There is evidence that some forms of gambling are more strongly linked to addictive behaviour. But these findings are not consistent across different studies. Most gambling prevalence studies use cross-sectional data, which can be influenced by selection bias. This report is important because it reviews the use of matching methods, (e.g., propensity score matching, coarsened exact matching, and sample selection models) in gambling prevalence studies. This report also includes an analysis of 9 British gambling prevalence surveys. The authors used these approaches to identify which gambling activities are associated with the most severe individual gambling harms. They found that engagement in some activities (i.e., online gambling, slots) is linked to greater problem gambling. Also, the results suggest it is important to control for selection biases in gambling research.
