Knowledge of random events and chance in people with gambling problems: an item analysis

ABSTRACT This paper examines the items of two scales, the Random Events Knowledge Test (REKT) and the Chance Test, and examines their relationship with problem gambling (N = 1375). Using exploratory and confirmatory factor analysis, the REKT was broken down into four sub-scales: Due to Win, Counterintuitive Nature of random chance, Odds Do Not Improve, and Biases and Wins. The Chance Test was broken down into three sub-scales: abstract Odds, Table Odds, and Chance Odds. These sub-scales were regressed onto of problem gambling severity and revealed that more knowledge about random chance on all sub-scales of the REKT and Abstract Odds from the Chance Test were negatively related to problem gambling. On the other hand, we found that higher score on the Table Odds and Chance Odds from the Chance Test were positively related to problem gambling. The results illustrate that compared to people who do not have a gambling problem, problem gamblers have a more accurate understanding of some aspects of the chances of winning specific games, but have a poorer understanding of various implications of the independence of random events. The findings suggest potential strategies for the prevention and treatment of problem gambling.