. Of the 36 gender identity choices offered, 14 were endorsed. Additionally, only 27 (20%) reported recent alcohol use, which is low compared to the general population (over 50%; SAMHSA, 2019). The mean number of drinks reported during a typical week was 8.4 (SD = 14.6) with a mode of 1. Participants endorsed, on average, about 8 consequences (M = 7.9, SD = 5.9). Typical drinks was significantly correlated with consequences experienced (r = 0.61, p </em>< 0.001).

Fewer people in our sample reported 30-day alcohol use compared to the national average. This challenges notions of high alcohol use among individuals with minoritized gender identities. Among those who did report alcohol use in the last 30 days, most reported drinking infrequently. Those who reported the heaviest use reported the most consequences from their alcohol use.
The recruitment protocol encouraged community participation in research effectively with the following benefits: the data cost less than some other sampling methods, promoted direct involvement of under-sampled identities, and allowed for direct communication resulting in feedback from participants on the structure and content. We were able to reorganize portions of the survey based on community opinion to address factors limiting participation among some identities. This study both demonstrated the effectiveness of recruiting under-sampled LGBTQ+ identities via Reddit and provided guidance on how researchers may adapt studies to be more accessible for participants that may feel alienated by scientific study.