The Prevalence of Pathological Gambling in Canada Robert Ladouceur
Laval University, Canada
This paper critically reviews prevalence estimates of problem and pathological gambling in Canada. Populations studied are adults, adolescents and primary school children. T h e proportions of pathological gamblers found in Canadian studies (ranging from 1.2% to 1.9% for adults) are similar to prevalence rates reported in the United States. Given the apparent link between gambling availability and increases in the prevalence of problem and pathological gambling, it is h o p e d that provincial and federal authorities in Canada will make investments in research and treatment of pathological gambling in the future.
As in the United States, gambling in Canada has expanded rapidly in the past two decades. Before 1970, legal gambling in Canada was restricted to occasional charity bingos and raffles, midway games of chance, parimutuel wagering on horse races and friendly bets between individuals (Campbell & Lowman, 1989). By 1993, legal gambling in Canada had expanded to include slot machines and video gaming devices, casinos, large-scale bingo operations, sports wagering and offtrack betting on horse races. Lotteries, bingo and parimutuel wagering are available in every Canadian province; casinos now operate in more than half of the provinces. Clearly, Canada is following the trend in The author wishes to thank Mark Freeston for his helpful comments. Portions of this paper were written while the author had grants from Counseil de Recherche en Sciences Hnmaines du Canada (410-91-1514) and from Loto-Quebec. Send correspondence to Robert Ladouceur, Ph.D., #-cole de Psychologie, Universit~ Laval, Quebec, Quebec, Canada, GIK 7P4.
Journal of Gambling Studies "CoL12(2~ Summer 1996 9 1996 Human Sciences Press, Inc.
129
130
JOURNAL O F GAMBLINGSTUDIES
Western industrialized countries toward the approval and legalization of gambling. Studies estimating the prevalence of pathological gambling in Canada are now available for six provinces: Alberta, New Brunswick, Nova Scotia, Ontario, Quebec and Saskatchewan. While prevalence surveys have been completed in British Columbia as well as in Manitoba, the reports on these surveys were not available at the time this paper was written. Prevalence estimates in both of these provinces are similar to prevalence estimates in other Canadian provinces. This paper will address the following topics: (1) prevalence estimates of adult pathological gambling in Canada; (2) prevalence estimates of adolescent problem and pathological gambling in Canada; and (3) a description of gambling behaviors among primary school children in Quebec. The results of these studies have been and will continue to be useful in evaluating the need for treatment and prevention services for pathological gamblers as well as the need for public education and information campaigns concerning the danger of becoming an excessive gambler. Quebec is the only province where prevalence studies of pathological gamblers were conducted without the specific request of a gove r n m e n t agency. In all other provinces, private companies or individuals were asked by provincial authorities to provide estimates of problem and pathological gambling in order to establish policy toward the legalization or expansion of various forms of gambling.
PATHOLOGICAL GAMBLING AMONG ADULTS Prevalence estimates of problem and pathological gambling among adults are available for six Canadian provinces. Detailed results of these studies are presented in Table 1 in the chronological order in which they were conducted.
Province of Quebec The first Canadian prevalence study was conducted in 1989 (Ladouceur, 1991) and was based on the South Oaks Gambling Screen developed by Lesieur and Blume (1987), adapted and extensively used by Volberg and her colleagues in their seminal work on the prevalence
131
ROBERT LADOUCEUR
Table I Prevalence Rates of Problem and Probable Pathological Gambling in Canada
Ever Response Gambled Rate Problem Pathological Total QuebeO New Brunswick 2 Nova Scotia 2 Alberta 2 Saskatchewan 1 Saskatchewan 2 Ontario 3
88% 87% 80% 93% 87% 87% 67%
68% 59% 34% 50% ? ? 65%
2.6 3.13 3.0 4.0 2.8 1.9 7.7
1.2 1.37 1.7 1.4 1.2 0.8 0.9
3.8 4.5 4.7 5.4 4.0 2.7 8.6
1Lifetime rate ~Current rate (one-year) Slt is not clear if these r a t e s a r e current or lifedme rates. In addition, the cut-off points for classifying respondents are different from those used in other Canadian studies.
of problem and pathological gambling in the United States. For the Quebec survey, the South Oaks Gambling Screen was translated for French-speaking Canadians (about 80% of the population in the province of Quebec speak French). According to the criteria for the South Oaks Gambling Screen, respondents scoring 3 and 4 points were classified as "problem gamblers" and those scoring 5 or more points were classified as "probable pathological gamblers." The Quebec sample was stratified to proportionally represent the population (aged 18 and over) from each area of the province. Random selection of telephone numbers from telephone books and r a n d o m selection of respondents within the households were used. One person from each household was interviewed. A total of 1,002 interviews were completed. Up to five attempts were made to contact each number. The response rate was 68%. The overall results indicated that 2.6% of the Quebec sample could be classified as "problem gamblers" and 1.2% could be classified as "probable pathological gamblers." If we sum the two figures, 3.8% of the adult population in Quebec have at some time had a moderate to severe problem with gambling. The typical problem gambler in Quebec is a single male, under the age of 30, with an income between $15,000 and
132
JOURNAL OF GAMBLINGSTUDIES
$25,000. The interval between the age at which problem and probable pathological gamblers began wagering and the age at which they became preoccupied by the amount they were wagering was 8.8 years. In contrast to many United States jurisdictions, a substantial proportion of problem and probable pathological gamblers had consulted or had already sought help for their gambling problem (8.9% and 2.9%, for a total of 11.8%). Since lotteries are the most popular form of legalized gambling in Quebec, it is not surprising that 88% of the respondents had purchased a lottery ticket at sometime in their life. Since the time this survey was conducted, casinos opened in Montreal and in the Charlevoix region, located about 90 minutes drive from Quebec City. The introduction of casino gambling in Quebec is expected to modify the nature and frequency of gambling behaviors in the population.
Province of New Brunswick The study in New Brunswick was conducted during September and October 1992 at the request of the Lotteries Commission of New Brunswick, Department of Finance (Baseline Market Research, 1992). A private local firm completed a province-wide telephone survey to assess the level of involvement of New Brunswickers in gaming activities and the prevalence of problem and probable pathological gambling. The South Oaks Gambling Screen was used to assess problem and pathological gambling. A total of 801 individuals were sampled in a similar m a n n e r to the previous study (Ladouceur, 1991). The response rate was 59%. Estimates of problem gambling in New Brunswick were as follows: lifetime problem gambling was 4.0%, probable pathological gambling was 2.0%, adding to 6.0% of individuals in the sample reporting some degree of difficulty with control over gambling activities. As expected, the current one-year prevalence rates were somewhat lower: 3.13% for problem gambling and 1.37% for pathological gambling (total -- 4.5%). Respondents in New Brunswick were further classified into four categories on the basis of their gambling involvement: (1) non-gamblers: respondents who had never participated in any of the gaming activities investigated in the survey; (2) infrequent gamblers: those who had participated in one or more gaming activities at some point in time; (3) occasional gamblers: respondents who had bet money on at least one gaming
ROBERT L A D O U C E U R
133
activity in the past year but less than once per week; and (4) regulargamblers: respondents who bet money on at least one gaming activity at least once a week. Among regular gamblers, 12% were classified as "problem gamblers" and 3% were classified as "probable pathological gamblers." None of the infrequent gamblers reported problems with gambling activities. This information may be useful in identifying at-risk individuals who should be included in prevention and treatment programs. The typical problem gambler in New Brunswick is a single male, with no more than high school education, who began gambling after the age of 18 (which is unusual!) with an income under $30,000 per year.
Province of Nova Scotia The survey in Nova Scotia was conducted in March and April 1993 by a private firm on behalf of the Nova Scotia Drug Dependency Services Division (Omnifacts Research, 1993). A sample of 810 randomly selected individuals, 18 years of age and over, selected in a similar fashion as the Quebec and New Brunswick surveys, were interviewed using the South Oaks Gambling Screen. An additional randomly selected sample of 300 adolescents (13 to 17 years of age) was also included in the survey. The data on adolescents are reported in the following section. Pooling together the two samples, the response rate was only 34%, which is unfortunately very low. According to the authors of the report, the low response rate associated with this sample is due to the inclusion of adolescents in the sample. Before a youth could be interviewed, both the adolescent and a parent had to agree to the interview. About 50% of parents contacted refused to allow their child to participate in the survey which contributed disproportionately to the low response rate for the sample as a whole. A m o n g the adult sample from Nova Scotia, 80% reported having gambled at some time. Similar to other provinces, the preferred game among those who had participated in gambling activities was the lottery. The current (one year) prevalence rates of problem and probable pathological gambling were 3.0% and 1.7% respectively, summing up to 4.7% of the adult population identified as having current moderate to severe difficulties with gaming activities. As in the New Brunswick report, subcategories were used to provide other ways to describe the sample. However, some of these classifications were based on arbitrary criteria.
134
JOURNAL OF GAMBLINGSTUDIES
For example, the authors classified individuals who gambled once a week or less as infrequent gamblers. The typical profile of the problem and probable pathological gambler in Nova Scotia was a young to middle aged male, single if a problem gambler and married if a pathological gambler. Probable pathological gamblers were almost twice as likely as the remainder of the sample to have been divorced or separated. Slightly over 50% of the problem and probable pathological gamblers reported a total family income over $40,000 per year, with a high school education or less. For problem gamblers, slot and video poker machines were the most frequent gambling activities. The authors of the report correctly point out that this demographic profile must be interpreted with caution since none of the demographic characteristics of the problem and probable pathological gamblers were statistically different than the other sampled subjects.
Province of Alberta The prevalence study of problem and pathological gamblers in Alberta took place between mid-June and mid-July 1993 and the report was made available in January 1994 (Smith, Volberg & Wynne, 1994). Again, it was conducted by a private firm on behalf of the Alberta Lottery and Gaming Commission. A representative sample of 1,803 citizens of Alberta were interviewed in the study. The instrument used to assess problem and pathological gambling was the South Oaks Gambling Screen and the response rate was 50%. Alberta is notable for having the highest rate of individuals who stated that they had at some time wagered on one or more gambling activities (93%). The current prevalence rates (one year) for problem and pathological gambling were 4.0% and 1.4% respectively, adding up to 5.4% of individuals who reported having less control than desired over gambling activities. The demographic profile of the current problem and pathological gamblers was as follows: not married, under the age of 30, with an annual household income under $25,000 and with less than high school education. As in the three other provinces, lotteries were the preferred type of gambling for the total sample. The Alberta study represents a major improvement in the investigation of prevalence rates of problem and pathological gambling. As suggested by Dickerson (1993), and implemented by Abbott and Vol-
ROBERT LADOUCEUR
135
berg (1992) in a study conducted in New Zealand, a second phase was used to further evaluate pathological gamblers. The Alberta study is the first effort in North America to use a systematic qualitative data-gathering m o d e to investigate "at large" problem and pathological gamblers. The purpose o f the second phase of the Alberta study was, first, to improve the internal validity of the telephone survey and, second, to estimate the social and financial costs created by problem and pathological gamblers. The respondents in the second phase of the Alberta study were interviewed in person, three months after the telephone survey. Respondents were randomly drawn from three groups in the sample. The three groups included: (1) frequent gamblers (N = 10) who reported gambling at least 3 hours per week and who scored 0 on the South Oaks Gambling Screen; (2) problem gamblers (N = 10) who scored 3 or 4 on the South Oaks Gambling Screen; and (3) pathological gamblers (N = 10) who scored 5 or higher on the South Oaks Gambling Screen. Although this procedure is a valuable enhancement of standard prevalence studies, the low response rates in the Phase Two of the Alberta study seriously limit the scope of the conclusions. In order to obtain an N of 10 per group, 26 pathological, 35 problem and 20 frequent gamblers were contacted and asked to participate in the second phase of the study. The response rates in each group were 39%, 29%, and 50% respectively for a total response rate of 39%. This low response rate is not related to this particular approach: Abbott and Volberg (1992) had an average response rate of 78% using the same procedure.
Province of Ontario The study in Ontario was conducted over a one-week period between July 23 and August 1, 1993 on behalf of the Canadian Foundation on Compulsive Gambling (Ontario). For this study, a private firm surveyed 1,200 Ontario residents between the ages o f 18 and 74, using a stratified, random-digit dialling sampling method (Insight Canada Research, 1993). The response rate was 65%. Before discussing the results of this study, two major problems must be pointed out. First, the questionnaire used was a modified version o f the South Oaks Gambling Screen for which no description is provided. The reader cannot evaluate the differences between the original South Oaks Gambling Screen and the version used in Ontario.
136
JOURNAL OF GAMBLINGSTUDIES
Second, and more importantly, respondents scoring 1 to 4 out of a possible 20 points are defined as having "some gambling problem" and those scoring 5 or more points are categorized as "probable pathological gamblers." Although this classification is nearly identical to the scoring used for the South Oaks Gambling Screen, it has never been validated. All other studies used a score of 3 or 4 on the South Oaks Gambling Screen to classify problem gamblers and 5 or more for reporting a pathological gambler. Two-thirds of the Ontario respondents (67%) had bet money on gaming activities in their lifetimes. According to the Ontario version of the South Oaks Gambling Screen, 7.7% of the respondents were classified as problem gamblers and 0.9% were probable pathological gamblers, for a total of 8.6%. It is not clear if these figures are lifetime or current prevalence rates. The demographic profile of these gamblers was as follows: male, not married and under 35. Considering the methodological flaws of this survey, the results must be interpreted with caution.
Province of Saskatchewan The prevalence study of problem and pathological gamblers in Saskatchewan was conducted late in 1993 although the data were not analyzed until the beginning of 1994 (Volberg, 1994). A representative sample of 1,000 adult citizens of Saskatchewan were interviewed in a telephone survey using the South Oaks Gambling Screen. The size of this sample is quite large if one considers that the total population aged 18 and over in this province is less than 3/4 of a million individuals. Unfortunately, the response rate for the sample was not reported. Among the Saskatchewan respondents, 87% reported having gambled in the past year. As in other provinces, the most popular type of gambling was the lottery. Lifetime problem and probable pathological gambling were 2.8% and 1.2% respectively, representing 4.0% of the sample reporting moderate to severe difficulties related to their gambling. These figures were similar to those reported by Ladouceur (1991) in the Province of Quebec concerning the lifetime prevalence rates. Compared to the total sample, these individuals are significantly more likely to be male, under the age of 30, non-Caucasian and unmarried. No significant differences were found in terms of education or income.
ROBERT LADOUCEUR
137
The current (one year) prevalence rates in Saskatchewan were 1.9% for problem gambling and 0.8% for probable pathological gambling, for a total of 2.7% of the adult sample from Saskatchewan. Interestingly, comparisons between current and lifetime problem and pathological gamblers in Saskatchewan reveal that the former are more likely to be under the age of 30, non-Caucasian, unmarried, and to have an annual household income under $30,000. In terms of identifying gambling activities associated with gambling problems, 60% of the lifetime problem and probable pathological gamblers had participated in 5 or more types of gambling in the past year while only 13% of the non-gamblers had done so. Also, in terms of frequency, 72% of lifetime problem or probable pathological gamblers participated weekly in one or more types of gambling compared to 38% of respondents who had gambled in the past year without problems.
PATHOLOGICAL GAMBLING AMONG ADOLESCENTS As this paper was written, only three surveys identifying problem and pathological gambling rates among adolescents were available in Canada: two were conducted in Quebec and one in Nova Scotia. An adolescent survey was underway in Alberta at the time this paper was drafted. When we consider that experts in the field of problem gambling have confirmed that the majority of pathological gamblers begin gambling early in adolescence (see Shaffer and Hall, this issue), it is surprising that so few provincial agencies responsible for administering gaming activities in Canada have as yet addressed the issue of adolescent gambling behavior. The first Canadian survey of adolescents was conducted by Ladouceur and Mireault in 1986 and published in 1988. These researchers surveyed 1,752 high school students at 9 randomly selected high schools in the Quebec City area. In each school, 2 to 3 classes from each grade were randomly selected. In total, 70 classes of 22 to 26 students were contacted. The final sample was composed of 1,612 respondents (the response rate was 92%). The instrument used was the French version of the questionnaire developed by Lesieur and Klein (1987). Results showed that according to the DSM-III criteria for pathological gambling, 3.6% of the adolescent could be classified as patho-
138
JOURNAL OF GAMBLINGSTUDIES
logical gamblers. The descriptive behaviors related to gambling among adolescents were most revealing. Surprisingly, 7.6% o f the respondents stated that one or both parents gambled too much, 5.4% of them admitted cutting classes at least once in ordgr to gamble, and 5% used illegal means to finance their gambling. A m o n g other signs of pathological gambling, 12% of the adolescents admitted chasing: returning to gamble in o r d e r to win back money lost in previous gambling sessions. A second study was conducted in 1993 by the same research team in Q u e b e c (Ladouceur, Dub6 & Bujold, 1994a). The total sample was 1,471 college (17-19 year old) students in Quebec. The instrument used was the French version of the South Oaks Gambling Screen. In addition, questions taken from theJacobs Health Survey (Jacobs, 1987) were included. The same procedure as described for the high school students in Quebec was used. Results showed that 2.8% of these college students were pathological gamblers (those who obtained a score o f 5 or m o r e on the SOGS) and a further 5.8% of the students could b e considered problem gamblers (with a score of 3 or 4 on the South Oaks Gambling Screen). There was a large difference in the prevalence rate for male and female respondents: only 0.6% of the female respondents were classified as pathological gamblers whereas 5.7% of the male respondents met these criteria. For problem gambling, the prevalence rates were 3% and 9.5% respectively. Scores obtained on the South Oaks Gambling Screen were positively correlated with missing school in order to gamble and the number of hours worked per week at a job unrelated to school activities. Problem and pathological gambling were positively associated with tobacco use and alcohol use as well as with frequency and number of illegal substance usage. Finally, the results o f this study indicate that 26.8% o f pathological gamblers, 8.2% o f problem gamblers and 7.2% of students without gambling problems had attempted suicide. A rather different adolescent study was conducted in the Province o f Nova Scotia as the same time as the adult study (Omnifacts Research, 1993). A total sample of 300 adolescent residents o f Nova Scotia completed the interview. This study is probably the first to use telephone methods to assess the prevalence o f problem and pathological gambling in adolescents in Canada. The procedure for the study involved stratifying the population o f Nova Scotia in clusters and then selecting a proportional r a n d o m sample of telephone numbers from the current tele-
ROBERT LADOUCEUR
139
phone directories for each cluster. If an adolescent was identified in a household and agreed to participate in the survey, the interviewer asked to speak with a parent. If the parent agreed, the interview was completed but if either the adolescent or the parent refused, the interview was terminated. The mean age o f the adolescent sample was 16 years, equally divided between males and females. Prevalence rates among adolescents in Nova Scotia were 8.7% and 3.0% for problem and pathological gambling respectively. Although many more males (12.5%) than females (4.7%) were considered problem gamblers, about an equal n u m b e r of males (3.3%) and females (2.7%) were identified as probable pathological gamblers (Omnifacts Research, 1993). As noted above in connection with the adult survey, there were serious methodological flaws with the adolescent survey in Nova Scotia. In particular, the response rate for the total population studied (adults and adolescents) was only 34%. The adolescent prevalence rates of problem and pathological gambling are therefore based on a response rate o f approximately 15%, which seriously limits the results of this study. Until a replication study is conducted in this province to adequately evaluate adolescent gambling behaviors, the results of this survey should be treated with extreme caution.
GAMBLING AMONG PRIMARY S C H O O L CHILDREN The large proportion of adults and adolescents who gamble regularly in Canada suggests that these activities are first engaged in earlier in an individual's life. The only studies that address the gambling behaviors of children use a retrospective design in which adolescents and adults are questioned about their childhood experiences concerning gambling to determine at what age their gambling first started. There is an urgent need in the field of gambling research to systematically investigate the gambling behaviors of primary school children. The first North American study to evaluate gambling behaviors among primary school children was conducted in 1993 in the Province of Quebec by Ladouceur and his colleagues (Ladouceur, Dub~ & Bujold, 1994b). The sample was comprised of 1,320 school children of the 4th, 5th and 6th grades from the Quebec City region. Fourth graders composed 98.9% of the sample, fifth graders made up 37.3% of the sam-
140
JOURNAL OF GAMBLINGSTUDIES
ple and sixth graders made up 33.8% o f the sample. There was no significant age difference between boys and girls. Two research assistants administered the questionnaires in primary school classrooms. Gambling was defined as an activity that involves an element of risk or chance where money is either won or lost. The assistant indicated that video games are not considered gambling unless the parties involved have m o n e y bet on the outcome of the game. The youngsters were asked if they had ever gambled and if so, further questioned about their choice o f games, amounts and objects wagered, sources of inc o m e used to bet and gambling behaviors. The results showed that 86% of the respondents had wagered m o n e y and 37.2% admitted having bet an object that was dear to them. Lottery tickets constituted the most popular game, with 61.1% of the children stating that they had purchased tickets. Results indicated that 8.4% of the children played cards at least once a week, followed by 8.2% who bet on sports and 7.7% who played the lottery weekly. Gambling behaviors differed with age and grade: 80.9% of fourth graders had wagered m o n e y as compared to 84.3% of fifth graders, and 92.3% of sixth graders. As with adolescents, primary school boys gambled m o r e than their female counterparts: 89.7% of the boys surveyed said that they had gambled in the past compared to 81.9% o f the girls. Boys had bet significantly greater amounts of money in a single day than had girls. More boys than girls had gambled with their parents at the lottery, on bingo or on cards. These data clearly show that participation in gambling is not limited to adolescents and adults. As early as primary school, gambling is a well-anchored activity for a large proportion of children. In fact, a majority o f the grade school respondents had already gambled at the time o f the survey and a large n u m b e r did so on a weekly basis, even those aged nine and under. These results confirm information from retrospective studies about the early age at which adolescents and adults in Canada first gamble.
CONCLUDING REMARKS Canada has clearly adopted a liberal attitude toward the legalization o f gambling. Some form o f gambling exists in all provinces and
ROBERT LADOUCEUR
141
casino gambling will soon be available in more than half of the provinces. This attitude has put Canada in a similar position to, if not ahead of, the United States with regard to the prevalence of problem and pathological gambling. With the increasing availability and liberalization of gambling leading to increases of problem and pathological gambling, research on the psychology of gambling and on the treatment of pathological gamblers must be supported by different funding agencies, including the administrators of different gaming activities. It is hoped that the high prevalence rates for problem and pathological gamblers in Canada will lead provincial and federal authorities to recognize that a minimal monetary investment in research and treatment of excessive gamblers is a good bet.
REFERENCES Abbott, M.W. & Volberg, R.A. (1992). Frequent gamblers and problem gandders in New Zealand: Report on Phase Two of the national survey. Research Series No. 14. Wellington: New Zealand Department of Internal Affairs. Baseline Market Research. (1992). Prevalence survey of problem gambling in New Brunswick. Report to the New Brunswick Department of Finance. Campbell, C. & Lowman, J. (1989). Gambling in Canada: Golden goose or Trojan horse? (Proceedings of the First National Symposium on Lotteries and Gambling.) Vancouver: Simon Fraser University. Dickerson, M.G. (1993). A preliminary exploration of a two-stage methodology in the assessment of the extent and degree of gambling-related problems in the Australian population. In W.R. Eadington & J.A. Cornelius (Eds.) Gambling behavior and problem gambling. (pp. 347-364). Reno: University of Nevada Press. Insight Canada Research. (1993). Prevalence of problem and pathological gambling in Ontario using the South Oaks Gambling Screen. Toronto: Canadian Foundation for Compulsive Gambling. Jacobs, D.F. (1987). Eff~ts on children of parental r.xc.e~e~in gambling. Paper presented at the Seventh International Conference on Gambling and Risk Taking, Reno, NV. Ladouceur, R. (1991). Prevalence estimates of pathological gamblers in Quebec, Canada. Canadian Journal of P~chiatry, 36, 732-734. Ladouceur, R., DubS, D., & Bujold, A. (1994a). Prevalence of pathological gambling and related problems among college students in the Quebec metropolitan area. Canadian Journal of Psychiatry, 39, 289-392. Ladouceur, R., DubS, D., & Bujold, A. (1994b). Gambling among primary school students.Journal of Gambling Studi~ 10, 363-370. Ladouceur, R., & Mireault, C. (1988). Gambling behaviors among high school students in the Quebec area.Jou~l of Gambling Behavior, 4, 3-12. Lesieur, H.R. & Blume, S.B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. AmericanJournal of Peychiatry, 144, 1184-1188. Lesieur, H.R. & Klein, R. (1987). Pathological gambling among high school students. Addictive Behavior~ 1~ 129-135. Omnifacts Research. (1993). An examination of the prevalence of gambling in Nova Scotia. Report to the Nova Scotia Department of Health Drug Dependency Services.
142
J O U R N A L O F GAMBLING STUDIES
ShatTer, HJ. & Hall, M.N. (1996). Estimating the prevalence of adolescent gambling disorders: A quantitative synthesis and guide toward standard gambling nomenclature.Journal of Gambling Studies, 12, 193-214. Smith, GJ., Volberg, R.A. & Wynne, HJ. (1994). Gamblingand problem gambling in AlbeTta. Report to the Alberta Lotteries and Gaming Commission. Volberg, R.A. (1994). Gamblingand problemgambling in S~lskatche.wan.Report to the Minister's Advisory Committee on Social Impacts of Gaming.
Laval University, Canada
This paper critically reviews prevalence estimates of problem and pathological gambling in Canada. Populations studied are adults, adolescents and primary school children. T h e proportions of pathological gamblers found in Canadian studies (ranging from 1.2% to 1.9% for adults) are similar to prevalence rates reported in the United States. Given the apparent link between gambling availability and increases in the prevalence of problem and pathological gambling, it is h o p e d that provincial and federal authorities in Canada will make investments in research and treatment of pathological gambling in the future.
As in the United States, gambling in Canada has expanded rapidly in the past two decades. Before 1970, legal gambling in Canada was restricted to occasional charity bingos and raffles, midway games of chance, parimutuel wagering on horse races and friendly bets between individuals (Campbell & Lowman, 1989). By 1993, legal gambling in Canada had expanded to include slot machines and video gaming devices, casinos, large-scale bingo operations, sports wagering and offtrack betting on horse races. Lotteries, bingo and parimutuel wagering are available in every Canadian province; casinos now operate in more than half of the provinces. Clearly, Canada is following the trend in The author wishes to thank Mark Freeston for his helpful comments. Portions of this paper were written while the author had grants from Counseil de Recherche en Sciences Hnmaines du Canada (410-91-1514) and from Loto-Quebec. Send correspondence to Robert Ladouceur, Ph.D., #-cole de Psychologie, Universit~ Laval, Quebec, Quebec, Canada, GIK 7P4.
Journal of Gambling Studies "CoL12(2~ Summer 1996 9 1996 Human Sciences Press, Inc.
129
130
JOURNAL O F GAMBLINGSTUDIES
Western industrialized countries toward the approval and legalization of gambling. Studies estimating the prevalence of pathological gambling in Canada are now available for six provinces: Alberta, New Brunswick, Nova Scotia, Ontario, Quebec and Saskatchewan. While prevalence surveys have been completed in British Columbia as well as in Manitoba, the reports on these surveys were not available at the time this paper was written. Prevalence estimates in both of these provinces are similar to prevalence estimates in other Canadian provinces. This paper will address the following topics: (1) prevalence estimates of adult pathological gambling in Canada; (2) prevalence estimates of adolescent problem and pathological gambling in Canada; and (3) a description of gambling behaviors among primary school children in Quebec. The results of these studies have been and will continue to be useful in evaluating the need for treatment and prevention services for pathological gamblers as well as the need for public education and information campaigns concerning the danger of becoming an excessive gambler. Quebec is the only province where prevalence studies of pathological gamblers were conducted without the specific request of a gove r n m e n t agency. In all other provinces, private companies or individuals were asked by provincial authorities to provide estimates of problem and pathological gambling in order to establish policy toward the legalization or expansion of various forms of gambling.
PATHOLOGICAL GAMBLING AMONG ADULTS Prevalence estimates of problem and pathological gambling among adults are available for six Canadian provinces. Detailed results of these studies are presented in Table 1 in the chronological order in which they were conducted.
Province of Quebec The first Canadian prevalence study was conducted in 1989 (Ladouceur, 1991) and was based on the South Oaks Gambling Screen developed by Lesieur and Blume (1987), adapted and extensively used by Volberg and her colleagues in their seminal work on the prevalence
131
ROBERT LADOUCEUR
Table I Prevalence Rates of Problem and Probable Pathological Gambling in Canada
Ever Response Gambled Rate Problem Pathological Total QuebeO New Brunswick 2 Nova Scotia 2 Alberta 2 Saskatchewan 1 Saskatchewan 2 Ontario 3
88% 87% 80% 93% 87% 87% 67%
68% 59% 34% 50% ? ? 65%
2.6 3.13 3.0 4.0 2.8 1.9 7.7
1.2 1.37 1.7 1.4 1.2 0.8 0.9
3.8 4.5 4.7 5.4 4.0 2.7 8.6
1Lifetime rate ~Current rate (one-year) Slt is not clear if these r a t e s a r e current or lifedme rates. In addition, the cut-off points for classifying respondents are different from those used in other Canadian studies.
of problem and pathological gambling in the United States. For the Quebec survey, the South Oaks Gambling Screen was translated for French-speaking Canadians (about 80% of the population in the province of Quebec speak French). According to the criteria for the South Oaks Gambling Screen, respondents scoring 3 and 4 points were classified as "problem gamblers" and those scoring 5 or more points were classified as "probable pathological gamblers." The Quebec sample was stratified to proportionally represent the population (aged 18 and over) from each area of the province. Random selection of telephone numbers from telephone books and r a n d o m selection of respondents within the households were used. One person from each household was interviewed. A total of 1,002 interviews were completed. Up to five attempts were made to contact each number. The response rate was 68%. The overall results indicated that 2.6% of the Quebec sample could be classified as "problem gamblers" and 1.2% could be classified as "probable pathological gamblers." If we sum the two figures, 3.8% of the adult population in Quebec have at some time had a moderate to severe problem with gambling. The typical problem gambler in Quebec is a single male, under the age of 30, with an income between $15,000 and
132
JOURNAL OF GAMBLINGSTUDIES
$25,000. The interval between the age at which problem and probable pathological gamblers began wagering and the age at which they became preoccupied by the amount they were wagering was 8.8 years. In contrast to many United States jurisdictions, a substantial proportion of problem and probable pathological gamblers had consulted or had already sought help for their gambling problem (8.9% and 2.9%, for a total of 11.8%). Since lotteries are the most popular form of legalized gambling in Quebec, it is not surprising that 88% of the respondents had purchased a lottery ticket at sometime in their life. Since the time this survey was conducted, casinos opened in Montreal and in the Charlevoix region, located about 90 minutes drive from Quebec City. The introduction of casino gambling in Quebec is expected to modify the nature and frequency of gambling behaviors in the population.
Province of New Brunswick The study in New Brunswick was conducted during September and October 1992 at the request of the Lotteries Commission of New Brunswick, Department of Finance (Baseline Market Research, 1992). A private local firm completed a province-wide telephone survey to assess the level of involvement of New Brunswickers in gaming activities and the prevalence of problem and probable pathological gambling. The South Oaks Gambling Screen was used to assess problem and pathological gambling. A total of 801 individuals were sampled in a similar m a n n e r to the previous study (Ladouceur, 1991). The response rate was 59%. Estimates of problem gambling in New Brunswick were as follows: lifetime problem gambling was 4.0%, probable pathological gambling was 2.0%, adding to 6.0% of individuals in the sample reporting some degree of difficulty with control over gambling activities. As expected, the current one-year prevalence rates were somewhat lower: 3.13% for problem gambling and 1.37% for pathological gambling (total -- 4.5%). Respondents in New Brunswick were further classified into four categories on the basis of their gambling involvement: (1) non-gamblers: respondents who had never participated in any of the gaming activities investigated in the survey; (2) infrequent gamblers: those who had participated in one or more gaming activities at some point in time; (3) occasional gamblers: respondents who had bet money on at least one gaming
ROBERT L A D O U C E U R
133
activity in the past year but less than once per week; and (4) regulargamblers: respondents who bet money on at least one gaming activity at least once a week. Among regular gamblers, 12% were classified as "problem gamblers" and 3% were classified as "probable pathological gamblers." None of the infrequent gamblers reported problems with gambling activities. This information may be useful in identifying at-risk individuals who should be included in prevention and treatment programs. The typical problem gambler in New Brunswick is a single male, with no more than high school education, who began gambling after the age of 18 (which is unusual!) with an income under $30,000 per year.
Province of Nova Scotia The survey in Nova Scotia was conducted in March and April 1993 by a private firm on behalf of the Nova Scotia Drug Dependency Services Division (Omnifacts Research, 1993). A sample of 810 randomly selected individuals, 18 years of age and over, selected in a similar fashion as the Quebec and New Brunswick surveys, were interviewed using the South Oaks Gambling Screen. An additional randomly selected sample of 300 adolescents (13 to 17 years of age) was also included in the survey. The data on adolescents are reported in the following section. Pooling together the two samples, the response rate was only 34%, which is unfortunately very low. According to the authors of the report, the low response rate associated with this sample is due to the inclusion of adolescents in the sample. Before a youth could be interviewed, both the adolescent and a parent had to agree to the interview. About 50% of parents contacted refused to allow their child to participate in the survey which contributed disproportionately to the low response rate for the sample as a whole. A m o n g the adult sample from Nova Scotia, 80% reported having gambled at some time. Similar to other provinces, the preferred game among those who had participated in gambling activities was the lottery. The current (one year) prevalence rates of problem and probable pathological gambling were 3.0% and 1.7% respectively, summing up to 4.7% of the adult population identified as having current moderate to severe difficulties with gaming activities. As in the New Brunswick report, subcategories were used to provide other ways to describe the sample. However, some of these classifications were based on arbitrary criteria.
134
JOURNAL OF GAMBLINGSTUDIES
For example, the authors classified individuals who gambled once a week or less as infrequent gamblers. The typical profile of the problem and probable pathological gambler in Nova Scotia was a young to middle aged male, single if a problem gambler and married if a pathological gambler. Probable pathological gamblers were almost twice as likely as the remainder of the sample to have been divorced or separated. Slightly over 50% of the problem and probable pathological gamblers reported a total family income over $40,000 per year, with a high school education or less. For problem gamblers, slot and video poker machines were the most frequent gambling activities. The authors of the report correctly point out that this demographic profile must be interpreted with caution since none of the demographic characteristics of the problem and probable pathological gamblers were statistically different than the other sampled subjects.
Province of Alberta The prevalence study of problem and pathological gamblers in Alberta took place between mid-June and mid-July 1993 and the report was made available in January 1994 (Smith, Volberg & Wynne, 1994). Again, it was conducted by a private firm on behalf of the Alberta Lottery and Gaming Commission. A representative sample of 1,803 citizens of Alberta were interviewed in the study. The instrument used to assess problem and pathological gambling was the South Oaks Gambling Screen and the response rate was 50%. Alberta is notable for having the highest rate of individuals who stated that they had at some time wagered on one or more gambling activities (93%). The current prevalence rates (one year) for problem and pathological gambling were 4.0% and 1.4% respectively, adding up to 5.4% of individuals who reported having less control than desired over gambling activities. The demographic profile of the current problem and pathological gamblers was as follows: not married, under the age of 30, with an annual household income under $25,000 and with less than high school education. As in the three other provinces, lotteries were the preferred type of gambling for the total sample. The Alberta study represents a major improvement in the investigation of prevalence rates of problem and pathological gambling. As suggested by Dickerson (1993), and implemented by Abbott and Vol-
ROBERT LADOUCEUR
135
berg (1992) in a study conducted in New Zealand, a second phase was used to further evaluate pathological gamblers. The Alberta study is the first effort in North America to use a systematic qualitative data-gathering m o d e to investigate "at large" problem and pathological gamblers. The purpose o f the second phase of the Alberta study was, first, to improve the internal validity of the telephone survey and, second, to estimate the social and financial costs created by problem and pathological gamblers. The respondents in the second phase of the Alberta study were interviewed in person, three months after the telephone survey. Respondents were randomly drawn from three groups in the sample. The three groups included: (1) frequent gamblers (N = 10) who reported gambling at least 3 hours per week and who scored 0 on the South Oaks Gambling Screen; (2) problem gamblers (N = 10) who scored 3 or 4 on the South Oaks Gambling Screen; and (3) pathological gamblers (N = 10) who scored 5 or higher on the South Oaks Gambling Screen. Although this procedure is a valuable enhancement of standard prevalence studies, the low response rates in the Phase Two of the Alberta study seriously limit the scope of the conclusions. In order to obtain an N of 10 per group, 26 pathological, 35 problem and 20 frequent gamblers were contacted and asked to participate in the second phase of the study. The response rates in each group were 39%, 29%, and 50% respectively for a total response rate of 39%. This low response rate is not related to this particular approach: Abbott and Volberg (1992) had an average response rate of 78% using the same procedure.
Province of Ontario The study in Ontario was conducted over a one-week period between July 23 and August 1, 1993 on behalf of the Canadian Foundation on Compulsive Gambling (Ontario). For this study, a private firm surveyed 1,200 Ontario residents between the ages o f 18 and 74, using a stratified, random-digit dialling sampling method (Insight Canada Research, 1993). The response rate was 65%. Before discussing the results of this study, two major problems must be pointed out. First, the questionnaire used was a modified version o f the South Oaks Gambling Screen for which no description is provided. The reader cannot evaluate the differences between the original South Oaks Gambling Screen and the version used in Ontario.
136
JOURNAL OF GAMBLINGSTUDIES
Second, and more importantly, respondents scoring 1 to 4 out of a possible 20 points are defined as having "some gambling problem" and those scoring 5 or more points are categorized as "probable pathological gamblers." Although this classification is nearly identical to the scoring used for the South Oaks Gambling Screen, it has never been validated. All other studies used a score of 3 or 4 on the South Oaks Gambling Screen to classify problem gamblers and 5 or more for reporting a pathological gambler. Two-thirds of the Ontario respondents (67%) had bet money on gaming activities in their lifetimes. According to the Ontario version of the South Oaks Gambling Screen, 7.7% of the respondents were classified as problem gamblers and 0.9% were probable pathological gamblers, for a total of 8.6%. It is not clear if these figures are lifetime or current prevalence rates. The demographic profile of these gamblers was as follows: male, not married and under 35. Considering the methodological flaws of this survey, the results must be interpreted with caution.
Province of Saskatchewan The prevalence study of problem and pathological gamblers in Saskatchewan was conducted late in 1993 although the data were not analyzed until the beginning of 1994 (Volberg, 1994). A representative sample of 1,000 adult citizens of Saskatchewan were interviewed in a telephone survey using the South Oaks Gambling Screen. The size of this sample is quite large if one considers that the total population aged 18 and over in this province is less than 3/4 of a million individuals. Unfortunately, the response rate for the sample was not reported. Among the Saskatchewan respondents, 87% reported having gambled in the past year. As in other provinces, the most popular type of gambling was the lottery. Lifetime problem and probable pathological gambling were 2.8% and 1.2% respectively, representing 4.0% of the sample reporting moderate to severe difficulties related to their gambling. These figures were similar to those reported by Ladouceur (1991) in the Province of Quebec concerning the lifetime prevalence rates. Compared to the total sample, these individuals are significantly more likely to be male, under the age of 30, non-Caucasian and unmarried. No significant differences were found in terms of education or income.
ROBERT LADOUCEUR
137
The current (one year) prevalence rates in Saskatchewan were 1.9% for problem gambling and 0.8% for probable pathological gambling, for a total of 2.7% of the adult sample from Saskatchewan. Interestingly, comparisons between current and lifetime problem and pathological gamblers in Saskatchewan reveal that the former are more likely to be under the age of 30, non-Caucasian, unmarried, and to have an annual household income under $30,000. In terms of identifying gambling activities associated with gambling problems, 60% of the lifetime problem and probable pathological gamblers had participated in 5 or more types of gambling in the past year while only 13% of the non-gamblers had done so. Also, in terms of frequency, 72% of lifetime problem or probable pathological gamblers participated weekly in one or more types of gambling compared to 38% of respondents who had gambled in the past year without problems.
PATHOLOGICAL GAMBLING AMONG ADOLESCENTS As this paper was written, only three surveys identifying problem and pathological gambling rates among adolescents were available in Canada: two were conducted in Quebec and one in Nova Scotia. An adolescent survey was underway in Alberta at the time this paper was drafted. When we consider that experts in the field of problem gambling have confirmed that the majority of pathological gamblers begin gambling early in adolescence (see Shaffer and Hall, this issue), it is surprising that so few provincial agencies responsible for administering gaming activities in Canada have as yet addressed the issue of adolescent gambling behavior. The first Canadian survey of adolescents was conducted by Ladouceur and Mireault in 1986 and published in 1988. These researchers surveyed 1,752 high school students at 9 randomly selected high schools in the Quebec City area. In each school, 2 to 3 classes from each grade were randomly selected. In total, 70 classes of 22 to 26 students were contacted. The final sample was composed of 1,612 respondents (the response rate was 92%). The instrument used was the French version of the questionnaire developed by Lesieur and Klein (1987). Results showed that according to the DSM-III criteria for pathological gambling, 3.6% of the adolescent could be classified as patho-
138
JOURNAL OF GAMBLINGSTUDIES
logical gamblers. The descriptive behaviors related to gambling among adolescents were most revealing. Surprisingly, 7.6% o f the respondents stated that one or both parents gambled too much, 5.4% of them admitted cutting classes at least once in ordgr to gamble, and 5% used illegal means to finance their gambling. A m o n g other signs of pathological gambling, 12% of the adolescents admitted chasing: returning to gamble in o r d e r to win back money lost in previous gambling sessions. A second study was conducted in 1993 by the same research team in Q u e b e c (Ladouceur, Dub6 & Bujold, 1994a). The total sample was 1,471 college (17-19 year old) students in Quebec. The instrument used was the French version of the South Oaks Gambling Screen. In addition, questions taken from theJacobs Health Survey (Jacobs, 1987) were included. The same procedure as described for the high school students in Quebec was used. Results showed that 2.8% of these college students were pathological gamblers (those who obtained a score o f 5 or m o r e on the SOGS) and a further 5.8% of the students could b e considered problem gamblers (with a score of 3 or 4 on the South Oaks Gambling Screen). There was a large difference in the prevalence rate for male and female respondents: only 0.6% of the female respondents were classified as pathological gamblers whereas 5.7% of the male respondents met these criteria. For problem gambling, the prevalence rates were 3% and 9.5% respectively. Scores obtained on the South Oaks Gambling Screen were positively correlated with missing school in order to gamble and the number of hours worked per week at a job unrelated to school activities. Problem and pathological gambling were positively associated with tobacco use and alcohol use as well as with frequency and number of illegal substance usage. Finally, the results o f this study indicate that 26.8% o f pathological gamblers, 8.2% o f problem gamblers and 7.2% of students without gambling problems had attempted suicide. A rather different adolescent study was conducted in the Province o f Nova Scotia as the same time as the adult study (Omnifacts Research, 1993). A total sample of 300 adolescent residents o f Nova Scotia completed the interview. This study is probably the first to use telephone methods to assess the prevalence o f problem and pathological gambling in adolescents in Canada. The procedure for the study involved stratifying the population o f Nova Scotia in clusters and then selecting a proportional r a n d o m sample of telephone numbers from the current tele-
ROBERT LADOUCEUR
139
phone directories for each cluster. If an adolescent was identified in a household and agreed to participate in the survey, the interviewer asked to speak with a parent. If the parent agreed, the interview was completed but if either the adolescent or the parent refused, the interview was terminated. The mean age o f the adolescent sample was 16 years, equally divided between males and females. Prevalence rates among adolescents in Nova Scotia were 8.7% and 3.0% for problem and pathological gambling respectively. Although many more males (12.5%) than females (4.7%) were considered problem gamblers, about an equal n u m b e r of males (3.3%) and females (2.7%) were identified as probable pathological gamblers (Omnifacts Research, 1993). As noted above in connection with the adult survey, there were serious methodological flaws with the adolescent survey in Nova Scotia. In particular, the response rate for the total population studied (adults and adolescents) was only 34%. The adolescent prevalence rates of problem and pathological gambling are therefore based on a response rate o f approximately 15%, which seriously limits the results of this study. Until a replication study is conducted in this province to adequately evaluate adolescent gambling behaviors, the results of this survey should be treated with extreme caution.
GAMBLING AMONG PRIMARY S C H O O L CHILDREN The large proportion of adults and adolescents who gamble regularly in Canada suggests that these activities are first engaged in earlier in an individual's life. The only studies that address the gambling behaviors of children use a retrospective design in which adolescents and adults are questioned about their childhood experiences concerning gambling to determine at what age their gambling first started. There is an urgent need in the field of gambling research to systematically investigate the gambling behaviors of primary school children. The first North American study to evaluate gambling behaviors among primary school children was conducted in 1993 in the Province of Quebec by Ladouceur and his colleagues (Ladouceur, Dub~ & Bujold, 1994b). The sample was comprised of 1,320 school children of the 4th, 5th and 6th grades from the Quebec City region. Fourth graders composed 98.9% of the sample, fifth graders made up 37.3% of the sam-
140
JOURNAL OF GAMBLINGSTUDIES
ple and sixth graders made up 33.8% o f the sample. There was no significant age difference between boys and girls. Two research assistants administered the questionnaires in primary school classrooms. Gambling was defined as an activity that involves an element of risk or chance where money is either won or lost. The assistant indicated that video games are not considered gambling unless the parties involved have m o n e y bet on the outcome of the game. The youngsters were asked if they had ever gambled and if so, further questioned about their choice o f games, amounts and objects wagered, sources of inc o m e used to bet and gambling behaviors. The results showed that 86% of the respondents had wagered m o n e y and 37.2% admitted having bet an object that was dear to them. Lottery tickets constituted the most popular game, with 61.1% of the children stating that they had purchased tickets. Results indicated that 8.4% of the children played cards at least once a week, followed by 8.2% who bet on sports and 7.7% who played the lottery weekly. Gambling behaviors differed with age and grade: 80.9% of fourth graders had wagered m o n e y as compared to 84.3% of fifth graders, and 92.3% of sixth graders. As with adolescents, primary school boys gambled m o r e than their female counterparts: 89.7% of the boys surveyed said that they had gambled in the past compared to 81.9% o f the girls. Boys had bet significantly greater amounts of money in a single day than had girls. More boys than girls had gambled with their parents at the lottery, on bingo or on cards. These data clearly show that participation in gambling is not limited to adolescents and adults. As early as primary school, gambling is a well-anchored activity for a large proportion of children. In fact, a majority o f the grade school respondents had already gambled at the time o f the survey and a large n u m b e r did so on a weekly basis, even those aged nine and under. These results confirm information from retrospective studies about the early age at which adolescents and adults in Canada first gamble.
CONCLUDING REMARKS Canada has clearly adopted a liberal attitude toward the legalization o f gambling. Some form o f gambling exists in all provinces and
ROBERT LADOUCEUR
141
casino gambling will soon be available in more than half of the provinces. This attitude has put Canada in a similar position to, if not ahead of, the United States with regard to the prevalence of problem and pathological gambling. With the increasing availability and liberalization of gambling leading to increases of problem and pathological gambling, research on the psychology of gambling and on the treatment of pathological gamblers must be supported by different funding agencies, including the administrators of different gaming activities. It is hoped that the high prevalence rates for problem and pathological gamblers in Canada will lead provincial and federal authorities to recognize that a minimal monetary investment in research and treatment of excessive gamblers is a good bet.
REFERENCES Abbott, M.W. & Volberg, R.A. (1992). Frequent gamblers and problem gandders in New Zealand: Report on Phase Two of the national survey. Research Series No. 14. Wellington: New Zealand Department of Internal Affairs. Baseline Market Research. (1992). Prevalence survey of problem gambling in New Brunswick. Report to the New Brunswick Department of Finance. Campbell, C. & Lowman, J. (1989). Gambling in Canada: Golden goose or Trojan horse? (Proceedings of the First National Symposium on Lotteries and Gambling.) Vancouver: Simon Fraser University. Dickerson, M.G. (1993). A preliminary exploration of a two-stage methodology in the assessment of the extent and degree of gambling-related problems in the Australian population. In W.R. Eadington & J.A. Cornelius (Eds.) Gambling behavior and problem gambling. (pp. 347-364). Reno: University of Nevada Press. Insight Canada Research. (1993). Prevalence of problem and pathological gambling in Ontario using the South Oaks Gambling Screen. Toronto: Canadian Foundation for Compulsive Gambling. Jacobs, D.F. (1987). Eff~ts on children of parental r.xc.e~e~in gambling. Paper presented at the Seventh International Conference on Gambling and Risk Taking, Reno, NV. Ladouceur, R. (1991). Prevalence estimates of pathological gamblers in Quebec, Canada. Canadian Journal of P~chiatry, 36, 732-734. Ladouceur, R., DubS, D., & Bujold, A. (1994a). Prevalence of pathological gambling and related problems among college students in the Quebec metropolitan area. Canadian Journal of Psychiatry, 39, 289-392. Ladouceur, R., DubS, D., & Bujold, A. (1994b). Gambling among primary school students.Journal of Gambling Studi~ 10, 363-370. Ladouceur, R., & Mireault, C. (1988). Gambling behaviors among high school students in the Quebec area.Jou~l of Gambling Behavior, 4, 3-12. Lesieur, H.R. & Blume, S.B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. AmericanJournal of Peychiatry, 144, 1184-1188. Lesieur, H.R. & Klein, R. (1987). Pathological gambling among high school students. Addictive Behavior~ 1~ 129-135. Omnifacts Research. (1993). An examination of the prevalence of gambling in Nova Scotia. Report to the Nova Scotia Department of Health Drug Dependency Services.
142
J O U R N A L O F GAMBLING STUDIES
ShatTer, HJ. & Hall, M.N. (1996). Estimating the prevalence of adolescent gambling disorders: A quantitative synthesis and guide toward standard gambling nomenclature.Journal of Gambling Studies, 12, 193-214. Smith, GJ., Volberg, R.A. & Wynne, HJ. (1994). Gamblingand problem gambling in AlbeTta. Report to the Alberta Lotteries and Gaming Commission. Volberg, R.A. (1994). Gamblingand problemgambling in S~lskatche.wan.Report to the Minister's Advisory Committee on Social Impacts of Gaming.
