Gambling and the trait of addiction in a sample of New Zealand university students
New Zealand Journal of Psychology, Jun 2003 by Clarke, Dave
Continued from page 1.
From their research and from other empirical studies reported in the literature, Gupta and Derevensky (1998; Gupta 2000) examined the possibility of an addictive personality trait among adolescent problem gamblers. For example, multiple addictions have been found among more than half of adolescents who have a compulsive behaviour problem (Griffin-Shelley, Sandier & Lees, 1992, in Gupta and Derevensky, 1998). Compulsive problems can include substance abuse, food, sex, relationships and gambling. From earlier surveys (e.g., Sharma, 1995, in Gupta and Derevensky, 1998) the trait of addiction among adolescents precedes the addiction itself; in other words, addiction to an activity does not create the addictive personality.
Eysenck and Eysenck (1975) have devised a scale from the original Eysenck Personality Inventory (EPQ) to measure the addictive personality trait. Their Addiction scale has been found to correlate with neuroticism and various drinking measures (Patton et al., 1994). Other investigators (Blaszczynski et al, 1985) reported similarities between heroin addicts and pathological gamblers on the Addiction scale, but did not examine if addiction was still involved in problem gambling when other variables were controlled. Hence, it was decided to include addiction in the present study and in the regression equation for the classification of problem gamblers.
The first aim of the present study was to compare the prevalence of gambling and problem gambling in a group of university students with a similar age group in a recent, representative New Zealand national sample, and with university students in other countries. The second aim was to examine the variables associated with problem gambling using the present student group.
On the basis of the literature reviewed above, the following hypotheses were generated. Compared to no problem gamblers, problem gamblers will:
Hypothesis 1. gamble more frequently and on more activities;
Hypothesis 2. spend more money on any one bet in the last twelve months;
Hypothesis 3. be more likely to think that their parents gambled excessively;
Hypothesis 4. be more involved in continuous gambling activities;
Hypothesis 5. have greater depression; and,
Hypothesis 6. be stronger on the addiction trait.
A seventh hypothesis was:
Hypothesis 7. the addiction trait will distinguish problem gamblers from non-problem gamblers, after controlling for the other variables noted in Hypotheses 1 to 6.
A gambling questionnaire was completed by a class of 137 female and 34 male students who were enrolled in an introductory psychology course at the Albany campus of Massey University in 2001. The age distribution was moderately skewed (0.98), with no outliers. Ages ranged from 15 to 57 years, with a median age of 23 years (M = 27.90, SD = 10.49). Seventy-two percent of the sample identified themselves as Caucasian or New Zealand European, 7.6% as Maori, 6.4% as Pacific Islander, and 9.4% as Asian. Most of the participants (84%) were in the average and above average socio-economic groups. Six of the students (all female) had never gambled. Their data were excluded from further analysis because there were too few of them to warrant meaningful comparisons to the two gambling groups. Thus, there were 165 students who had gambled for money at least once on any activity in their lifetimes, and their data were used in this analysis. The sample consisted predominately of young, female, Caucasian adults with moderate to high socio-economic status.
An anonymous questionnaire consisting of five sections was given to the participants. The sections appeared in the order below.
Prevalence of Gambling. The participants indicated which of 13 gambling activities they played for money at least once in their lifetime, and if they gambled, the frequency for each activity during the past 12 months, with categories of 0 (not in the last 12 months), 1 (less than monthly), 2 (monthly), 3 (weekly), or 4 (daily). The activities appear in Table 1. From the data, two scores were calculated for each person: the total number of activities tried at least once in their lifetime, ranging from 1 to 13, andfreqitency, the total ratings for all the games played during the past 12 months. For gambling frequency, the 13 rating scales could be considered equivalent to Likert five-point scales, ranging from "never" to "very frequently", so that total scores reflect a continuous range from 0 to 52. Respondents also indicated the largest amount of money gambled in the last 12 months: 1 ($1 or less), 2 (more than $1, up to $10), 3 (more than $ 10, up to $49), 4 ($50 to $99), 5 ($ 100 to 199), or 6 ($200 or more). To compare problem to non-problem gamblers and for logistic regression, the data were re-classified into two categories: $10 or less, and more than $10.
Problem Gambling. Problem gambling and respondents' perceptions of parents' excessive gambling were measured by the Revised South Oaks Gambling Screen (SOGS-R). The SOGS-R is the same 20-item questionnaire as the original SOGS which screens for problem and pathological gambling, except that it surveys over a limited time frame (Abbott & Volberg, 1999; Ladouceur et al., 1997). The SOGS-R used in the present study asks questions on a "yesno" basis about problems associated with gambling in the last 12 months, rather than in one's lifetime, for comparisons with the national sample, also based on gambling within the past year. It is also more reliable and produces fewer false negatives than the SOGS (Abbott, 2001). A score of 3 or more indicates that a person is at least a problem gambler, and less than 3, a non-problem gambler. Probable pathological gamblers have a score greater than 4, and were included in the present study as problem gamblers. Abbott and Volberg (1996) have provided some support for the reliability and validity of the SOGS-R with New Zealand samples. For the present sample, the coefficient of internal consistency was .94. Although it is not included in the SOGS-R scoring scheme, one item asks the respondents if they thought that either of their parents gambled excessively. Respondents circled either "yes", no", or "don't know". The item was used to measure parents' gambling behaviour as a dichotomous variable, with either "yes" or "no" defining the dichotomy.
Addiction. The Eysenck Addiction scale (Eysenck & Eysenck, 1991) was developed from the EPQ by comparing the responses of 221 drug addicts to those of 310 normal people. It consists of 32 items from the original Psychoticism (P), Extraversion (e), Neuroticism (N), and Lie (L) scales answered on a yes-no basis. Addicts score high on P and N, and low on E and L. The coefficient of internal consistency for the present sample was .69.
Depression. The depression symptom inventory is an existing instrument developed with medical inpatients and outpatients, and respondents in cross-sectional surveys in New York State (Bell et al., 1982). It was used rather than the Beck Depression Inventory (BDI; Beck et al., 1961) because the BDI was designed to assess the severity of depression in clinically diagnosed groups (Davison & Neale, 2001), whereas the depression symptom inventory is more applicable for non-clinical populations. Each item is rated on a five-point scale, from "often" to "never", with scores ranging from 18 to 90. For the present study, item one was re-worded from "Do you feel in good spirits?" to "Do you have a feeling of well-being?", to avoid Maori connotations relating to the spirits of their ancestors; and item 18 from "How does the future look to you?" to "Does your future seem uncertain to you?" to make sense in terms of the rating scale. The modified depression scale had an internal consistency coefficient of .82 for the present sample.
Demographics. The participants were asked to indicate their gender, age, which socio-economic group best described their family background (low, moderately low, average, moderately high, high) and which ethnic group they primarily identified with (Caucasian, New Zealand European or Pakeha, New Zealand Maori, Pacific Island group, Asian, or Other).