What Influences Gambling Participation?

When he was 19, Gavin Oregan left home to go to college in Ottawa. He had friends who gambled and got increasingly involved in online sports betting. He always had savings and was always financially responsible. Living with ADHD, he said later, it was not even so much about the money but the thrill of winning that kept him playing. Only when he saw the seriousness of the problem did he know he owed over $30,000. His story suggests his friends, his ADHD, and easy accessibility to online gambling all contributed to the development of his gambling problem (Breaking the Cycle - Mental Health Commission of Canada, 2025). This illustrates the modern reality of gambling, which Encyclopedia Britannica defines as betting or staking of something of value, with consciousness of risk and hope of gain, on the outcome of a game, a contest, or an uncertain event (Glimne, 2019).

Over the past 50 years, Canada has undergone significant social and economic changes, which are reflected in the prevalence of gambling among young people. Normalization began with legalization, evolving from being a prohibited social vice into being an approved part of state-led economic strategy. The 1969 enactment of Bill C-150, which legalized lotteries, served as a catalyst for this legal evolution. In 1985, the federal government transferred regulatory authority to provincial jurisdictions, resulting in a significant jurisdictional shift. By introducing casinos and Electronic Gambling Machines (EGMs), the state was able to transform from a prohibitor to an active entrepreneur, growing the market. This led to a major increase in government revenue. Most recently, single-event sports were made legal in 2021 with the passage of Bill C-128, which further solidified authorized gambling integration into the mainstream economy (Cosgrave & Cormack, 2023).

This phenomenon is linked to the modern crisis of meaning. A crisis of meaning occurs when the dominant meaning-giving myths of society become less “expressive”, causing more to have a vulnerable a sense of purpose. Gambling is potentially one way to fill this void, as it generates excitement, connection, “meaningful” activity, and allows the gambler to hope for a better life.

This paper investigates the extent to which gambling participation is influenced by broader social conditions, as opposed to being solely determined by individual choice. Sociological evidence reveals how exposure to gambling is shaped by social inequalities, that gambling exposure is unequal, individuals who face difficult social circumstances may use gambling to cope with their lives and that income inequality increases the odds of online gambling.It also reviews psychological evidence about individual and behavioral factors associated with problem gambling, including gambling-related risk factors such as mood disorders or ADHD, participation in electronic gambling machines (EGMs), and participating in multiple gambling formats. It presents economic evidence on the incentives to increase gambling and the social costs of excessive participation to examine why gambling is still promoted despite its potential harm. Based on these findings, recommendations will be directed to policymakers like Loto Québec and gambling regulators who have the authority to regulate gambling availability, advertising and harm-reduction strategies. The paper argues that the normalization of gambling is not simply the result of individual choice. It is a structural social issue driven by social inequalities, psychological vulnerabilities, and economic incentives that increase gambling participation and expose vulnerable populations to greater gambling-related harm.

Sociological Factors

Sociology is the systematic study of society, social institutions, and those structural forces that end up shaping human behavior. It examines how social, political, economic, and cultural processes end up influencing the patterns of human action and the way social life is organized. This disciplinary view is especially useful when we talk about understanding gambling participation it recognizes the social anchors that influence understands that the act of gambling is not a set of separate individual decisions but rather "socially anchored" through conditions of inequality, geography, and social networks individuals into playing like inequality, geographic location, and social networks. This section will be based on three articles. First, Papineau et al. (2024) show that priority areas in Quebec have a high density of gambling machines, converging with profitability and with social vulnerability. Beckert & Lutter (2013) argues that people with lower income and people with no or only a high school diploma gamble to escape the monotony of their lives. And Pabayo et al. (2024) shows that income inequality increases the odds of online gambling among adolescent males.

Papineau et al

Papineau et al. (2024) provides empirical evidence of how gambling is physically impacting the population of Québec. This article maps out how gambling machines are distributed in Quebec, showing that the state’s revenue-seeking results in targeting the most vulnerable communities. To identify areas in Quebec where exposure and vulnerability to all types of gambling are high, Loto-Québec developed two indices: the Gambling Exposure Index (GEI), which combines density, accessibility, and the theoretical risk of gambling, and the Gambling Vulnerability Index (GVI), which is based on six weighted socio-economic indicators linked to problem gambling (Papineau et al., 2024, p. 62). A neighborhood is considered highly vulnerable if it has a high concentration of the following factors: areas where a large portion of households earn $60,000 or less annually (approx. 65%), neighborhoods where many residents have a high school diploma or less (50%), a high proportion of residents who are single, widowed, or divorced (approx. 58%), a significant concentration of people between the ages of 20 and 44, higher than average unemployment rates within the area, and a higher proportion of male residents in vulnerable areas (51%) because statistically men have an odds ratio of 2.124 compared to women, meaning they are more than twice as likely to experience problem gambling (According to a compilation of data from ENHJEU 2009-2012, p. 71). The authors carried out analyses to identify areas showing a convergence of high vulnerability and gambling exposure. The data was put onto an interactive online map. The work carried out reveals that 2,599 of the 13,420 dissemination areas in Quebec show high values of vulnerability and exposure to gambling (Papineau et al., 2024, p. 71). 1,394,042 people (17% of the population) live in these areas, designated priority areas for placing gambling infrastructure by Loto Québec. Mauricie–Centre-du-Québec, Abitibi-Témiscamingue, Outaouais, Côte-Nord and Gaspésie–Îles-de-la-Madeleine regions have higher values than the Quebec average. Approximately 48.5% of vulnerable neighborhoods in Quebec have gambling machines (Papineau et al., 2024 pp. 76-77). The reason the target occurs is because Loto-Quebec, the state-owned corporation, has a specific mandate to generate dividends for the Ministry of Finance. The distribution of the machines is driven by where they will be most productive in terms of revenue (vulnerable areas having a higher percentage of machines than non-vulnerable areas).

Pabayo et al

The source identities three primary harms caused by placing gambling machines in vulnerable neighborhoods: people in low-income households spend a proportionally larger share of their available budget on gambling, the presence of the gambling sites outside stimulates the desire to gamble, and those environments make responsible gaming campaigns ineffective because it places the burden of risk of the person while ignoring the commercial determinants of health that drive the addiction (Papineau et al., 2024, p. 66). This source helps my research by providing empirical evidence that the vulnerable have been targeted by a state agency to extract revenue. It demonstrates that the government's role as an entrepreneurial revenue beneficiary prioritizes financial extraction over public welfare. Pabayo et al. (2024) shifts the focus from individual behavioral risk factors to structural social conditions by examining how income inequality influences online gambling participation among Canadian adolescents. By highlighting how unequal environments foster relative deprivation, the study suggests that gambling can be used as a maladaptive strategy for upward mobility, where individuals attempt to improve their financial situation quickly despite the high risk of loss. However, the evidence indicates that gambling is not an effective solution and often worsens financial and psychological outcomes. The researchers conducted a quantitative three-level multilevel modeling analysis. Data were collected through a self-administered paper-based survey as part of the 2018/2019 COMPASS Study, which targeted secondary school students across 136 schools in Canada’s four largest provinces: Ontario, Quebec, British Columbia, and Alberta (Pabayo et al., 2024, p. 291). The study included a sample of 74,501 students aged 13 to 18, with a final analytic sample of 70,134 respondents (Pabayo et al., 2024, p. 291-292). These student-level data were then linked with 2016 Canadian Census data to calculate the Gini coefficient as a measure of income inequality at the school census division level (Pabayo et al., 2024, p. 292-293). The researchers also used bivariate and mediation analyses to examine whether income inequality increased symptoms of depression and anxiety, using beta coefficients (β) to measure the strength of these relationships (Pabayo et al., 2024, p. 294-296). Firstly, the study found that online gambling participation was more prevalent among males than females (Pabayo et al., 2024, p. 296). 2,217 students out of 70,134 (3.2%) reported participating in online gambling for money. 1,813 out of 34,917 young men reported online gambling, and a standard deviation increase in the Gini coefficient (income inequality) was associated with an increased odds ratio (OR) of 1.12, 95% confidence interval (Pabayo et al., 2024, p. 296). Among young women, 404 out of 35,217 students reported online gambling. While the odds ratio was technically higher at 1.19, it was not statistically significant (OR = 1.19, 95% CI 0.95, 1.50), meaning the association between income inequality and online gambling was not statistically significant for women (Pabayo et al., 2024, p. 296). Secondly, it was shown that income inequality increased symptoms of depression and anxiety. An increase in income inequality was significantly associated with higher depressive symptoms (β = 0.07, 95% CI 0.04, 0.11) and anxiety symptoms (β = 0.05, 95% CI 0.01, 0.09). They also found that for everyone standard deviation increases in these symptoms, the likelihood of a student participating in online gambling increased (OR = 1.55 for depression; OR = 1.38 for anxiety) (Pabayo et al., 2024, pp. 294, 295). To be clear, Pabayo et al. (2024) cannot confirm whether gambling causes these symptoms. This source helps answer my research question by demonstrating that gambling participation is influenced not only by individual characteristics but also by broader social inequalities. It provides evidence that income inequality is correlated with online gambling, perhaps through effects on mental health and psychosocial well-being, supporting the argument that structural social conditions increase vulnerability to gambling-related harm.

Bercket and Lutter

Finally, Bercket and Lutter (2013) provides empirical evidence that lottery participation is strongly linked to socio-economic position and social networks rather than just individual choice. It critiques individualistic and behavioral approaches to gambling by testing three sociological frameworks: socio-structural (strain theory), cultural, and social network accounts. Specifically, Bercket & Lutter seek to answer this question: Why do the poor spend more on lottery tickets than their wealthier and educated peers? Beckert & Lutter (2013) conducted a quantitative analysis using a nationwide probability sample of 1,508 residents in Germany (Beckert & Lutter, 2013). To arrive at their conclusions, they applied negative binomial regression models to evaluate how different social variables including education, work dissatisfaction, and peer groups impact monthly lottery expenditure (Beckert & Lutter, 2013). This method allowed them to distinguish between individual psychological factors and the social embeddedness of gambling habits.

The work reveals that lottery participation is a mass phenomenon, as 40% of the sample play at least once a year, 22.3% play regularly (at least once a month), and 17% play at least once a week (Beckert & Lutter, 2013, p. 1160). The researchers found that three distinct sociological factors drive these patterns: first, socio-structural strain: Individuals with lower educational levels and those who express feelings of futility or monotony about their everyday routines spend significantly more money on lottery tickets (Beckert & Lutter, 2013). Second, income inequality: for regular players in the lowest income quintile, expenditure represents 3.72% of their monthly income, while heavy players (the top 25% of spenders) in that same bracket spend an average of 6.73% (Beckert & Lutter, 2013, p. 1161). Finally, social network influence: the social networking of the player is the single strongest predictor of expenditure (Beckert & Lutter, 2013). The model fit improved significantly when including network variables, with the Pseudo-R2 increasing from 0.018 to 0.030 (Beckert & Lutter, 2013, pp. 1162-1163). A limitation of purely psychological models is that they treat gambling as a cognitive error, whereas this source argues it is a socially anchored behavior. Because the state often relies on lottery tickets for fiscal revenue, it essentially extracts funds from the most disadvantaged populations, who use the lottery as a strategy for upward mobility despite the low probability of winning (Beckert & Lutter, 2013). This source is helpful because it provides quantitative proof that social context and structure are just as important as cognition in understanding gambling behavior.

These sociological sources demonstrate that gambling is influenced by social structures and not just individual choice. Research indicates that governments policies and gambling infrastructure disproportionately expose vulnerable communities to gambling, that income inequality increases gambling participation and related mental health risks, and that social networks and socio-economic disadvantage influence gambling behavior. Taking together these findings support the case for the normalization of gambling as a structural social issue.

Psychological Factors

Psychology is the scientific study of human thoughts, feelings, behaviors and those mental processes that guide how people see their surroundings and then respond to them. It looks at how cognitive processes, emotional states, personality factors, and behavioral patterns end up shaping individual choices and actions. This disciplinary perspective is important in the examination of the determinants of gambling involvement and the development of gambling related harm, as it recognizes gambling as a structural social problem, supported by biopsychosocial vulnerabilities, psychological motivations and specific design characteristics of gambling activities, rather than an issue of individual choice or self-control. This section will be based on evidence from threee articles: Murch et al. (2025) first study limitations of Loto-Québec’s Problem Gambling Severity Index (PGSI), finding that the moderate-risk category includes a range of gamblers, the majority of whom have occasional problems, and a smaller group that experiences fewer but more persistent and serious harms. Second, Ioannidis et al. (2019) meta-analyze a previous literature examining the psychological processes affected by gambling disorder, finding evidence that impairments in impulsivity-related processes specifically decision-making, inhibitory control and delay discounting can be linked to gambling disorder. Third, Williams et al. (2021) identify demographic, behavioral, and environmental predictors of problem gambling using data from the Canadian Community Health Survey. They find that factors such as participation in electronic gambling machines, male gender, lower income, and lower educational attainment are linked to a higher risk of gambling. Finally, Gooding and Williams (2024) investigate whether certain gambling formats pose greater risks than others. They find that electronic gambling machines, casino table games, and online gambling are associated with higher rates of problem gambling compared to lower-risk formats like lottery games. Together, these studies show that gambling harm arises from the interaction of different risk profiles, psychological vulnerabilities, social characteristics, and the design of gambling products.

Murch et al

Murch et al. (2025) provide evidence that Quebec’s provincial government harm reduction assessment tool (Problem Gambling Severity Index (PGSI) has important limitations, potentially reducing its effectiveness. The PGSI is a tool used Murch et al. (2025) to assess gambling problems in the general population over the past 12 months (Murch et al., 2025, p. 1072). PGSI uses a hybrid approach that covers 3 main areas: specific behaviors, loss of control, and negative consequences (Murch et al., 2025, p. 1073). By exposing the moderate-risk category as an ambiguous transitional space where individuals may be escalating toward severe addiction, the research challenges the efficacy of the tools the Provincial Government of Quebec uses to manage the social consequences of legalized gambling (Murch et al., 2025, p. 1073). The researchers conducted a quantitative secondary analysis of data collected from the two self-report phases in 2019 and 2022(Murch et al., 2025, p. 1074).The sample consisted of 18,494 Canadian online gamblers recruited directly from the provincially operated Loto Quebec website (Murch et al., 2025, p. 1074). To arrive at their conclusions, they applied unsupervised K-means clustering algorithms to a specific subset of 3,868 moderate risk respondents who scored between 3 and 7 on the PGSI (Murch et al., 2025, p. 1071). This method allowed them to group individuals based on their response variance across the nine PGSI items, distinguishing between persistent and episodic gambling behaviors (Murch et al., 2025, p. 1075.

The study concluded that the moderate-risk category is not uniform but a transitional space, meaning that people’s problems are often moving toward becoming more severe or getting better (Murch et al., 2025, p. 1073). The researchers found that three distinct types of people are grouped together under this one label: The possibly episodic group (61.83%) is the largest group. These individuals report a wide range of problems (guilt, loss chasing; returning another day to try to win back the money that was lost, suspecting/feeling they have a problem, and betting more than they can afford to lose) but they only experience them sometimes (Murch et al., 2025, p. 1077). The specific problems group (8.85%) is the smallest group with a different pattern. They report fewer types of problems (guilt, loss chasing and suspecting a problem), but they experience them much more intensively, often almost always (Murch et al., 2025, p. 1077). Intermediate cases (29.32%) group falls right in the middle, showing a mix of the patterns from the other two groups (Murch et al., 2025, p. 1077). A limitation of the PSGI scoring system is that the state calculates these scores using a mathematically arbitrary approach (Murch et al., 2025, p. 1076). Because the PGSI simply adds points, it treats someone who has one severe, constant problem (3+0+0=3) the same as someone who has 3 minor, occasional problems (1+1+1=3) (Murch et al., 2025, p. 1081). This scoring approach creates ambiguity because the total score does not tell you if the person is dealing with a variety of small issues or a few very dangerous, persistent ones (Murch et al., 2025, p. 1081). After receiving your PSGI score, the AI-based prevention system relies on PGSI scores, its ability to accurately identify individuals at risk depends on the validity of those scores (Murch et al., 2025, p. 1082). Consequently, the algorithm potentially may flag individuals whose gambling behavior does not warrant clinical intervention. For example, it could recommend a clinical treatment service where the respondents would not need clinical treatment, which is an inappropriate use of resources (Murch et al., 2025, p. 1081). This source is good because it provides evidence that the state’s primary method for detecting and preventing gambling harm is flawed. Its risk is that it may contribute to inaccurate identification of individuals who need intervention (Murch et al., 2025, p. 1082).

Ioannidis et al

Ioannidis et al. (2019) examine whether impulsivity is a consistent psychological characteristic of individuals with gambling disorder and problem gambling by synthesizing evidence across previous empirical studies. The researchers note that although impulsivity is central to psychological and neurobiological theories of gambling disorder, no previous study had systematically examined all major domains of impulsive cognition together (Ioannidis et al., 2019, pp. 1354–1355). To address this gap, they conducted a systematic review and meta-analysis of case-control studies comparing individuals with gambling disorder or problem gambling to healthy controls across multiple cognitive domains related to impulsivity (Ioannidis et al., 2019, pp. 1354–1355). Their review included 52 independent studies, and random-effects meta-analyses were performed to calculate standardized effect sizes while moderation analyses examined whether age, gender, geographical region, study quality, and psychiatric comorbidities influenced the results (Ioannidis et al., 2019, pp. 1355–1356). This approach allowed the researchers to identify psychological characteristics consistently associated with gambling disorder rather than relying on findings from a single sample. The study concluded that individuals with gambling disorder demonstrate significant impairments across several domains of impulsive cognition. Compared with healthy controls, they showed poorer motor inhibition, meaning greater difficulty stopping impulsive actions, attentional inhibition, meaning greater difficulty ignoring gambling-related cues, delay discounting, meaning a stronger preference for immediate rewards over larger delayed rewards, and decision-making, indicating a tendency to make riskier and less advantageous choices (Ioannidis et al., 2019, pp. 1356–1359). Among individuals classified as problem gamblers, impaired decision-making was the only cognitive domain with sufficient evidence for meta-analysis, suggesting that poor decision-making may represent an early psychological vulnerability before gambling disorder fully develops (Ioannidis et al., 2019, pp. 1358–1359). Overall, the researchers concluded that heightened impulsivity is a robust psychological feature of gambling disorder and that deficits in decision-making may help identify individuals who are at risk of progressing toward more severe gambling problems (Ioannidis et al., 2019, pp. 1359–1360).

This source is valuable because it explains why some individuals are more vulnerable to gambling-related harm from a psychological perspective. Unlike studies that focus primarily on gambling behavior or demographic characteristics, Ioannidis et al. (2019) identify underlying cognitive processes that increase susceptibility to gambling disorder. Their findings suggest that impaired decision-making and impulsivity are important psychological mechanisms contributing to gambling harm, helping explain why some individuals develop persistent gambling problems while others do not (Ioannidis et al., 2019, pp. 1359–1360). This source also complements research on environmental risk factors by demonstrating that gambling behavior is influenced not only by gambling availability but also by individual cognitive vulnerabilities.

Williams et al

Williams et al. (2021) provides evidence about which demographic groups are statistically most likely to develop gambling problems using demographics, behavioral predictors, and social characteristics (Williams et al., 2021, p. 521). The research identifies key predictor categories that help explain which individuals are at greater risk for problem gambling (Williams et al., 2021, p. 521). The researchers conducted an analysis of data collected from a national population study known as the 2018 Canadian Community Health Survey (CCHS) (Williams et al., 2021, p. 522). The sample consisted of a broad group of 23,952 Canadian adults recruited for the purpose of identifying characteristics most strongly associated with problem gambling (Williams et al., 2021, p. 523). To arrive at their conclusions, they applied a stepwise binary logistic regression to a specific subset of variables including demographics, mental health, and game play patterns (Williams et al., 2021, p. 525). The logistics is a statistical process used to predict a single “either/or” outcome distinguishing problem gamblers from non-problem gamblers by entering variables into the model one by one and only retaining those that provide unique predictive power not already explained by other factors (Williams et al., 2021, p. 525). This method allowed them to estimate how strongly different variables predict the likelihood of being a problem gambler (Williams et al., 2021, p. 525). The study concluded that the gambling profile is not a single uniform but a multifaceted set of predictors, meaning that people’s risks are often tied to specific social and behavioral indicators (Williams et al., 2021, p. 525). The researchers found that three distinct types of data are the most important predictors: firstly, the Electronic Gambling Machine (EGM) participation is the single strongest independent behavioral predictor of problem gambling status (Williams et al., 2021, p. 525). In the regression model, EGM participation had a Wald statistic (a measure of how strongly and significantly a variable predicts an outcome) of 137.33, which was significantly higher than any other variable, and an odds ratio of 15.14, meaning EGM players are over 15 times more likely to be problem gamblers (Williams et al., 2021, p. 527). Other high-risk behavioral patterns that added predictive power included participation in casino table games (Wald 15.81), speculative financial activities (Wald 6.25), and instant lottery tickets (Wald 5.94) (Williams et al., 2021, p. 527).

Secondly, social predictors highlight that problem gambling rates vary significantly by gender, income, and education. For example, 0.7% of males in the sample were categorized as problem gamblers, compared to 0.4% of females (Williams et al., 2021, p. 526). Out of household incomes of $40,000-59,900, 0.8% were problem gamblers and 3.5% were at risk, and 0.8% who earned $60,000-79,900 were problem gamblers and 2.5% were at risk (Williams et al., 2021, p. 526). Compared to the highest income bracket ($150,000+), only 0.2% were problem gamblers (Williams et al., 2021, p. 526). Out of educational attainment, 0.7% of problem gamblers only had a secondary school diploma and 3.2% at-risk (Williams et al., 2021, p. 526). Individuals with any form of post-secondary education less than a bachelor's degree were recorded 0.7% to be problem gamblers and 3.0% at-risk (Williams et al., 2021, p. 526). Environmental factors were found to be exceptionally strong predictors of harm at the provincial level. A very strong correlation (r = 0.93) between the number of EGMs available per 1,000 people and the combined provincial at-risk and problem gambling rates (Williams et al., 2021, p. 525). In Ontario and British Columbia, the only two provinces that do not permit EGMs outside of dedicated gambling venues have the lowest problem gambling rates in Canada at 0.3% (Williams et al., 2021, p. 526). By contrast, Manitoba (9.2 EGMs per 1,000 people) has a problem gambling rate of 1.2%, which is the highest in the country (Williams et al., 2021, p. 525, 526, 527). Quebec’s problem gambling rate is not too far with 0.7% (Williams et al., 2021, p. 526). This source helps answer my research question by identifying the demographic, behavioral, and environmental characteristics that are statistically associated with problem gambling in Canada. However, individual vulnerability alone does not explain why some of gambling create greater risks than others. The structure and design of gambling products themselves also influence the likelihood of harmful participation.

Gooding and Williams

Gooding and Williams (2024) extend this analysis by examining the intrinsic risks associated with different gambling formats. The researchers were trying to prove whether certain types of gambling like slot machines are naturally more addictive than others, or if gambling problems just happen to people who play a lot of everything. To do this, they studied a massive group of 10,199 Canadian gamblers, which included 1,346 people who were already struggling with serious gambling problems. They even followed up with 4,707 of these people one year later to see who developed new problems over time, which represented 82.5% of the participants who had agreed to stay in the study (Gooding & Williams, 2024).The follow-up was conducted through an online panel survey administered by Leger Opinion, where participants were about specific gambling behaviors they did over the past 12 months, including: types of gambling, (Lottery, EGMs, casino table games, sports betting, etc.) whether they played in-person or online, the frequency of play, and total time total amount of money spent on gambling.

The researchers proved that the biggest warning sign of a gambling problem is breadth, which is simply playing too many kinds of games at once. They found that while a typical gambler only plays about 1.8 different types of games, people with gambling problems play an average of 3.2 types (Gooding & Williams, 2024, p. 562). In fact, if someone plays four or more different types of gambling, they are statistically more likely to have a problem than not (p. 563). This happens because playing more types of games usually means a person is spending more time and money gambling overall. However, even when they accounted for people playing many different games, the researchers proved that Electronic Gambling Machines (EGMs) like slot machines and casino table games are still the most dangerous. These games are continuous, meaning you can play them repeatedly very fast, getting a constant hit of excitement. The study showed that 60.6% of people who play casino table games monthly and 44.0% of those who play slot machines monthly are problem gamblers (p. 561). In contrast, only 12.1% of monthly lottery players had those same problems (Gooding & Williams, 2024, p. 561). The most important proof came from the one-year follow-up. The researchers found that people who played slot machines were 1.92 times more likely to develop a gambling problem in the future (Gooding & Williams, 2024, p. 563). Even after the researchers adjusted the math to account for the fact that these people were playing many different games, the risk for slot machines was still 1.39 times higher than normal (Gooding & Williams, 2024, p. 563). Finally, when they asked the gamblers themselves which game caused them the most trouble, 61.7% pointed directly to slot machines (Gooding & Williams, 2024, p. 564). They also found that online gambling is a major risk, with 31.1% of problem gamblers saying their issues were mostly related to playing online (Gooding & Williams, 2024, p. 564). Collectively, this proves that while playing too many games is bad, slot machines and online platforms are designed to be much more addictive than things like a weekly lottery ticket.

Psychology Conclusion

As can be seen from the above psychological findings, gambling is an activity that depends on both personal vulnerability and gambling environment, thus creating conditions for developing gambling harm. The results of the research indicate that the development of gambling issues is a rather complicated process that does not happen in the same manner for everyone, because the moderate-risk category combines two different groups: a majority of people who experience occasional and infrequent gambling problems and a smaller group who experience fewer types of problems but more persistent and severe gambling-related harms (Murch et al., 2025). In addition, it should be noted that there are individuals who are more at risk when it comes to gambling issues because of specific demographic, psychological, and behavioral characteristics (Williams et al., 2021). Specifically, young adult males and Indigenous populations, as well as individuals with mood disorders. Psychological vulnerabilities are also an important factor. Impairments of impulsivity-related processes are observed in individuals with gambling disorders, such as decision-making, inhibitory control, and the ability to delay immediate rewards (Ioannidis et al., 2019). Finally, there are gambling products that may involve certain elements that increase the risk of problematic participation and addictive behavior (Gooding & Williams, 2024). For example, gambling formats like electronic gambling machines and casino table games. Taken together, these findings support the view that gambling-related harm arises from an interaction between individual vulnerability, psychological processes and characteristics of gambling environments. Impulsivity, chasing losses, participation in multiple gambling activities, demographic risk factors and exposure to highly reinforcing formats of gambling are all factors contributing to the development of problematic gambling behavior.

Economic Factors

Economics is the study of how society manages its scarce resources. Resources such as money, time and opportunities are limited. It is the study of how people, groups and nations make choices within limits. It also studies how incentives, markets and external forces affect these choices. From this perspective, gambling should not be considered simply as a personal choice based on personal preferences or risk-taking behavior. Rather, gambling participation takes place in an economic context influenced by accessibility, technology, consumer incentives, and the availability of finances. These factors influence both the motivations for gambling and the outcomes of gambling activity. To demonstrate this argument, the section examines three studies that analyze different dimensions of the relationship between economics and gambling behavior. First, Humphreys et al. (2021) show that gambling participation is influenced by economic accessibility, finding that greater availability of gambling facilities increases participation while recreational gambling can provide limited consumption benefits. Second, Effertz et al. (2018) examine the negative externalities created by the expansion of online gambling, demonstrating that increased access to digital gambling platforms raises the risk of problem gambling and generates significant healthcare costs. Finally, Cesarini et al. (2016) analyze whether gambling-related wealth gains improve long-term economic and social outcomes, finding that large lottery winnings do not produce lasting improvements in health or children’s educational achievement. Together, these studies provide evidence that gambling behavior is not simply the result of individual preferences but is shaped by economic environments that influence opportunities, incentives, and consequences.

Humphreys et al

Humphreys et al. (2021) examine the broader costs and benefits of legalized gambling by investigating whether recreational gambling affects health and well-being. Using economic models, the study analyzes the relationship between recreational gambling and various health outcomes. This study suggests that recreational gambling may provide psychological and consumption benefits when it does not develop into problem gambling. Humphreys et al. (2021) conducted a quantitative econometric analysis using data from five cycles (2002-2009) of the Canadian Community Health Survey (CCHS), which is a nationally representative random-digit-dial telephone survey. The study utilized a dataset of 184,221 total observations, resulting in a final analytic sample of 119,399 recreational gamblers and 65,768 at-risk gamblers, both compared to non-gamblers. To examine whether gambling participation causes changes in health outcomes, the authors used instrumental variable (IV) probit and recursive bivariate probit models. These methods address potential endogeneity by accounting for unobserved factors that may influence both gambling behavior and health. They used the density of gambling facilities per capita in each province as an instrumental variable, if gambling availability influences participation but does not directly affect health outcomes except through gambling behavior. This source concludes that many gambling participants are recreational rather than problem gamblers, with 92.55% of active gamblers classified as non-problem gamblers. Problem gambling was measured using a modified version of the Canadian Problem Gambling Index (CPGI), where active gamblers were categorized as non-problem if they reported experiencing no negative impacts or behavioral issues. Problem gambling was found to be statistically rare, affecting only 0.35% of the total sample and 0.73% of all gamblers. The study also provides evidence that recreational gambling (anyone who participated in any form of wagering at least once in the previous 12 months) generates positive consumption benefits by functioning as a form of adult leisure, reducing the probability of experiencing high life stress by 9.3%. Stress is measured using a binary variable that takes a value of 1 if a respondent reports that most days are quite a bit stressful or extremely stressful and 0 otherwise, based on a CCHS question. It is also decreasing the probability of being dissatisfied with life by 7.1%. In addition, recreational gambling was associated with a 3.2% reduction in the probability of stroke and a 2.7% reduction in the probability of stomach ulcers, while having no significant effect on the likelihood of heart disease, high blood pressure, or diabetes (Humphreys et al., 2021, pp. 43-44). Lastly, the study found that greater regional access to gambling facilities directly increases gambling participation, playing an important role in consumer behavior (Humphreys et al., 2021, p. 41). In the IV-probit model, the coefficients for gambling facilities per is 0.076 (p < 0.01) (Humphreys et al., 2021, p. 50), and the bivariate probit models, it is even higher, ranging from 0.201 to 0.213 (p < 0.001) (Humphreys et al., 2021, p. 54). They indicate that as the density of facilities increases, the probability of participation rises. Humphreys et al. (2021) concluded that gambling availability is a causal driver of participation. The instrument produced Kliebergen-Paap Wald F-statistics (a test used to determine whether an instrumental variable is strong and reliable) between 17.77 and 18.48, exceeding the critical threshold of 16.38 required to show regional access is a reliable predictor of consumer behaviour (Humphreys et al., 2021, p. 45). For example, in 2002, Newfoundland/Labrador had a high-access density of 4.62 facilities per 1,000 population, whereas Ontario had a low-access density of only 1.14 facilities per 1,000 population (Humphreys et al., 2021, p. 37). This source helps answer my research questions by showing that legalized gambling has both potential benefits and costs. It suggests that recreational gambling may provide benefits, while major harms are concentrated among problem gamblers. However, this raises an important question: what happens when gambling is no longer recreational and begins to develop into problematic behavior?

Effertz et al

Effertz et al. (2018) addressed this question by examining the economic costs associated with legalized online gambling, particularly when gambling participation leads to problem gambling and increased health-related expenses. It shows the negative externalities and societal costs created by excessive gambling behavior. They used data from PAGE (Pathological Gambling and Epidemiology), a cross-sectional survey that was conducted in Germany via a computer-assisted-telephone-interview (CATI) procedure. A total of 15,023 randomly selected individuals aged 14 to 64 participated in the study. Participants were asked about their gambling behaviors using validated questionnaires, including whether they gambled online or offline. The survey also collected demographic information, such as age and other socioeconomic characteristics, to support the analysis. Furthermore, problem gambling was measured using DSM-5 criteria, where participants were classified as having problem gambling if they fulfilled at least three of the nine diagnostic criteria. To quantify online activity, the study introduced a relative fraction measure, which calculated the ratio of online gambling days to the total number of online and offline gambling days. Finally, Effertz et al. (2018) used a bottom-up approach to estimate annual medical costs by integrating the survey results with external health insurance claims data. The results showed that online gambling was strongly associated with a greater likelihood of developing problematic gambling. Across all statistical models, increasing the proportion of gambling done online increased the risk of gambling problems. Specifically, every 10% increase in online gambling participation raised the probability of problematic gambling by approximately 8.8% to 12.6%, suggesting that features unique to online gambling may contribute to gambling-related harm (Effertz et al., 2018, p. 967). The analysis also found that gambling more frequently and having a stronger attachment to the internet significantly increased the risk of problem gambling (Effertz et al., 2018, p. 972). Frequency is defined as the total sum of days an individual reported engaging in any form of gambling over the previous 12 months. In contrast, higher levels of education were associated with a lower risk. Other factors linked to a greater likelihood of problematic gambling included being male, unemployed, single or divorced, and having a migration background (Effertz et al., 2018, p. 972). Younger individuals who gambled online were especially vulnerable, and the highest risk was observed among participants with both an extreme level of internet attachment and a migration background. Finally, Effertz et al. (2018) estimated that gambling-related health care costs in Germany totaled approximately €218.4 million per year, with €27.2 million (12.5%) directly attributable to online gambling. They also estimated that about 139,000 of the approximately 1.077 million problematic and pathological gamblers in Germany developed gambling problems because of online gambling. It was estimated by calculating the difference between the average probability of being a problematic gambler with current online activity and the predicted probability if the fraction of online gambling were reduced to zero, then multiplying that difference by the total population frequency for each age group in Germany. This source is useful because it highlights that online gambling has created significant economic costs for the health care system in addition to increasing the generality of gambling-related harm.

Cesarini et al

Cesarini et al., (2016) examines whether sudden increases in wealth, often called wealth shocks, improve the long-term health and development of players and their families. By analyzing nearly half a million prizes awarded to Swedish lottery players, Cesarini et al. (2016) tracked how massive financial windfalls ranging from roughly $20,000 to $800,000 affected families over a 24-year period. This research included participants from the Triss lottery, which is a scratch-ticket format where winners often appear on a morning TV show to draw for lump-sum prizes or monthly installments that can last up to 50 years. It also featured players from the Kombi lottery, which is a monthly subscription service where subscribers are automatically entered into draws while supporting the Swedish Social Democratic Party.

Cesarini et al. (2016) employed a suite of rigorous econometric techniques to ensure their findings were not skewed by a person's pre-existing lifestyle. Their primary approach was a natural experiment, a method widely respected in economics that uses a random event like a lottery win to mimic a controlled laboratory setting. To refine this, they used cell fixed effects and matched controls, where they grouped individuals into specific cells based on their exact probability of winning before the draw occurred. While these methods are common in high-level economics to eliminate omitted variable bias, this study is unique for applying them to a dataset of this magnitude and duration. They also used ordinary least squares (OLS) for most outcomes and an exponential proportional hazard model to analyze how the risk of death changed over time. Finally, they conducted Monte Carlo simulations, which are computational tests used to verify that their statistical results were not just a product of random chance or a small sample size.

The evidence shows that in a country with a strong social safety net, money itself does not magically improve long-term physical health or a child's potential (Cesarini et al., 2016, p. 687). For adults, Cesarini et al. (2016) proved that even a massive win had no significant effect on the risk of dying within ten years, allowing them to rule out any benefit even one-sixth as large as the typical wealth-health gap (Cesarini et al., 2016, pp. 687, 708). The only adult benefit was a small decrease in the use of anti-anxiety and sleeping pills (Cesarini et al., 2016, pp. 687, 690, 717). When it came to children, the results were just as striking; a prize of one million SEK had no impact on a child's cognitive skills or their school grades, with the 95% confidence interval for GPA falling between -0.08 and 0.03 standard deviation units (Cesarini et al., 2016, pp. 687, 723). While they did find that wealth lowered a child's risk of obesity by 2.1 percentage points, it also led to a 3.4 percentage point increase in hospital visits shortly after the win, representing a 19% jump in relative risk (Cesarini et al., 2016, pp. 721, 722). Ultimately, the study suggests that the differences seen between the rich and poor in affluent nations are likely due to complex social factors rather than a simple lack of cash (Cesarini et al., 2016, pp. 690, 733). It shows that the financial gains from winning big do not lead to lasting improvements in a person's health or their family's success. By examining what happens after people win the lottery, the study debunks the idea that gambling is a reliable way to solve money problems, proving instead that chance-based wealth has some very real limits.

An increase in gambling activity is a balance of private consumption gains and public sector costs. For many people who gamble, it works as adult entertainment which brings certain psychological gains, for instance, 9.3% lower chances of having high stress levels in life. At the same time, the spread of gambling activity to virtual platforms results in huge negative externality effects, as a 10% increase in the number of gamblers leads to a 12.6% increase in risks of getting hooked on gambling, which in turn poses a huge burden on the health care system. In the end, the use of gambling as a solution to financial issues is an illusion, because neither huge windfalls nor losses have any impact on a person’s health condition and his or her children’s performance at school.

From an economic perspective, gambling represents a trade-off between the leisure benefits experienced by many participants and the systemic costs created by gambling-related harm. For example, Humphreys et al. (2021) found that 92.55% of active gamblers were classified as non-problem gamblers, suggesting that recreational gambling can function as a form of leisure. Their findings also indicate that recreational gambling reduces the probability of experiencing high levels of life stress by 9.3% (Humphreys et al., 2021). However, the expansion of digital gambling platforms creates significant negative externalities, as Effertz et al. (2018) found that every 10% increase in online gambling participation raises the probability of problematic gambling by 8.8% to 12.6%. These harms extend beyond individuals, as gambling-related health care costs in Germany were estimated at €218.4 million annually (Effertz et al., 2018). Furthermore, the belief that gambling can provide a pathway to financial improvement is challenged by Cesarini et al. (2016), who found that even large lottery winnings did not produce significant long-term improvements in physical health or children's educational outcomes. Therefore, while gambling may provide short-term entertainment benefits, it does not represent a reliable solution to financial or personal difficulties.

Conclusion

All the facts presented in this paper clearly prove that gambling involvement is determined by several social, psychological, and economic determinants rather than individual preferences only. Inequality, unequal exposure to gambling, psychological vulnerability, and economic determinants lead to higher gambling involvement and increase risks of gambling harm. Gambling should stay a legal entertainment for adults; however, the state should pay more attention to these factors rather than concentrate on personal responsibility. To decrease gambling-related harms, policy makers need to develop measures aimed at reducing the impact of factors influencing gambling behavior. The measures should include limiting the presence of electronic gambling machines (EGMs) in disadvantaged communities, limiting advertising of risky gambling products, such as online and sports gambling, enhancing screening and intervention programs, which will involve improvement of Problem Gambling Severity Index (PGSI) and other responsible gambling measures, providing gambling education for children at school level and for the general public, and increasing gambling revenues investment into gambling prevention, treatment, and independent research. The measures mentioned above are addressed to Loto-Québec, the Government of Quebec, and gambling regulatory agencies. All these studies show the necessity to apply public health strategies that minimize exposure to gambling hazards without banning gambling altogether as a legitimate form of adult entertainment.

The present study has investigated the factors affecting participation in gambling activities from the perspectives of sociology, psychology, and economics. Taken together, the results presented above indicate that gambling participation is not simply an outcome of individuals' choices. Sociologically speaking, the participation in gambling activities is affected by various social inequalities and vulnerable environment that provide greater opportunities for gambling. Psychologically, it is established that personal vulnerabilities such as impulsivity, psychiatric disorders, demographic features, and involvement in riskier gambling products increase the chances of developing gambling-related problems. From an economic point of view, it is found that gambling is affected by accessibility and gambling expansion to the Internet market as well as by the fact that gambling cannot ensure people's success or personal development in any way. In summary, the above-presented results confirm the thesis that gambling is the result of the interaction between social, psychological, and economic factors and not of personal responsibility. It should be emphasized that many people do participate in gambling without being affected negatively but still vulnerable population groups suffer most from the gambling environment and products. Accordingly, public policies need to consider the wider factors affecting gambling behavior as well as minimize exposure to dangerous situations where gambling can take place.

References

Beckert, J., & Lutter, M. (2013). Why the Poor Play the Lottery: Sociological Approaches to Explaining Class-based Lottery Play. Sociology, 47(6), 1152–1170. https://doi.org/10.1177/0038038512457854

Breaking the Cycle - Mental Health Commission of Canada. (2025, October 21). Mental Health Commission of Canada. https://mentalhealthcommission.ca/catalyst/breaking-the-cycle/

Cesarini, D., Lindqvist, E., Östling, R., & Wallace, B. (2016). Wealth, Health, and Child Development: Evidence from Administrative Data on Swedish Lottery Players *. The Quarterly Journal of Economics, 131(2), 687–738. https://doi.org/10.1093/qje/qjw001

Effertz, T., Bischof, A., Rumpf, H.-J., Meyer, C., & John, U. (2018). The effect of online gambling on gambling problems and resulting economic health costs in Germany. The European Journal of Health Economics, 19(7), 967-978. http://www.doi.org/stable/45156506

Glimne, D. (2019). gambling | Definition, History, Games, & Facts. In Encyclopædia Britannica. https://www.britannica.com/topic/gambling

Gooding, N. B., & Williams, R. J. (2024). Are there riskier types of gambling?Journal of Gambling Studies, 40(2), 555-569. https://doi.org/10.1007/s10899-023-10231-0

Humphreys, B. R., Nyman, J. A., & Ruseski, J. E. (2021). The Effect of Recreational Gambling on Health and Well-Being. Eastern Economic Journal, 47(1), 29–75. https://www.doi.org/stable/48762761

Ioannidis, K., Hook, R., Wickham, K., Grant, J. E., & Chamberlain, S. R. (2019). Impulsivity in gambling disorder and problem gambling: A meta-analysis. Neuropsychopharmacology, 44(8), 1354–1361. https://doi.org/10.1038/s41386-019-0393-9

Murch, W. S., Scheurich, R., & Monson, E. (2025). Distinguishing persistent versus episodic clusters of at-risk respondents on the Problem Gambling Severity Index. Journal of Gambling Studies, 41, 1071–1085. https://doi.org/10.1007/s10899-025-10386-y

Pabayo, R., Patel, P., Patte, K. A., & Leatherdale, S. T. (2024). Income Inequality and the Odds of Online Gambling Among a Large Sample of Adolescents in Canada.Journal of Gambling Studies (Online), 40(1), 289-305. https://doi.org/10.1007/s10899-023-10194-2

Papineau, E., Robitaille, É., Prisca Samba, C., Lemétayer, F., Kestens, Y. & Raynault, M.-F. (2024). Se donner des outils pour réduire les inégalités d’exposition aux jeux de hasard et d’argent. Drogues, santé et société, 22(2), 58–88. https://doi.org/10.7202/1115694ar

Williams, R. J., Leonard, C. A., Belanger, Y. D., Christensen, D. R., el-Guebaly, N., Hodgins, D. C., McGrath, D. S., Nicoll, F., Smith, G. J., & Stevens, R. M. G. (2021). Predictors of gambling and problem gambling in Canada. Canadian Journal of Public Health / Revue Canadienne de Santé Publique, 112(3), 521–529. https://doi.org/10.17269/s41997-020-00443-x