Journal Information
Journal ID (publisher-id): jgi
ISSN: 1910-7595
Publisher: Centre for Addiction and Mental Health
Article Information
© 1999-2003 The Centre for Addiction and Mental Health
Received Day: 16 Month: October Year: 2001
Publication date: February 2002
Publisher Id: jgi.2002.6.7
DOI: 10.4309/jgi.2002.6.7

Slot Machine Gamblers — Why Are They So Hard to Study?
Jonathan Parke, MA Affiliation: Psychology Division, Nottingham Trent University, Nottingham, U.K. E-mail:
Mark Griffiths, PhD Affiliation: Psychology Division, Nottingham Trent University, Nottingham, United Kingdom, E-mail:

[This article prints out to about 11 pages.]
The Opinion section has many purposes including being a forum for authors to offer provocative hypotheses.

–The Editor

This Opinion piece was not peer-reviewed.
For correspondence: Jonathan Parke, Psychology Division, Nottingham Trent University, Burton Street, Nottingham, U.K. NG1 4BU, Phone: +44 115 848 5635, Fax: +44 115 848 6826
Jonathan Parke is a postgraduate researcher and part-time lecturer at Nottingham Trent University and a visiting lecturer at Salford University, Manchester. In the past year, he entered the field of gambling and gaming research with many conference and research papers in the United Kingdom. He has also as a consultant for government and industry.
Mark Griffiths, PhD, is a reader in Psychology at Nottingham Trent University and is internationally known for his research on gambling and gaming addictions. In 1994, he was the first recipient of the John Rosecrance Research Prize for “outstanding scholarly contributions to the field of gambling research.” He has published over 90 refereed research papers, numerous book chapters and over 250 other articles. His current interests are technological addictions, especially computer games and the Internet.

The literature examining the psychology of slot machine gambling is limited. The lack of research seems surprising given the billions of dollars generated from slot machine gambling worldwide coupled with the fact that a small proportion of the population plays them pathologically (Griffiths, 1995). However, we have both spent over 10 years playing in and researching this area and we can offer some explanations on why it is so hard to gather reliable and valid data.

The explanations represent experiences of several research efforts to examine the psychology of slot machine gamblers in the United Kingdom, Canada and the United States. They are roughly divided into three categories:

Player-specific factors

A number of player-specific factors can impede the collection of reliable and valid data. These include such factors as activity engrossment, dishonesty, social desirability, motivational distortion, fear of ignorance, guilt, embarrassment, infringement of player anonymity, unconscious motivation, lack of self-understanding, chasing and lack of incentive to participate in research.

Activity engrossment. Slot machine gamblers can become fixated on their playing almost to the point where they “tune out” to everything else around them. We have observed that many gamblers will often miss meals and even utilise devices (such as catheters) so that they do not have to take toilet breaks. Given these observations, there is sometimes little chance that we as researchers can persuade them to participate in research studies — especially once they are already gambling on a machine

Dishonesty and social desirability. It is well known that some gamblers are dishonest about their gambling behaviour. Social and problem gamblers alike are subject to social desirability factors and may be dishonest about the extent of their gambling activities to researchers as well as to those close to them. This obviously has implications for the reliability and validity of any data collected.

Motivational distortion. Many slot machine gamblers experience low self-esteem and when participating in research may provide ego-boosting responses that lead to motivational distortion. For this reason, many report that they win more (or lose less) than they actually do. Again, this self-report data has implications for the reliability and validity of the data.

Fear of ignorance. We have observed that many slot machine gamblers claim to understand how slot machines work when in fact they know very little. This appears to be a face-saving mechanism so that they do not appear ignorant.

Guilt and embarrassment. Slot machine gamblers may often be guilty and/or embarrassed to be in the gambling environment in the first place. They may like to convince themselves that they are not “gamblers” but simply “social players” who visit gambling environments infrequently. We have found that gamblers will often cite their infrequency of gambling as a reason or excuse not participate in an interview or fill out a questionnaire. Related to this, some gamblers just simply do not want to face up to the fact that they gamble.

Infringement of player anonymity. Some slot machine gamblers play on machines as a means of escape. Many gamblers perceive the gaming establishment in which they gamble as a private arena rather than a public one. Researchers who then approach them may be viewed as infringing on their anonymity.

Unconscious motivation and lack of self-understanding. Unfortunately, many slot machine gamblers do not themselves understand why they gamble. Therefore, articulating this accurately to researchers can be difficult. Furthermore, many gamblers experience the “pull” of slot machines, the feeling of being compelled to play despite better judgment, but they cannot articulate why.

Chasing. Many frequent gamblers do not want to leave “their” slot machine in case someone “snipes” their machine while they are elsewhere. Therefore, it is understandable that most gamblers are also more concerned with chasing losses than participating in an interview or filling out a questionnaire for a researcher.

Lack of incentive. Some slot machine gamblers simply refuse to take part in research because they feel that there is nothing in it for them (i.e. a lack of incentive). Moreover, few gamblers view research about their gambling habits and experiences as potentially helpful to others.

Researcher-specific factors

In addition to player-specific factors, there are also some researcher-specific factors that can impede the collection of data from slot machine gamblers. Most of these factors concern research issues relating to such participant and non-participant observational techniques as blending in, subjective sampling and interpretation, and lack of gambling knowledge.

Blending in. The most important aspect of non-participant observation research while monitoring fruit machine players is the art of being inconspicuous. If the researcher fails to blend in, then slot machine gamblers soon realise they are being watched and are therefore highly likely to change their behaviour. For instance, some players may get nervous, perhaps agitated and stop playing. Others may do the opposite and try to show off by exaggerating their playing ritual. Furthermore, some gamblers will discourage spectators if they consider them to be “skimmers” (i.e. individuals who try to win by playing “other peoples machines”). Blending into the setting depends upon a number of factors, including whether the venue is crowded and easy to wander around in without looking suspicious.

The researcher's experience, age and sex can also affect the situation. In the United Kingdom, amusement arcades are generally frequented by young men and elderly women. If the arcade is not crowded and the researcher does not fit the general profile, then there is little choice but to be one of the “punters.” The researcher will probably need to spend lengthy periods of time in the arcade; therefore, spending money is unavoidable unless the researcher has a job there — an approach which may have benefits (see below).

Subjective sampling and interpretation. It is impossible for the researcher to study everyone at all times and locations in the gambling environment. Therefore it is a matter of personal choice as to what data are recorded, collected and observed. This affects the reliability and validity of the findings. Furthermore, many of the data collected during observation will be qualitative in nature and therefore, will not lend themselves to quantitative data analysis.

Lack of gambling knowledge. Lack of “street knowledge” about slot machine gamblers and their environments (e.g., knowledge of the terminology players use, machine features, gambling etiquette, etc.) can lead to misguided assumptions. For instance, non-participant observation may lead to recording irrelevant data and idiosyncratic interpretation of something that is widely known amongst gamblers. This can also lead to subjective interpretation issues.

External factors

In addition to player- and researcher-specific factors, there are also external factors that can impede the collection of data from slot machine gamblers. Most of these factors involve the gaming industry's reactions to the presence of researchers in their establishments, but there are other factors as well.

Gaming establishment design. Years of research experience have demonstrated that many arcades and casinos are not ideally designed for doing covert research. Non-participant observation is often difficult in small establishments or in places where clientele numbers are low.

Gatekeeper issues and bureaucratic obstacles. The questions of how and where access to the research situation can be gained raise ethical questions. According to Burgess (1984), access is usually determined by an informant (often an acquaintance of the researcher) or gatekeeper (usually the manager). Obtaining permission to carry out research in a gambling establishment can be difficult and is often the hardest obstacle that a researcher has to overcome to collect the required data. Many establishments do not have the power to make devolved decisions and must seek permission from the head office. The industry may prevent access for many reasons. The main ones are described below.

Management concerns. From the perspective of arcade and casino managers, the last thing they want are researchers disturbing gamblers, their customers, by taking them away from their gambling. Furthermore, they do not want researchers to give their customers any chance to feel guilty about gambling. In our experience, management sees researchers in this light, which influences whether they give permission to carry out research.

Industry perceptions. From the many years we have spent researching (and gambling on) slot machines, it has become clear that some people in the gaming industry view researchers as anti-gambling and expect research to report negatively about their clientele, establishment or organisation. As with management concerns, this also has an impact on obtaining permission to carry out research.

Practical advice for collecting data on slot machine gamblers

Having presented what we believe to be the main impediments to collecting data about slot machine gambling, we offer some practical advice in this section on how to get around these potential problems.

Network with the gaming industry. Since gaining formal access to gambling establishments is difficult, it is sensible to network with the gatekeepers in order to facilitate access. The more they know about the researchers and what their goals are, the more likely they are to make a decision based on informed choice.

Be flexible and adaptable in fieldwork. Researchers must constantly monitor their activities, and they have to be flexible and adaptable. For instance, if a researcher enters the field with certain hypotheses, misconceptions may result which will need rapid revisions. Redefining methodology and hypotheses on the basis of early observations may also be necessary (Burgess, 1984)

Collect relevant data. There are few guidelines on what are relevant data when engaged in observational work. Schatzman and Strauss (1973) suggest categorising behaviour into these categories; (a) routine events, in which activities are part of the daily round of life, (b) special events, which are fortuitous but can be anticipated and (c) untoward events, which cannot be anticipated or predicted. Alternatively, Spradley (1980) suggests three different types of observation. These are (a) descriptive observations, which describe the setting, the people and the events that took place, (b) focussed observations, which give the descriptive observations a more detailed portrait and (c) selective observations, which link the questions posed by the researcher.

Introduce incentives to take part in research. To get participants involved, it may be useful to pay the participants, give them gifts or include them in prize draws, etc. There are of course ethical issues concerning giving potential problem gamblers more money with which to gamble, but such issues may be handled on an individual basis.

Utilise data that are already there. For observational purposes it may be possible to use observational behavioural data through such sources as surveillance footage. However, ethical issues here are paramount and may affect if such approaches can be employed at all.

Idiographic methodology. When it is difficult to recruit the appropriate participants, it may be necessary to study a smaller sample size to gain valuable insights through collecting content-rich data through means such as in-depth explorative interviews or observational analysis (see following section) rather than simply doing questionnaires. Researchers' evaluations can thus be triangulated with other methods of data collection in order to be more confident about the validity and reliability of their findings. For example, Griffiths (1995) researched adolescent gambling utilising a range of methodologies including questionnaires, interviews and participant and non-participant observation. If a participant appears to have given socially desirable responses in the questionnaire or initial interview, additional evaluations can be made through observational sessions or a more probing interview.

Observational methodologies. Fieldwork can be ideal for studying “social worlds,” described by Lindesmith, Strauss and Denzin (1975) as “those groupings of individuals bound together by networks of communication or universes of discourse and who share perspectives on reality” (pp. 439-440). There are countless social worlds frequently segmented into various subworlds (Strauss, 1978), many of which go unnoticed, and so-called “invisible worlds” of socially problematic populations (Unrah, 1983).

Whenever possible, it is recommended to supplement self-report data with the use of observational methodologies. Non-participant observation usually relies on the researcher being unknown to the group under study. The one distinct advantage of non-participant observation is that the researcher can study a situation in its natural setting without altering the conditions — but only if the researcher can blend in naturally, as previously discussed. The one obvious advantage is that non-participant observation relies only on observing behaviour. Since the researcher cannot interact in the social behavioural processes, most data collected will be qualitative, interpretative, and to some extent, limited. However, by using other methodological research tools (e.g., structured interviews), suspicions, interpretations and even hypotheses can be confirmed.

Contact treatment agencies. Recovering pathological gamblers may be more helpful in participating in research than gamblers found in gambling establishments. However, there are problems with utilising these populations. They will have distinctive viewpoints on gambling, and gamblers recruited from treatment agencies to participate in research do not represent a cross-section of the continuum of gamblers. These individuals may have gambled much more frequently and taken more risks than the average gambler. Furthermore, they may have experienced significantly higher levels of life disruption as a result of their gambling. Thus, they view gambling as a problem and are motivated and taking positive steps to combat related problems. For these reasons, their opinions and attitudes may well be different from those of the average gambler. Nevertheless, provided that conclusions and generalisations are not based solely on such a population, the data can often make a rich contribution to research findings.

Get employed in a gaming establishment. One way to collect invaluable data is to work in a gaming venue, an approach that has been taken by prominent researchers in this field. For example, Sue Fisher collected all of her observational data while employed behind the change counter of her local amusement arcade. Employment within the environment can be used to establish the researcher's identity and allow blending into the environment. Slot machine gamblers are usually unaffected by onlooking staff because there is no real risk of staff playing their machine when they have finished their gaming (see “skimming” referred to above). Hence, staff are fully permitted to observe playing behaviour and are often required to do so to be vigilant for fraudulent practices. Furthermore, while submerged in this social world, researchers can gather large amounts of relevant and fruitful information indirectly through participation in the gambling environment. We recently utilised this approach to obtain data and it proved effective.

Become a gambler. By becoming a gambler, the researcher can take an auto-ethnographic approach in the collection of data. Auto-ethnography literally means the study of one's own group (Rosecrance, 1986) and involves research processes as well as research methods (Burgess, 1984). It can have a number of advantages; for instance, it may allow acceptance by the group under study, familiarisation with gambling terminology, longitudinal perspective and development of tacit knowledge. According to Hayano (1979), the criteria for auto-ethnographic research are knowledge of the people, culture and language, and the ability to pass as a “native” member of the group.

Obviously, the choice of fieldwork is dictated by the identity of the researcher and it is quite possible for researchers to use this type of methodology without knowing their approach was auto-ethnographic. However, it needs to be remembered that the “insider role” (Rosecrance, 1986) can result in a lack of objectivity resulting in a research bias in interpreting and reporting information. Hayano (1979) countered this argument by stating that subjectivism and personal involvement may not be methodological problems but rather assets that can deepen ethnographic understanding. Furthermore, first-hand experiences of gambling used in conjunction with some form of objective analysis may enhance the researcher's understanding and outlook.

It is hoped that these proposed explanations will benefit future research in this area by providing researchers with an understanding of some of the difficulties of gathering data and offering practical advice on what can be done to facilitate data collection, and thus, improve validity and reliability. Unfortunately, identification of slot machine gamblers is often accomplished by a “search and seek” method of trawling local gambling establishments. Therefore, researchers are often limited to collecting data during playing time and not outside it. Data acquisition would be improved if gamblers were not occupied by playing their slot machine.

Burgess, R.G.. ( 1984). In the Field: An Introduction to Field Research. London: George Allen and Unwin.
Griffiths, M.D.. ( 1995). Adolescent Gambling. London: Routledge.
Hayano, D.. ( 1979). Auto-ethnography: Paradigms, problems and prospects. Human Organization, 38, 99-104.
Lindesmith, A.. Strauss, A.. Renzin, N.. ( 1975). Social Psychology. New York: Holt.
Rosecrance, J.. ( 1986). You can't tell the players without a scorecard: A typology of horse players. Deviant Behaviour, 7, 77-97.
Schatzman, L.. Strauss, A.L.. ( 1973). Field Research: Strategies for a Natural Sociology. Englewood Cliffs, NJ: Prentice Hall.
Spradley, J.P.. ( 1980). Participant Obsevation. New York: Holt, Rinehart and Winston.
Strauss, A.. ( 1978). A social world perspective. In N. Renzin (Ed.), Studies In Symbolic Interaction, Vol.1, pp. 119–128. Greenwich, CN: In Press.
Unrah, D.. ( 1983). Invisible Lives: Social Worlds of the Aged. Beverly Hills, CA: Sage Publications.

Article Categories:
  • Opinion