Journal Information
Journal ID (publisher-id): jgi
ISSN: 1910-7595
Publisher: Centre for Addiction and Mental Health
Article Information
The Centre for Addiction and Mental Health
Received Day: 4 Month: June Year: 2012
Accepted Day: 5 Month: September Year: 2012
Publication date: December 2013
First Page: 1 Last Page: 19
Publisher Id: jgi.2013.28.10
DOI: 10.4309/jgi.2013.28.10

Stereotypes of problem gambling
Jenny Horch
David Hodgins
University of Calgary, Calgary, Alberta, Canada
This article was peer-reviewed. All URLs were available at the time of submission.

For correspondence: Jenny Horch, PhD, University of Calgary, Department of Psychology, University of Calgary, 2500 University Dr. NW, Calgary, AB, T2N 1N4, Canada, 403-210-9500. E-mail:
Competing interests: None declared.
Ethics approval: This research was approved on June 10, 2008, by the Conjoint Faculties Research Ethics Board at the University of Calgary under the title “Private Stigma of Disordered Gamblers – Part A” (File No. 5686).
Contributors: JH was involved in research design, conducting the study, data analysis, writing, and editing the manuscript. DH was involved in research design, statistical analyses, and editing versions of the manuscript.
Jenny D. Horch, PhD, is a post doctoral research fellow in the Addictive Behaviours Laboratory at the University of Calgary. Her research has focused on stigma, self-esteem, and reduced treatment seeking in problem gambling and she has several publications in this area. Dr. Horch is a clinical psychologist and maintains a clinical position treating patients on the Burn and Trauma Units at the Foothills Medical Centre. She is also involved in a medical research team that aims to improve physical and psychological outcomes of skin grafts.
David C. Hodgins, PhD, is a professor in the Program in Clinical Psychology in the Department of Psychology, University of Calgary. Dr. Hodgins is also the Head of the Department of Psychology and a coordinator with the Alberta Gambling Research Institute. His research interests focus on various aspects of addictive behaviours including relapse and recovery from substance abuse and gambling disorders. He has conducted a number of randomized clinical trials of both brief and more traditional addiction treatment models. In 2010, he received the Scientific Achievement Award from the US National Center for Responsible Gaming. Dr. Hodgins teaches in the clinical psychology program and has an active cadre of graduate students. He maintains a private practice in Calgary and provides consultation to a number of international organizations.


Introduction: Research supports the notion that problem gambling is stigmatized, yet little is known about stereotypes, a key variable in the stigmatization process. Method: University students (41 male, 110 female) generated words when presented with one of three labels: gambler, problem gambler, and gambling addict. An adjective checklist permitted participants to select words characteristic of problem gamblers and was administered to additional student (N = 790) and problem gambling samples (N = 74). Results: Content and frequency analyses revealed that problem gamblers were considered compulsive, impulsive, desperate, irresponsible, risk-taking, depressed, greedy, irrational, antisocial, and aggressive. Problem gambling and gambling addict labels generated more words regarding negative gambling consequences. Gambler resulted in more miscellaneous words (e.g., casino, money). Conclusions: Stereotype content was not entirely inaccurate and the label gambler was not neutral. Future research could examine which aspect of stereotype content invites stigmatization.


Introduction: La recherche appuie l'hypothèse selon laquelle le jeu compulsif est stigmatisé (Horch et Hodgins, 2008; Hodgins et el-Guebaly, 2001). Toutefois, on en sait très peu sur les stéréotypes, qui sont l'une des variables clés dans le processus de stigmatisation. Méthodologie: Des étudiants (41 hommes, 110 femmes) devaient proposer des mots lorsqu'on leur présentait l'une des trois étiquettes suivantes : « joueur », « joueur compulsif » et « dépendant au jeu ». Une liste d'adjectifs permettait aux participants de choisir des mots caractéristiques décrivant un « joueur compulsif ». Cette liste a été soumise à des étudiants supplémentaires (N  =  790) et à un échantillon de joueurs compulsifs (N  =  74).Résultats: Des analyses de contenu et de la fréquence ont révélé que le « joueur compulsif » était considéré comme impulsif, capable de tout, irresponsable, preneur de risques, déprimé, avide d'argent, asocial, agressif et irrationnel. Les étiquettes « joueur compulsif » et « dépendant au jeu » ont produit davantage de mots pour ce qui est des conséquences négatives de la dépendance au jeu. Le mot « joueur » a produit davantage de mots divers (p.ex. casino, argent). Conclusions: Le contenu des stéréotypes n'était pas tout à fait inexact et l'étiquette « joueur » n'était pas neutre. De nouvelles études pourraient examiner quel aspect du contenu des stéréotypes entraîne la stigmatisation.


Research supports the notion that problem gambling is stigmatized. Both the general public (Rockloff & Schofield, 2004) and those with gambling problems (Hodgins & el-Guebaly, 2000; Tavares, Martins, Zilberman, & el-Guebaly, 2002) perceive stigma associated with problem gambling. Furthermore, individuals with gambling problems report stigma as a major motive for treatment avoidance (Hodgins & el-Guebaly, 2000) and delay (Tavares et al., 2002). A recent review of barriers indicated intent to handle the problem on one's own (i.e., self-change), shame/embarrassment/stigma, and denial were the most endorsed barriers to problem gambling treatment-seeking (Suurvali, Cordingley, Hodgins, & Cunningham, 2009). However stigma is a broad and often poorly defined construct. Stigma has been conceptualized as a process consisting of several components, including cues, stereotypes, prejudice, and discrimination that roughly correspond to perception, cognition, affect, and behaviour respectively (Corrigan, 2004; Teachman, Wilson, & Komarovskaya, 2006). While the problem gambling literature is beginning to consider the topic of stigma, the construct has not been particularly well conceptualized. Furthermore, little is known about problem gambling stereotypes, a key variable in the stigmatization process. Stereotype content regarding problem gamblers may be important both to better understand cultural conceptions of problem gamblers and to perhaps modify inaccurate or harmful conceptions in order to reduce stigmatization or to conduct other interventions aimed at increasing treatment-seeking in this population.

Stereotypes have been defined as “characteristics that are descriptive of, attributed to, or associated with members of social groups or categories” (Stangor & Lange, 1994, p. 361, italics in original). Stereotypes about social groups are commonly-used sets of expectations that allow us to predict how individuals will behave and therefore also determine our responses (Stangor & Lange, 1994). Stereotypes include judgements of others that influence what information is sought out, attended to and remembered about others, how that information is interpreted and the ease with which it is processed, as well as the affective and emotional responses we experience (Stangor & Lange, 1994). Stereotypes could be considered a useful cognitive heuristic or mental representation if it were not for often being unfairly deleterious.

Stereotype content regarding problem gamblers may be important both to better understand cultural conceptions of problem gamblers and to perhaps modify these conceptions to improve accuracy or in order to reduce stigmatization and increase treatment-seeking. Both the general public and individuals struggling with problem gambling have indicated that they perceive stigma associated with the condition (Hodgins & el-Guebaly, 2000; Horch & Hodgins, 2008; Rockloff & Schofield, 2004; Tavares et al., 2002) yet little is known regarding what aspect of problem gambling invites stigmatization. Without determining stereotype content about individuals with gambling problems it may prove challenging to modify any possibly inaccurate perceptions or to conduct other interventions aimed at increasing treatment-seeking in this population.

Although little empirical literature exists regarding stereotypes of problem gamblers, the broader stereotype literature is substantial. Stereotypes have been studied at both an individual (personal) and a cultural (consensual) level. These have been associated with process and content studies, respectively. Stereotype process studies examine individual level stereotyping and investigate what functions stereotypes serve for mental processing (i.e., they answer the questions “how?” or “why?” people stereotype). Stereotype content refers to the characteristics and descriptors typically assigned to a social category (Operario & Fiske, 2004). Stereotype content studies examine cultural level stereotyping and investigate “what” people believe about a particular group of individuals (Stangor & Lange, 1994). The present study examines stereotype content in problem gambling.

Several methodologies have been developed to measure stereotype content, including checklists, rating scales, and free response measures (Madon, 1997). Checklist methodologies were the first to be employed and were used in the pioneering Katz and Braly (1933) study on ethnic stereotype content. Checklists, or adjective checklists, contain a number of predetermined traits from which participants select adjectives pertaining to the social category of interest. Rating scale or differential methodologies are similar, except that participants rate the extent to which a trait characterizes a member of a social category rather than simply checking whether an attribute applies to such individuals. Finally, open or free response methods have participants generate attributes and list them on a blank sheet of paper, typically with only the category label at the top of the page.

An advantage of checklist measures is that they can sample a broad range of traits. However, as the attributes are researcher-generated, it is possible that important attributes may be omitted as an oversight. Rating scales have a similar disadvantage but are able to provide a measure of relative stereotype strength. Open-response measures, in contrast, are more likely to capture central components of a stereotype, and they do not constrain participants to consider only attributes provided by the researchers that might not automatically come to mind (Madon, 1997). However, possible incomplete responding may be a drawback of this method (e.g., participants may only generate the first attribute to come to mind). To remedy the weaknesses of particular stereotype measures, researchers have suggested using a combined methodology (Stangor & Lange, 1994), an approach adopted in the present study.

Stereotype content for problem gambling is difficult to predict. As problem gambling has been classified as a mental illness, stereotype content may coincide with beliefs about individuals with mental illness. Some commonly held stereotypes about such individuals include beliefs that they are unpredictable, incompetent, immoral, unlikely to recover, and to blame for their condition (Corrigan, 2004; Corrigan et al., 2000; Watson & River, 2005). Others have classified problem gambling as an addiction, which may suggest stereotypes of dangerousness and responsibility issues (Rasinski, Woll, & Cooke, 2005).


The primary objective of the present investigation was to determine stereotype content associated with problem gambling as held by university students. University students tend to be high frequency gamblers (Adlaf, Demers, & Gliksman, 2005) and have higher rates of problem gambling than the general population (National Research Council, 1999), making this a particularly interesting sample. Two separate samples of university students are reported and checklist results from individuals with a gambling problem were also obtained. A secondary objective was to examine the utility of free-response and checklist methodologies and compare content produced by each method. Finally, exploration of open response data was also made, including examining the impact of different labels (i.e., gambler, problem gambler, or gambling addict) on words generated, inquiring as to whether a prototype came to mind, and exploring additional non-adjective stereotype information.

It was expected that stereotype content would reveal constructs such as impulsivity and negative personality traits. Furthermore, it seemed likely that the term addiction would be generated. Regarding free response and checklist methodologies, it was expected that both were likely to generate overlapping but different stereotype content, with free response generating physical descriptors (see Niemann, Jennings, Rozelle, Baxter, & Sullivan, 1994) and other non-adjective words.


Three samples are presented with Sample 1 as the primary sample and only the adjective checklist of interest in Samples 2 and 3. Sample 1 consisted of 152 participants (41 men, 110 women, 1 undisclosed) averaging 21.9 years old (SD = 5.0, range 17–55) who were recruited from the University of Calgary's Research Participation System. Students were granted partial course credit for their participation in what was described as a study on their “knowledge of characteristics of different groups,” in order to avoid pre-selecting those with prior interest in gambling. Sample 2 consisted of 790 university students (142 men, 648 women, 2 undisclosed) averaging 20.5 years old (SD = 3.8, range 17–51), also recruited as indicated above. For Sample 3, 74 individuals with gambling problems (50 men, 24 women) averaging 41.5 years old (SD = 13.8, range 18–70) were recruited largely via newspaper classified advertisements and community posters. Advertisements invited individuals with “concerns about [their] gambling” to participate in a study on gambling stigma. Individuals who were 18 years of age, able to read English, and obtained a score of three or greater on the lifetime National Opinion Research Center DSM Screen for Gambling Problems (NODS; Gerstein et al., 1999; administered over the telephone) were invited to participate. Participants completed paper and pencil questionnaire measures and received a $20.00 grocery store gift card for participation.

Sample size for Sample 1 was based on previous qualitative stereotype research that considered ethnic, sexual orientation, criminal, and substance use stereotypes. Previous studies have used between 100 subjects (Marín, 1984; Williams & Best, 1977; Katz & Braley, 1933) and 200 subjects (i.e., 187, Devine & Baker, 1991; 197, Gardner, Kirby, Gorospe, & Villamin, 1972). A sample size of 150 was considered sufficient to gain an understanding of the general cultural stereotype of problem gamblers.


Stereotype content. The Adjective Checklist (ACL; Gough & Heilbrun, 1983) is a 300 item checklist that uses adjectives (e.g., from absent-minded to zany) which may be applied to oneself or any particular social group. The measure has been used in past stereotype research and requires 10 to 15 minutes to complete. For the purposes of this study the ACL was modified in order to develop a checklist more representative of the problem gambling stereotype and to incorporate a rating scale methodology.

Prior to administration of Sample 1, four PhD candidates from the Addictive Behaviours Laboratory at the University of Calgary were asked to generate several words to describe problem gamblers and some words were added to the ACL based on their responses (i.e., antisocial, compulsive, crazy, depressed, desperate, distressed, disturbed, dysfunctional, irrational, isolated, manipulative, risk-taking, secretive, sensation seeking, shameful, stupid, uncontrolled, unfulfilled, and untrustworthy). The ACL format was also modified in that participants were asked to review the traits checked and select the five words most and least typical of problem gamblers. Only one label was used in the ACL and the label problem gambler was considered mostly widely used. Each of the selected words was then rated on a 5-point Likert-type scale with 1 (very uncharacteristic of problem gamblers), 3 (no more characteristic of problem gamblers than of any other group), and 5 (very characteristic of problem gamblers).

Before the administration of Samples 2 and 3, the ACL was again modified. Words added included addictive, broke, competitive, dishonest, fun-loving, obsessive, wasteful, and weak-willed. Finally, the 5-point Likert-type scale was modified to a 7-point scale with the same anchors. Due to these modifications, results from Sample 1 and from Samples 2 and 3 are presented separately in the Results section.

Instructions for the adjective checklist were not always well understood and analyses were conducted on valid data only. Specifically, data from individuals who did not appear to understand the rating scale were not included in the analysis (i.e., some individuals described a word as uncharacteristic but then selected a rating of characteristic and vice versa).

Demographics and gambling involvement. Demographic information was collected to examine potential covariates. Participants were asked to indicate their sex, age, marital status, income, ethnicity, and political orientation. Political orientation categories were taken from Kemmelmeier, Danielson, and Basten (2005). Frequency of gambling behaviour (daily, weekly, monthly, occasionally, never) was solicited for the following gambling activities: instant/scratch tickets, slot machines, video lottery terminals (VLTs), casino table games, lottery, raffles/fundraising tickets, bingo, sport select, horse/dog racing, sport betting with a bookie, sports pools (workplace, friends, others), games of skill (darts, golf, pool), keno, gambling on cards with friends/family, internet gambling, speculative investments/stocks. Familiarity with individuals with a gambling problem was measured with the Level of Contact Report (LOC; Holmes, Corrigan, Williams, Canar, & Kubiak, 1999); a 12-item checklist that describes varying levels of intimacy with an individual with mental illness, ranging from the most intimate contact (“I have a mental illness”) to the least intimate contact (“I have never observed a person that I was aware had a serious mental illness”). For the purposes of this study the measure was adapted to refer to someone with a “gambling problem” rather than a “mental illness.”


For sample 1, groups of 10 to 20 students were asked to complete paper and pencil questionnaire measures following ethics approval in order to determine problem gambler stereotype content. Participants were given a consent form and advised that they could withdraw from the study at any time without academic penalty. After obtaining informed consent participants were given a questionnaire package and asked to complete the package of test materials from front to back, without looking ahead or referring back. Anonymity was emphasized in order to minimize the likelihood of social desirability bias. Participants were randomly distributed questionnaires containing one of three category labels (i.e., problem gambler, gambler, gambling addict) centered at the top of the page and instructed to “write down the characteristics that you believe capture the cultural conception of the group…” and that words listed “…may or may not reflect your own personal beliefs about the group.” Instructions stated that words could include “traits, behaviours, beliefs, and so on” and ten blank lines were provided. Questions pertaining to participants' perceptions of gamblers' demographics, appearance, and typical gambling activities followed. Participants were also asked if a particular person (i.e., an exemplar) came to mind. Familiarity with problem gambling, personal gambling involvement, and demographic information was also obtained. Samples 2 and 3 completed a number of questionnaires (Horch, 2011). For the purposes of this report, only the Adjective Checklist data and demographic information are presented. Sample 2 completed the adjective checklist online and Sample 3 completed a paper and pencil version of the checklist.


Analyses were conducted in the Statistical Package for the Social Sciences (SPSS) version 15.0. Table 1 presents demographic information for all samples. Results are presented for Sample 1 first with results of the adjective checklist in Samples 2 and 3 presented at the end of the analysis section. The two student samples did not differ in terms of their gambling involvement. Averaging across both samples, participants reported engaging in between three and four different gambling activities on at least an occasional basis (Range: 0–16, M = 3.65, SD = 3.19), with the most frequently endorsed gambling activities being bingo (62.8%), scratch tickets (49.9%), cards with friends and family (47.2%), raffles (40.7%), and gambling on games of skill (34.1%). Familiarity with problem gamblers items revealed 90.5% of participants had watched a movie or television show depicting a problem gambler, 34.9% had watched a television documentary, 17.8% indicated a friend of the family has a gambling problem, 16.6% had a relative with a gambling problem, 7.5% had a co-worker with a gambling problem, and 2.5% provided services or treatment to individuals with gambling problems. Three individuals (2.9%) stated that they themselves had a gambling problem.

Sample 3 participants were current (74.3%) or past (20.3%; 5.4% missing) gamblers of whom 47.3% had sought treatment. Sixty-six (89.3%) individuals met criteria for pathological gambling (NODS ≥ 5; Gerstein et al., 1999). Participants reported playing card games (36.5%), VLTs (32.4%), slots (29.7%), casino table games including roulette and cards (24.3%), the lottery (16.2%), scratch tickets (8.1%), the internet (6.8%), and betting on races (5.4%) most frequently.

Principal Analyses

Content analysis. To determine stereotype content associated with problem gambling, a content analysis was first conducted on the free response data before a frequency analysis was conducted for both the checklist and free-response data. Participants generated an average of eight words each (M = 7.72, SD = 2.78), with 149 individuals providing at least one word and a total of 1173 words. For the content analysis, data from all labels were collapsed and categories to organize free-response answers were arrived at by consensus of the author and three other researchers in a bottom-up approach. Synonyms, similar words, and similar constructs were first grouped together before broad themes, or major categories, began to emerge. Subcategories within each of the broad categories were then further refined.

When all the data were considered together, three major categories of responses were determined: (1) consequences of gambling, (2) addiction, and (3) characteristics of problem gamblers. A fourth (4) miscellaneous category was also created consisting of three subcategories: motivations to gamble, gambling words, and miscellaneous. Consequences of gambling, from most frequently mentioned to least, included: financial, social, personal/ psychological, and occupational. The addiction category included concurrent addictions, addictive personality, the term addiction or addict, and the construct of denial. Concurrent addictions generated included tobacco, alcohol, and other substance and behavioural addictions. Finally, characteristics of problem gamblers generated could be grouped into the following subcategories: demographics, affect, and personality traits. The most frequent demographics generated suggested that an uneducated, lower class divorced male in his mid 40s to 50s came to participants' minds, although other descriptors were also generated. Affect consisted of words reflecting depression, anxiety, mood lability, and brokenness or neediness. Table 2 lists the categories under personality traits, provides examples of words in each category, and also indicates the frequency with which words were generated based on labels and in total (i.e., collapsed across labels).

Frequency Analyses

Frequency analyses were conducted for both the free response and checklist data. Chi square analyses were run for each major content domain (i.e., consequences of gambling, addiction, characteristics of problem gamblers, and miscellaneous) and for the 16 personality traits subcategories to test the null hypothesis that participants in all conditions (i.e., labels) endorsed a content domain equally. Chi square analyses based on labels for the four major categories indicated that label produced significantly different frequencies of gambling-related consequences, χ2 (2) = 8.93, p = .011, and miscellaneous words, χ2 (2) = 11.54, p = .003. A follow-up chi-square indicated that problem gambler resulted in participants listing more negative consequences than gambler, χ2 (1) = 8.81, p = .003, and that gambling addict resulted in more negative consequences than gambler, χ2(1) = 4.61, p = .033. Gambler produced more miscellaneous words than problem gambler, χ2 (1) = 4.79, p = .027, and gambling addict, χ2 (1) = 10.0, p<.002. Regarding personality traits (Table 2), analyses revealed significant differences based on label for thrill seeking, χ2 = 6.42, p = .040, cognitive style, χ2 = 6.64, p = .036, and extraversion, χ2 = 6.33, p = .042. Furthermore, the label gambler resulted in the most words associated with risk while the labels problem gambler or gambling addict resulted in the most words associated with compulsion, obsession, or lack of control.

A frequency analysis was also conducted with the checklist data provided by 106 participants. The ten most frequently endorsed words thought to be most characteristic of problem gamblers were: compulsive (endorsed by 26% of participants), impulsive (26%), irresponsible (25%), risk-taking (22%), greedy (19%), desperate (17%), depressed (14%), deceitful (11%), irrational (11%), and careless (11%). Least characteristic words were: dependable (27%), responsible (25%), realistic (23%), clear thinking (22%), self-controlled (19%), stable (17%), rational (17%), cautious (16%), practical (11%), and honest (11%).

Additional Stereotype Descriptions

After completing the free response task, participants in the present study were asked if they imagined a specific individual, what the individual looked like, and the individual's age, gender, race, marital status, and employment type and status. They were asked if the person was someone they knew and whether the individual was engaged in a specific type of gambling. Sixty-seven participants (44%) indicated that they were thinking of a specific person. Of these participants, 57 (85%) thought of a male, six (9%) thought of a female, two (3%) said both, and two (3%) did not indicate gender or were unsure. The majority of participants imagined a Caucasian individual (n = 48, 71%), seven (10%) imagined an Asian person, four (6%) imagined a First Nations individual, one imagined an Iranian, and one imagined a South Asian individual, with six (9%) not indicating ethnicity. Marital status of imagined individuals was mostly single (23, 34%), married (17, 25%), or divorced (15, 22%), and the remainder (8, 12%) saying something other or a combination of the above. Only 25 (37%) individuals stated they thought of someone they knew. Most thought of either a friend, friend of the family, or friend's relative (13, 19%), or of a family member such as an uncle, grandfather, or brother-in-law (7, 10%). One individual indicated they thought of a neighbour, two (3%) thought of a customer at their bar, one “a work colleague,” one “a person begging for money in Vegas,” one person wrote “idol,” and one individual indicated “it was me.” Seventy participants (46%) indicated they thought of a specific type of gambling activity, with the majority thinking of slots/VLTs (23, 33%), card games such as poker or blackjack (21, 30%), and some indicating both slots and card games (11, 16%). The remainder (14, 20%) thought of bingo, lotteries, horse racing, sports betting or craps, either alone or in combination with the above activities.

Adjective Checklist Data in Two Additional Samples

The ACL was modified as described in the methods section and administered to two additional samples. Table 3 presents and contrasts the ten words most and least characteristic of problem gamblers in both samples. Both frequency (i.e., the percent of participants who endorsed the item) and mean weighting on the rating scale (1  =  very uncharacteristic of problem gamblers, 7 =  very characteristic of problem gamblers) are provided.


The primary objective of this study was to elicit the content of the stereotypes that university students hold about problem gambling. Content and frequency analysis of the open response and checklist data appear to paint a similar picture. As predicted, addiction, impulsivity and several negative personality traits were elicited. Specifically problem gamblers were considered to be compulsive, impulsive, desperate, irresponsible, risk-taking, depressed, greedy, irrational, antisocial, and aggressive. They were considered to not be dependable, responsible, realistic, clear thinking, self-controlled, stable, rational, cautious, practical, or honest. When later samples responded to a modified checklist, university students and individuals with a gambling problem consistently indicated problem gamblers were addictive, obsessive, compulsive, impulsive, broke, and desperate. Uncharacteristic words were largely consistent between all three samples with both university students and problem gamblers endorsing dependable, cautious, clear thinking, stable, logical, and honest. When considering this data in comparison with previously known stereotypes of mental illness and addiction it seems stereotypes were more associated with addiction (e.g., compulsive, risk-taking, antisocial, aggressive, irresponsible) than with mental illness (e.g., depressed, irrational).

A secondary objective was to determine the utility of free response as compared to checklist-derived data. Characteristics generated were consistently produced using free response and checklist methodologies, although to different extents. Furthermore, as predicted, both methods produced overlapping and different content with free response generating more physical descriptors and non-adjective words (e.g., money, casino, lifestyle). Specifically, all words endorsed on the checklist measure were also generated using open response, although with a lower frequency. There were, however, some important differences in information gained from the two different methodologies. Overall, the open response method generated richer data that were more difficult to interpret. Open response permitted participants to generate information in addition to traits and characteristics, including demographics, expected comorbidities, and impairments or consequences of problem gambling, but also resulted in miscellaneous and perhaps idiosyncratic or irrelevant words. It has been argued that checklist methodology constrains participants to the researchers' conceptions and may miss important elements of a stereotype that are not provided on the checklist. As such, open response is particularly important in a first study on a subject and is perhaps more thorough. However, free response with ten data points and a sample of 152 was unwieldy. Future stereotype content research might use a smaller sample or request that fewer words be generated. Furthermore, the checklist clearly resulted in a similar list of characteristics and was much easier to interpret. However, this task did not appear to be well-understood by participants.

A third objective was to examine the impact of labels on words generated with open response. There were significant differences in the frequencies with which words or content domains were represented based on the label presented. The problem gambler and gambling addict labels generated more words regarding negative consequences of gambling and gambler resulted in more miscellaneous words. Despite the possible importance of distinctions in labels, it seemed most characteristics were generated with both problem gambler and gambling addict labels with minor variations in frequency. Instead, the perspectives of individuals receiving labels may be a more important consideration in the selection of labels.

Remarkably, the term gambler did not appear to be neutral in valence (although it was perhaps somewhat less negative than problem gambler and gambling addict labels) as many negative characteristics were generated that overlapped with the problem gambler and gambling addict labels. Scull and Woolcock (2005) explored the meaning of the word gambling in non-English speaking background communities in Australia and noted that Chinese participants stated that using the term gambling, as opposed to play or “try your luck,” tended to imply that a person's behaviour was in some way problematic. Accordingly, the use of the term gambler might be considered akin to the term drinker and when an individual is so labelled it may typically be assumed that the person is not merely a social drinker. Furthermore, the term gambler labels an individual based on behaviour. It would seem that participants may have inferred that the gambler was actually a problem gambler rather than a social or recreational gambler. However, it is uncertain as to whether a term such as person who gambles or other such descriptions would have produced different results. Research examining this question in the domain of schizophrenia found that more politically correct labels (i.e., consumer of mental health services) were associated with less negative reactions and greater perceived likelihood of change, but resulted in greater attributions of responsibility and did not result in greater intention to interact with such persons (Penn & Nowlin-Drummond, 2001).

An interesting consideration concerning stereotypes is the importance of accuracy. The grain of truth hypothesis (Allport, 1979; Campbell, 1967) holds that stereotypes exist because, at some level, there is some truth to them. In keeping with this, participants listed traits that were somewhat consistent with the current conceptualization of pathological gambling in the DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, 4th ed., text rev.; American Psychiatric Association, 2000). Of the ten DSM criteria, preoccupation, the inability to quit, lying, gambling to escape, and relational problems were all represented by words generated (i.e., “frequently thinking about it,” “lack of self-control,” “deception,” “seeks escape,” and “has lost valuable relationships,” respectively). Furthermore, common comorbidities with alcohol, depression, and nicotine dependence were also generated. Research has shown that these comorbidities exist in the problem gambling population at higher rates than in the general population. For instance, individuals who have developed a gambling problem have a six-fold increased risk of having a diagnosable alcohol use condition in their lifetime (Petry, Stinson, & Grant, 2005), are about three times more likely to suffer from a mood disorder (Petry et al., 2005), and 41.6% of heavy gamblers are nicotine users (Smart & Ferris, 1996). A recent meta-analysis of eleven population surveys found that problem gamblers had high rates of comorbid substance use disorders (57.5%), mood disorders (37.9%), and nicotine dependence (60.1%; Lorains, Cowlishaw, & Thomas, 2011). Finally, the demographic of an uneducated, lower class, divorced male in his mid 40s to 50s is not entirely inaccurate either, as research supports that more men than women develop gambling problems (Afifi, Cox, Martens, Sareen, & Enns, 2010), although this gender gap is closing (Wong, 2005), and individuals with lower SES are also more likely to develop gambling problems (Toneatto & Nguyen, 2007). Despite these potential grains of truth, many words were generated which were not particularly accurate or which may apply to some individuals but not all (e.g., “dangerous,” “unintelligent,” “rude”). Furthermore, some terms were downright discriminatory (e.g., “chump,” “evil,” “weak”). Such unfairly deleterious terms are why stereotypes are considered harmful.

This study was not without limitations. First, the above findings on problem gambling stereotypes may not generalize beyond the populations sampled. Prejudicial attitudes are deeply embedded in social and cultural norms (Goffman, 1963), and demographic variables such as age, ethnicity, and culture are likely to play a role in the content of social norms and attitudes, including stereotypes. Past research has found that discrimination towards individuals with a gambling problem may differ with age (Rockloff & Schofield, 2004) and ethnicity (Horch & Hodgins, 2008). Past research has also determined that some cultures are much more condoning of gambling (e.g., Chinese culture, Raylu & Oei, 2004). Cultural variables such as values, beliefs, and level of acculturation may all play a role in whether an individual initiates gambling activity or develops problem gambling (Raylu & Oei, 2004). Given the roles of these demographic variables, a community sample would have been more representative of at least the Alberta population. At the same time, stereotypes are, by definition, broadly held beliefs and participants were asked to provide “the cultural conception” of the group which “may not reflect… personal beliefs.” Therefore, as a preliminary investigation of labelling and stereotyping of problem gambling, a university sample is sufficient, and the additional problem gambling sample is of added interest.

A second limitation is that the adjective checklist and the corresponding rating scale were not well understood by participants. It seemed that many participants did not understand the rating scale following the checklist measure or perhaps misread the valence. This was the case whether presented on paper or online. The ACL was modified to include the use of non-characteristic words and to rate all words on a scale of very uncharacteristic to very characteristic of problem gamblers. This was done in keeping with suggestions in the literature to combine open and checklist strategies and to provide a measure of stereotype strength. However, the resulting measure was apparently too complex and would have benefited from being piloted.

Stereotype content regarding problem gamblers may be important both to better understand cultural conceptions of problem gamblers and to perhaps modify inaccurate or harmful conceptions in order to reduce stigmatization or to conduct other interventions aimed at increasing treatment-seeking in this population. Despite results indicating that the problem gambling stereotypes include traits such as compulsive, irrational, impulsive, risk-taking, antisocial, aggressive, irresponsible, greedy, depressed, addictive, obsessive, broke, and desperate, it is still unclear which of these descriptions invite stigmatization. Furthermore, stereotypes contained more than a grain of truth and were a somewhat accurate description of diagnostic criteria, comorbidities, and demographic characteristics of individuals with a gambling problem. However, some terms generated were unfairly deleterious and may therefore warrant modification.

Adlaf, E. M.. , Demers, A.. , & Gliksman, L. . ( 2005). Canadian campus survey 2004. Toronto, ON: Centre for Addiction and Mental Health.
Afifi, T. O.. , Cox, B. J.. , Martens, P. J.. , Sareen, J.. , & Enns, M. W. . ( 2010). Demographic and social variables associated with problem gambling among men and women in Canada. Psychiatry Research, 178, 395–400.
Allport, G.W. . ( 1979). The nature of prejudice. New York: Doubleday Anchor Books.
(Original work published 1954.)
American Psychiatric Association ( 2000). Diagnostic and statistical manual of mental disorders, (4th ed., Text Rev.). Washington DC: Author.
Campbell, D. . ( 1967). Stereotypes and the perception of group differences. American Psychologist, 22, 817–829.
Corrigan, P. W. . ( 2004). How stigma interferes with mental health care. American Psychologist, 59, 614–625.
Corrigan, P. W.. , River, L. P.. , Lundin, R. K.. , Wasowski, K. U.. , Campion, J.. , Mathisen, J.. , . ( 2000). Stigmatizing attributions about mental illness. Journal of Community Psychology, 28, 91–102.
Devine, P. G.. , & Baker, S. M. . ( 1991). Measurement of racial stereotype subtyping. Personality and Social Psychology Bulletin, 17, 44–50.
Gardner, R. C.. , Kirby, D. M.. , Gorospe, F. H.. , & Villamin, A. C. . ( 1972). Ethnic stereotypes An alternative assessment technique, the stereotype differential.. The Journal of Social Psychology, 87, 259–267.
Gerstein, D.. , Murphy, S.. , Toce, M.. , Hoffman, J.. , Palmer, A.. , Johnson, R.. , . ( 1999). Gambling impact and behavior study: Report to the national gambling impact study commission. Chicago: National Opinion Research Centre.
Goffman, E. . ( 1963). Stigma: Notes on the management of spoiled identity. Englewood Cliffs, NJ: Prentice Hall.
Gough, H. G.. , & Heilbrun, A. B. . ( 1983). The adjective checklist manual. (1983 edition). Mountain View, CA: Consulting Psychologists Press.
Hodgins, D. C.. , & el-Guebaly, N. . ( 2000). Natural and treatmentassisted recovery from gambling problems A comparison of resolved and active gamblers.. Addiction, 95, 777–789.
Holmes, E. P.. , Corrigan, P.W.. , Williams, P.. , Canar, J.. , & Kubiak, M. A. . ( 1999). Changing attitude about schizophrenia. Schizophrenia Bulletin, 25, 447–456.
Horch, J. D.. , & Hodgins, D. C. . ( 2008). Public stigma of disordered gambling Social distance, dangerousness and familiarity.. Journal of Social and Clinical Psychology, 27, 505–528.
Katz, D.. , & Braly, K. . ( 1933). Racial stereotypes of one hundred college students. The Journal of Abnormal and Social Psychology, 28, 280–290.
Kemmelmeier, M.. , Danielson, C.. , & Basten, J. . ( 2005). Whats in a grade Academic success and political orientation.. Personality and Social Psychology Bulletin, 31, 1386–1399.
Lorains, F. K.. , Cowlishaw, S.. & Thomas, S. A.. ( 2011). Prevalence of comorbid disorders in problem and pathological gambling Systematic review and metaanalysis of population surveys.. Addiction, 106, 490–498.
Madon, S. . ( 1997) What do people believe about gay males A study of stereotype content and strength.. Sex Roles, 37, 663–685.
Marín, G. . ( 1984). Stereotyping Hispanics The differential effect of research method, label, and degree of contact.. International Journal of Intercultural Relations, 8, 17–27.
National Research Council ( 1999). Pathological gambling: A critical review. Washington, DC: National Academy Press.
Niemann, Y. F.. , Jennings, L.. , Rozelle, R. M.. , Baxter, J. C.. , & Sullivan, E. . ( 1994). Use of free responses and cluster analysis to determining stereotypes of eight groups, Personality and Social Psychology Bulletin, 20, 379–390.
Operario, D.. , & Fiske, S. T. . ( 2004). Stereotypes Content, structures, processes, and context.. In Brewer, M. B. & Hewstone, M.
(Eds.), Social cognition (pp. 120–141). Malden, MA: Blackwell Publishing.
Penn, D. L.. , & Nowlin-Drummond, A. . ( 2001). Politically correct labels and schizophrenia: A rose by any other name?Schizophrenia Bulletin, 27, 197–203.
Petry, N. M.. , Stinson, F. S.. , & Grant, B. F. . ( 2005). Comorbidity of DSMIV pathological gambling and other psychiatric disorders Results from the National Epidemiological Survey on Alcohol and Related Conditions.. Journal of Clinical Psychology, 66, 564–574.
Rasinski, K. A.. , Woll, P.. , & Cooke, A. . ( 2005). Stigma and substance use disorders. In Corrigan, P. W.
(Ed.), On the stigma of mental illness: Practical strategies for research and social change (pp. 219–236). Washington, DC: American Psychological Association.
Raylu, N.. , & Oei, T. P. . ( 2004). Role of culture in gambling and problem gambling. Clinical Psychology Review, 23, 1087–1114.
Rockloff, M, J. , & Schofield, G.. ( 2004). Factor analysis of barriers to treatment for problem gambling. Journal of Gambling Studies, 20, 121–127.
Scull, S.. , & Woolcock, G. . ( 2005). Problem gambling in non-English speaking background communities in Queensland, Australia: A qualitative exploration. International Gambling Studies, 5, 29–44.
Smart, R. G.. , & Ferris, J. . ( 1996) Alcohol, drugs and gambling in the Ontario adult population, 1994. Canadian Journal of Psychiatry, 41, 36–45.
Stangor, C.. , & Lange, J. E. . ( 1994). Mental representations of social groups Advances in understanding stereotypes and stereotyping.. Advances in Experimental Social Psychology, 26, 357–416.
Tavares, H.. , Martins, S. S.. , Zilberman, M. L.. , & el-Guebaly, N. . ( 2002). Gamblers seeking treatment: Why haven't they come earlier?Addictive Disorders and Their Treatment, 1, 65–69.
Toneatto, T.. , & Nguyen, L. . ( 2007). Individual characteristics and problem gambling behavior. In Smith, G. , Hodgins, D. C. , & Williams, R. J.
(Eds.)Research and measurement issues in gambling studies (pp. 279–303). San Diego, CA: Elsevier Inc.
Watson, A.. , & River, P. . ( 2005). A social-cognitive model of personal responses to stigma. In Corrigan, P. W.
(Ed.), On the stigma of mental illness: Practical strategies for research and social change (pp. 145–164). Washington, DC: American Psychological Association.
Williams, J. E.. , & Best, D. L. . ( 1977). Sex stereotypes and trait favorability on the adjective check list. Educational and Psychological Measurement, 37, 101–110.
Wong, T. . ( 2005). The vulnerable faces of pathological gambling. Psychiatry (Edgemont), 2, 34–42.

Table 1 

Participant Demographic Characteristics for Three Samples

Click for larger image

Table 2 

Frequency of Personality Trait Words Generated with Free Response by Label and Total

Click for larger image
Click for larger image

Table 3 

Ten Words Most and Least Characteristic of “Problem Gamblers”

Click for larger image

Article Categories:
  • Research

Related Article(s):