Do Predictors of Gambling Involvement Differ Across Male and Female Adolescents?

Sensitivity checking was conducted between the sample of adolescents (N = 3,767) who had less than 50% missing on both the gambling measure and the sample of adolescents who had more than 50% missing data on either of the gambling behavior measures (n = 3,253). Mean comparisons using one-way ANOVAs were based on demographics (age and gender) and variables/behaviors that appeared before the mid-part of the survey (e.g., school grades, attachment to mother and to father, depression, self-esteem). Due to the large sample size, many comparisons based on group averages were statistically significant. In general, the excluded group was slightly younger, comprised of a slightly greater percentage of males, and reported lower average grades. A discriminant function analysis was conducted to determine the extent to which the two samples could be differentiated based on these variables. Together, the demographic and intra/interpersonal variables explained only 2.2% of the separation between groups, Wilkes Λ = .978, χ² = 149.67, df = 9, p < .001. Discriminant function analyses also were conducted using the full sample (N = 7,290) with missing data imputed. Analyses revealed consistent results with the smaller sample (n = 3,767) suggesting that, when compared to participants excluded from the sample due to missing data, the sample of respondents in this study was not meaningfully different in terms of demographics or inter/intrapersonal variables.