A recent national survey (Amey, 2001) found that proportionately more young people in New Zealand gambled on cards, scratch tickets and gaming machines than older age groups. From North American surveys, young urban males in college are at risk of being problem gamblers. The present study compared the prevalence of gambling and problem gambling among a sample of 171 first year university students with the prevalence found nationally and internationally. The students, median age 23 years, completed a questionnaire consisting of demographic items, questions about gambling behaviour, the Revised South Oaks Gambling Screen (SOGS-R), a depression inventory, and the Eysenck Addiction scale. Approximately 97% of the sample gambled for money and 16% of the 165 gamblers were classified as problem gamblers. Maori gambled on more activities than the European/Pakeha, Pacific Island or Asian groups did, but there were no other significant sociodemographic differences. There were significant differences between problem and non-problem gamblers on gambling frequency, number of activities, amount gambled, parents' gambling, continuous gambling, depression and Addiction. Logistic regression analysis showed that after controlling for gambling frequency, number of activities, amount gambled, continuous gambling and parents' gambling, Addiction was a significant predictor in discriminating between problem and non-problem gamblers. Combining initiatives in the prevention and treatment of drug and gambling addiction was suggested.
In a previous study (Clarke & Rossen, 2000) the prevalence of adolescent gambling and problem gambling recalled by a sample of New Zealand university students was compared with that of Reid & Searle's (1996) representative national sample. Generally, the students reported that they were more involved during adolescence in all forms of gambling for money than either the national sample or samples of adolescents overseas were. The sample also had a larger proportion of problem gamblers than the other groups.
Problem gambling is typically defined by three or more symptoms on the widely-used South Oaks Gambling Screen (SOGS; Abbott & Volberg, 1996, 1999; Lesieur & Blume, 1987). Probable pathological gambling is defined by five or more symptoms. The inclusion of "probable" is to distinguish SOGS' respondents from pathological gamblers identified in clinical interviews (Abbott, 2001). The SOGS is a 20-item, self-report questionnaire based on DSM-III-R criteria. It consists of questions about problems associated with gambling in the respondent's lifetime on a "yes-no" basis. The symptoms include trying to recoup losses, borrowing or stealing money to gamble, and escaping from dysphoric moods (anxiety, depression). However, the symptoms do not include frequency of gambling, range of activities, or amount of money spent on gambling. Nine of the items are sources of borrowing money, so that effectively only 12 items measure distinct symptoms of problem gambling. Abbott and Volberg (1996) have provided data to support the validity of the SOGS' criteria for problem and probable pathological gambling in New Zealand.
In a partial replication of the earlier survey, the first purpose of the present study was to ascertain the extent of gambling in a group of first year university students compared with a similar age group in the most recent representative national sample (Amey, 2001). The second purpose was to compare problem gamblers to non-problem gamblers on frequency of gambling, number of different gambling activities, amount gambled, continuous gambling, parents' gambling, depression and the trait of addiction. The third purpose was to assess the relative importance of each of these variables in predicting the classification of problem gamblers.
Problem versus Non-problem Gamblers
A number of variables have been associated with problem gambling. …