Studying Cannabis Use Behaviors With Facebook and Web Surveys: Methods and Insights

Abstract

The rapid and wide-reaching expansion of internet access and digital technologies offers epidemiologists numerous opportunities to study health behaviors. One particularly promising new data collection strategy is the use of Facebook’s advertising platform in conjunction with Web-based surveys. Our research team at the Center for Technology and Behavioral Health has used this quick and cost-efficient method to recruit large samples and address unique scientific questions related to cannabis use. In conducting this research, we have gleaned several insights for using this sampling method effectively and have begun to document the characteristics of the resulting data. We believe this information could be useful to other researchers attempting to study cannabis use or, potentially, other health behaviors. The first aim of this paper is to describe case examples of procedures for using Facebook as a survey sampling method for studying cannabis use. We then present several distinctive features of the data produced using this method. Finally, we discuss the utility of this sampling method for addressing specific types of epidemiological research questions. Overall, we believe that sampling with Facebook advertisements and Web surveys is best conceptualized as a targeted, nonprobability-based method for oversampling cannabis users across the United States.

Addressing Public Health Emergencies via Facebook Surveys: Advantages, Challenges, and Practical Considerations

“…It is important to note that our study is not the first to use targeted Facebook advertisements for participant recruitment in health research. Earlier research has used this approach to address topics such as smoking behavior [6], cannabis use [7], and mental health [8]. Furthermore, Whitaker et al [9] and Thornton et al [10] have performed systematic reviews of related literature.…”

Mechanical Turk data collection in addiction research: utility, concerns and best practices

“…Due to the nature of the platform, workers are also typically computer-savvy, an effect which is exacerbated among the most prolific workers by the need for browser extensions and supportive scripts to help workers reduce down-time between HITs and maximize earnings. However, MTurk has the potential (as does other on-line, social media-based recruitment [26]) to offer a non-probability-based method for oversampling participants with particular conditions relevant to addiction, such as heavy drinkers or individuals reporting chronic pain-similar to flyer-based community recruitment of individuals with a history of substance use. These sampling methods must be carefully considered to recruit ideal participants (e.g.…”

Digital Clinical Trials for Substance Use Disorders in the Age of Covid-19

“…Social media, internet, and other online platforms may provide opportunities and larger, more diverse pools of prospective participants as location limitations are eroded if physical attendance in the clinic is not necessary or is less frequent. 16 , 17 Recruitment across a state or potentially from multiple states enhances diversity by allowing for inclusion of individuals in rural areas or participants who have had trouble with transportation to research sites and who historically have not been well represented in clinical research. 18 In doing so, researchers will need to address challenges in working competently with unfamiliar communities.…”

“…Digital recruitment strategies can target characteristics outlined in inclusion criteria such as demographics (eg, age, sex) and the substance of interest (eg, “alcohol treatment”). 17 , 16 However, this form of advertising can be expensive as popular “key words” such as “addiction” drive up costs, and many researchers have limited experience in optimizing these platforms. 20 Partnerships with advertising experts or hiring staff with past experience in this area may be fruitful.…”