This study used principal components analysis to examine the structure of variables associated with pathological gambling. A large battery of questionnaires was administered to a sample of 141 gamblers who ranged from non-problem gamblers to severe pathological gamblers. We found a significant relationship between severity of pathological gambling and various measures of impulsivity, depression, anxiety, erroneous beliefs, and reports of early wins. Component analysis of these variables found four distinct components: emotional vulnerability, impulsivity, erroneous beliefs, and the experiences of wins. Component scores based on these components were regressed onto pathological gambling. Emotional vulnerability had the largest effect (beta = 0.54), followed by early wins (beta = 0.32), erroneous beliefs (beta = 0.31), and impulsivity (beta = 0.23). The overall model accounted for 53.4% of the variance of pathological gambling. The findings confirm the idea that there may be several different risk factors that explain the development of pathological gambling.