Assessment tools used to identify the variables or conditions believed to maintain gambling behaviors are currently limited to indirect assessment methods such as the Gambling Functional Assessment (GFA; Dixon & Johnson, 2007) and the Gambling Functional Assessment-Revised (GFA-R; Weatherly, Miller, & Terrel, 2011). To date, no experimental functional analysis has been conducted targeting gambling behavior. Similarly, little is known regarding the convergent validity of the GFA, GFA-R, and experimental functional analyses with gambling behavior. The current study developed an experimental functional analysis for use with gambling behavior(s) in order to assess the convergent validity results between the experimental functional analysis and both the GFA and GFA-R. Participants completed a demographic questionnaire, the South Oaks Gambling Screen (SOGS; Leisure & Bloom, 1987), the GFA, and the GFA-R. Gambling behavior was observed during brief, recurrent phases of exposure to a series of analog conditions. Each condition differed along variations of the following dimensions: (a) attention, (b) access to tangibles, (c) escape from aversive stimuli, and (d) sensory stimulation. Both inter-response time and frequency of spins were compared to the control condition for all 20 participants.
Results of the experimental functional analysis (EFA) were compared with both indirect assessments utilized within the study. Results found that the GFA matched for inter-response time 30% of the sample and the GFA-R for 40% of the sample. Additionally, the GFA matched with the EFA for frequency of spins for 25% of the sample and with the GFA-R for 45% of the sample.