Journal of Gambling Studies (2023) 39:1189–1205
https://doi.org/10.1007/s10899-022-10166-y
ORIGINAL PAPER

An Empirical Study of the Pathway Model Link Between
Cognitive Distortions and Gambling Problems
Kahlil S. Philander1,2

· Sally M. Gainsbury2

Accepted: 6 November 2022 / Published online: 21 November 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022

Abstract
This research empirically tests the relationship between gambling-related cognitive distortions and the development of gambling problems. In two separate studies using methodologies designed to support non-experimental causal inference, we demonstrate that holding
false beliefs about gambling experiences is related to current and future risk of developing problems with gambling. In our first study, we use an instrumental variable estimation
strategy on an internet sample (n = 184) and observe a statistically significant relationship
between Gamblers’ Belief Questionnaire scores and measures of loss chasing, overspending, and gambling problems. These findings were robust to linear and ordinal estimation
strategies and multiple model specifications. In our second study, we examine five-year
prospective longitudinal data (n = 1,431) to validate our initial findings and test whether
irrational thoughts are also related to future problems with gambling. While controlling for
current fallacies, we find that past Gambling Fallacies Measure scores are related to present
gambling problems across two survey waves. The effect size of each of the past fallacy levels is roughly half of the effect size of present levels, suggesting meaningful impacts. Our
findings support the Pathways Model of Problem and Pathological Gambling.
Keywords Cognitive distortions · Gambling disorders · Gambling fallacies

Introduction
The current understanding of gambling research suggests that individuals with gambling
problems often hold false beliefs about their likelihood of winning or influencing game outcomes (Fortune & Goodie, 2012; Hodgins et al., 2011). Correcting these cognitive distortions is believed to be an important part of treatment programs (Hodgins et al., 2011; Petry
et al., 2017; Yau & Potenza, 2015), but despite the importance of addressing these schema
in treatment, cognitive distortions’ role in contributing to the formation of problems with
* Kahlil S. Philander
Kahlil.philander@wsu.edu

School of Hospitality Business Management, Carson College of Business, Washington State
University, 915 North Broadway, Everett, WA 98201, USA

Brain and Mind Centre, School of Psychology, Faculty of Science, University of Sydney, 94 Mallet
St, Camperdown, NSW 2050, Australia

Vol.:(0123456789)

Journal of Gambling Studies (2023) 39:1189–1205

gambling is unclear. For instance, most individuals hold some distorted beliefs but do not
develop problems with gambling (Welte et al., 2015).
The leading model of problem gambling etiology, the Pathways Model of Problem and
Pathological Gambling (Pathways Model, Blaszczynski & Nower, 2002), proposes that
cognitive distortions precede overconsumption and loss chasing, which subsequently can
lead to problems. However, empirical evidence of this link is primarily retrospective and
correlational since the related habits may take a long period of time to develop. Validation
of this proposed path is an important question to gambling studies, as it has implications
for prevention programs and appropriate responsible gambling practices in product design
and marketing.
This research contributes to the literature by empirically examining the relationship
between gambling-related cognitive distortions and the development of gambling problems. The contribution is methodological and empirical, as we use two estimation strategies that are novel to the existing literature. In our first study, we use an instrumental
variable approach to correct for endogeneity biases in the impact of cognitive distortions
on gambling. This allows us to improve our understanding on the direction of the effect
and potential confounding issues from missing variables. In our second study, we use panel
data from a multiyear longitudinal study of adults to understand whether cognitive distortions are related to gambling problems in future periods.
As gambling grows in social influence, it is important to understand how product and
design choices will contribute to social harms and what steps can be taken to improve consumer protection. If cognitive distortions contribute to the risk of developing a gambling
problem, regulators and other stakeholders should invest in prevention programs that specifically target these false beliefs. In addition, operators can make better product and design
choices that minimize the likelihood that individuals will hold false beliefs about their likelihood of winning or influencing outcomes.

Literature
Cognitive Distortions
Gambling is an activity that spans several large industries including casinos, pokies,
pachinko parlors, video lottery terminals, sports wagering, lottery games, and online gaming. Gambling-related cognitive distortions are a set of loosely related errors in human
decision-making that are connected by their relationship to the gambling consumption
experience and false perceptions about individuals’ ability to predict or control outcomes
that are determined by chance (Blaszczynski & Nower, 2002; Fortune & Goodie, 2012;
Goodie & Fortune, 2013; Hodgins et al., 2011; Toneatto, 1999).
Much of the early research on gamblers’ cognitive distortions derived from ‘thinkaloud’ methods, where gamblers expressed their thoughts while gambling or performing
related tasks (Gaboury & Ladouceur, 1989; Griffiths, 1994; Ladouceur et al., 1988; Sharpe,
2002). Analysis of transcripts from these studies revealed several categories of maladaptive
thoughts, which conflicted with an objective assessment of reality. For example, irrational
verbalizations in Ladouceur et al. (1988) included misattributions about skill while playing
a slot machine, “I’m getting good at this game. I think I’ve mastered it” (p.414).
Gambling related biases include a class of heuristics and other misattributions that
emerge throughout the gambling experience (Fortune & Goodie, 2012), which lead to false

Journal of Gambling Studies (2023) 39:1189–1205

beliefs by the player about the likely outcomes of their play (Hodgins et al., 2011). These
beliefs often do not correct over time or experience, and are related to persistence in gambling (Billieux et al., 2012). Across a range of logic or probability tasks, individuals often
give irrational responses due to a use of heuristics instead of deliberate reasoning processes
(Kahneman et al., 1974). Although the use of heuristics can provide fast choices that do not
burden limited executive resources, they can systemically fail to provide reasonable decisions (De Neys & Goel, 2011; Stanovich, 2000).
Several validated scales of gambling cognitive distortions were developed in the early2000s (Goodie & Fortune, 2013), leading to many studies that demonstrate a correlation
between distorted thinking and gambling problems across multiple populations (Cunningham et al., 2014; Joukhador et al., 2003; Raylu et al., 2016; Xian et al., 2008). Etiological
models of gambling problems, most notably the “Pathway Model”, propose a causal role
of cognitive distortions in the development of gambling disorders (Blaszczynski & Nower,
2002; Hodgins et al., 2011; Sharpe, 2002). Cognitive distortions are described in this literature as false beliefs that lead to wagering, and therein, classical and operant conditioning
effects from the reward and punishment structure of games. This further leads to increased
frequency of play and exposure, and potential development of more pervasive cognitive
schema that shape and distort consumers’ experiences with gambling.
One of the largest categories of distorted thinking by gamblers is related to availability heuristics, which are biases where events that are more easily recalled in memory are
perceived as more likely to occur (Tversky & Kahneman, 1973). These include gamblers
drawing illusory correlations between events, such as salient environmental features or
the impact of superstitious beliefs, talismans, and behaviors (Fortune & Goodie, 2012).
It also includes biases from selective memory, including more prominent recall of own
and others’ wins rather than losses (Griffiths, 1994; Hodgins et al., 2011). There is some
evidence that availability heuristics respond to commercial practices, such as advertisement of recent lottery winners at local stores drawing more customers (Guryan & Kearney,
2008) and making wins more salient than losses through slot machine sounds (Dixon et al.,
2014). These types of effects appear to drive use of superstitious elements in marketing and
design choices.
A second class of errors is related to the representativeness heuristic, where randomly
distributed small sequences of numbers are believed to be representative of the distributions from which they are derived (Tversky & Kahneman, 1971). This set of errors includes
the hot hand fallacy, which is a false belief that past success in randomly determined events
is indicative of future success. This cognitive distortion has most notably been reinforced
in slot machines, where multiline games respond with lights and sounds like a win, despite
a net loss occurring across multiple simultaneous wagers. Those outcomes, known as
‘losses disguised as wins’, are believed to be associated with overestimation of win rates,
and are in machines found to be preferred by problem gamblers (Barton et al., 2017). The
gambler’s fallacy is also an error in representativeness. It refers to a mistaken belief that
if an outcome has occurred more often in the past, it will occur less often in the future,
despite the trials being statistically independent (Croson & Sundali, 2005). Observational
studies confirm that consumers bet consistently with the gambler’s fallacy and hot hand
beliefs (Croson & Sundali, 2005; Xu & Harvey, 2014).
Perhaps the most frequently reinforced cognitive distortion in the gambling industry
is the illusion of control. This bias refers to an erroneous belief by players that they can
influence games of chance through skillful behavior or decisions, or that they have influence over chance elements in games of mixed skill and chance (Ladouceur et al., 1984;
Langer, 1975). Many illusions appear due to core aspects of the games. For example,

Journal of Gambling Studies (2023) 39:1189–1205

craps games require a player to roll dice, and players tend to place higher and more
risky wagers on their own rolls than rolls by other players (Davis et al., 2000). Similarly, sports fans may believe that their knowledge about teams or players will give them
an advantage in placing sports wagers. Illusions of control are also reinforced through
unnecessary game elements or other firm practices. For example, lottery advertisements
often imply gamblers have control over outcomes by picking numbers (Miyazaki et al.,
2001), and electronic gaming machines have features like stop buttons that appear to
change reel outcomes, but have no actual influence on outcomes that were determined
before the button was pressed (Inoue, 1998). A related design feature that has received
attention in both slot machine and instant ticket research is the artificial near-miss,
where losses are purposefully manipulated to appear close to a winning outcome, rather
than a randomly distributed loss appearance (Muir, 2009; Stange, et al., 2017). Near
misses are interpreted by gamblers as progress in mastering the game, and are therefore
viewed as biases in illusion of control (Clark, 2010; Clark et al., 2009; Griffiths, 1991).
A small number of studies have attempted to estimate the relationship between cognitive distortions and gambling problems using longitudinal estimation strategies. Two
studies used generalized estimating equations, finding small but significant effect sizes
(Leonard & Williams, 2016; Schulte et al., 2021) but also noted concerns about reverse
causality in the relationship. One study used logistic regression (Nicholson et al., 2016),
which likely led to dependent observations that are nested within subjects, making inference unreliable.

Hypotheses
Our hypotheses closely follow propositions in the Pathways Model literature (Blaszczynski & Nower, 2002), which suggests that gambling-related cognitive distortions lead to
excessive loss chasing and overspending, and then to subsequent problems with gambling.
A behavioral response to irrational beliefs about winning by gamblers is likely to be an
increase in play volume – gamblers that falsely believe that they can win at gambling may
be more likely to continue to play. This may lead to instances of overspending and attempts
to win back money that results in loss chasing. The emotional, social, and financial consequences of this behavior and increased volume may lead to more pervasive problems with
gambling.

H1a Gambling-related cognitive distortions lead to increases in loss chasing.
H2a Gambling-related cognitive distortions lead to wagering more than an individual can
subjectively afford to lose.
H3a Gambling-related cognitive distortions lead to current gambling problem severity.
Cognitive distortions may be deeply held, and we may expect their effects to perpetuate
in the future. Since addictions may develop over time as a result of conditioning effects
from gambling or ongoing personal and social consequences (Blaszczynski & Nower,
2002), we expect that cognitive distortions will also be indicative of future risks.

H1b Gambling-related cognitive distortions lead to increases in future loss chasing.

Journal of Gambling Studies (2023) 39:1189–1205

H2b Gambling-related cognitive distortions lead to wagering more in the future than an
individual can subjectively afford to lose.
H3b Gambling-related cognitive distortions lead to future gambling problem severity.

Study 1
Cognitive distortions in gambling are formed and reinforced over time. The usefulness of
experimental design to evaluate impacts on behavior and gambling problems is therefore
constrained. To test H1a, H2a, and H3a, we employ an instrumental variable method in
Study 1 that allows us to control for arbitrary forms of endogeneity that would appear in
ordinary linear regressions, such as reverse causality or omitted variable bias.

Methods
Due to the low prevalence of persons with gambling problems in the general population,
which is estimated at less than 1% of adults (Kessler et al., 2008), we recruited gamblers
through the Amazon Mechanical Turk (MTurk) panel. Online panels are noted to have over
four times the incidence of persons with gambling disorders than random-digit dialed samples (Lee et al., 2015). Most workers are jointly motivated by financial compensation and
enjoyment of the tasks (Paolacci & Chandler, 2014).
Participants in this study were restricted to those with an MTurk approval rating of at
least 95 percent (Goodman et al., 2013; Peer et al., 2014). To further reduce social desirability bias, respondents were told their data would be anonymous and the specific nature
of the research question was concealed (Paolacci & Chandler, 2014). Participants were
English speaking North Americans of the legal gambling age that had lived in or visited
the jurisdictions that contain all of the games discussed in the study (Nevada, New Jersey, Connecticut, and California) in the past 12 months. A total of 232 respondents were
recruited, and 48 responses were removed due to failing at least one of two attention checks
or not completing the survey. The sample was more male (67.93%) with a mean age of
34.0 (SD = 9.3). Study procedures were approved by the [redacted] Human Research Ethics
Committee.
We estimate linear instrumental variable models using questions of loss chasing, overspending, and overall problem gambling severity index scores as the dependent variable,
and we estimate ordered probit models using problem gambling severity index categories
as the dependent variable. Instrumental variable methods can improve the consistency
of empirical models when there are endogenous effects as they simulate random assignment when instruments are sufficient strong (Angrist & Imbens, 1995; Angrist et al., 1996;
Baiocchi et al., 2014; Rassen et al., 2009). These corrections are important to this model,
as it seems likely that estimates will be biased by omitted variables (e.g. genetic predisposition) or reverse causality. Instrumental variable models are often used in social sciences
to infer causality when there are unmeasured confounding variables (Lousdal, 2018). We
estimate the models using Stata 15.1 and its extended regression models package (StataCorp LLC, 2017).
The study was approved by the [redacted] Human Research Ethics Committee and all
participants provided informed consent. Replication data will be made available by the
authors for institutional review board approved studies.

Journal of Gambling Studies (2023) 39:1189–1205

Measures
The Gamblers Belief QuestionnaireWe measure distortions using the Gamblers Belief
Questionnaire (GBQ), a validated and self-administered psychometric measure of gambling cognitive distortions (Steenbergh et al., 2002; Winfree et al., 2015). The GBQ
was originally validated through five studies described in Steenbergh et al. (2002), and
further studies have found it to be a valid scale (Marchetti et al., 2016; Philander et al.,
2019; Winfree et al., 2013; Wong & Tsang, 2012). The scale includes 20-items measured on a 7-point scale (1 = strongly agree, 7 = strongly disagree), which are reverse
coded and summed (m = 71.12, sd = 9.48). Higher scores indicate higher levels of cognitive distortions.
The Problem Gambling Severity Index Gambling disorders are measured using the
problem gambling severity index (Currie et al., 2013; Ferris & Wynne, 2001). The index
is a self-reported epidemiological measure of gambling addiction, designed to measure
harmful consequences of gambling behavior in the general population. The index has
been validated across multiple populations (Currie et al., 2013; Holtgraves, 2009; Jackson et al., 2010; Miller et al., 2013; Orford et al., 2010; Sharp et al., 2012) and in multiple languages (Loo et al., 2011; Lopez-Gonzalez et al., 2018; Marchetti et al., 2016; So
et al., 2019).
Two of the nine items from the index are used to test H1a and H2a, respectively: (1)
[thinking about the last 12 months] “When you gambled, did you go back another day to
try to win back the money you lost?” (Loss Chasing); and (2) “Have you bet more than
you could really afford to lose?” (Overspending). Item (1) originally appeared in the
fourth edition of the American Psychiatric Association’s diagnostic manual (American
Psychiatric Association, 2012). Item (2) originally appeared in the South Oaks Gambling Screen (Lesieur & Blume, 1987). Both screens were developed for use with clinical populations. All items responses occur on a four-point Likert scale: 0, “Never”; 1,
“Sometimes”; 2, “Most of the time”; or 3, “Almost always”. Item scores are summed
and total scores range from 0 to 27, with four interpretative categories (Currie et al.,
2013): 0, non-gambler/non-problem gambler (n = 83, 45.11%); 1–4, low-risk gambler
(n = 48, 26.09%); 5–7, moderate-risk gambler (n = 13, 7.07%); and > 7, problem gambler
(n = 40, 21.74%).
Skill and chance variables (instruments) Respondents were asked to order a series of
gambling games (e.g. slot machines) and non-gambling games (e.g. chess) along a 100point scale from all skill (0) to all chance (100), which we exploit to form our instrumental variables. The instrumental variables are the difference in an individual’s perceived slot machine skill/chance (m = 89.93, sd = 14.65) from its mean score across all
participants (perceived slot skill), and the difference in an individual’s perceived poker
skill/chance (m = 19.77, sd = 30.67) from its mean score across all participants (perceived poker skill). Intuitively, our instrument exploits subjective differences in a game
that objectively has no skill component, slot machines, from a game that objectively
has a skill component, poker. We propose that individual’s experiences playing, observing, or otherwise learning about particular games are likely to influence their potentially
distorted thoughts on skill and chance elements of that game, and therein impact their
overall cognitive distortions in gambling as measured by the GBQ.
Slot machines are games of pure chance, and we expect respondents with stronger
cognitive distortions to overvalue the role of skill in slot machines relative to other
respondents. Poker is a game of mixed skill and chance, and with the slot machine

Journal of Gambling Studies (2023) 39:1189–1205

instrument capturing illusion of control effects, we expect the poker variable will capture further cognitive distortions related to an undervaluation of the role of skill in
poker, either through attributing negative outcomes to luck (i.e. attribution bias) or
viewing it as similar to other card-based table games (i.e. representativeness bias). We
test the power of our instruments as part of our empirical analysis, and control for frequency of play of slot machines and poker.
Other variables Respondents are asked about frequency of play in various games over
the past 12-months on a six-point Likert scale ranging from “Not at all in the past year” to
“Daily”. Summary statistics of these variables are provided in Table 1. Respondents are
asked several demographic questions, including age, gender identity, race, income level,
and education level. Respondents were mostly ‘White’ (n = 105, 57.07%), followed by
‘Asian’ (n = 34, 18.48%), ‘American Indian or Alaskan Native’ (n = 16, 8.70%), ‘Hispanic,
Latino, or Spanish origin’ (n = 15, 8.15%), ‘Black or African American’ (n = 13, 7.07%),
and ‘Other’ (n = 1, 0.54%). In terms of highest level of education, 26 individuals reported a
Master’s degree (14.13%); 91 reported a Bachelor’s degree (49.46%); 36 reported an Associate’s degree (19.57%); and 31 reported a high school diploma or equivalent (16.85%).
The median income category was $50,000-$70,000 per year, with three respondents not
disclosing their income range.

Results
To test if the specific behavioral items related to H1a and H2a are implicated in the relationship with GBQ, we produce an ordered probit model with endogenous regressors for
both dependent variables in Table 2. All model estimates imply a positive relationship
between loss chasing and GBQ scores, supporting H1a. Similarly, we estimate a positive
relationship between reported over-spending and GBQ, supporting H2a. After controlling
for frequency of game play, gender, age, race, income, and education, and other endogenous effects, higher rated gambling cognitive distortions score are associated with higher
likelihood to reply “Sometimes” or more, and a lower likelihood of reporting “Never”, as
shown in the estimated marginal effects.
We further test the validity of our instruments in our linear two-stage least squares model
estimating problem gambling severity index scores. We note that all models in this study have
the same first-stage equation variables. The fully specified linear two-stage regression model
(Table 3, Model 6) is statistically significant (adjusted R2 = 0.63, χ2 (22, n = 184) = 619.97,
p < 0.001). The perceived slot skill variable appears to be a good instrument (β = −0.596,
p < 0.001), while the perceived poker skill variable is not significant (β = 0.123, p = 0.071). We
retain perceived poker skill as an instrument, as it is significant in the similar ordered regression model appearing in Table 4 (β = 0.330, p < 0.001), which is functional form that may better align with the distribution of the scale (Currie et al., 2013). Jointly, the instruments perform

Table1  Summary statistics of
frequency variables

Variable

N

Mean

Std. Dev

Min

Max

Slot Machine Frequency

2.435

1.333

Poker Frequency
Video Game Frequency
Mobile Game Frequency

2.120
4.152
4.304

1.350
1.656
1.658

− 0.959***
(0.138)
0.352***
(0.079)
− 0.005***
0.002***
0.002**
0.001*

− 0.963***
(0.136)
0.349***
(0.079)

− 0.008***
0.003***
0.003***
0.002**

Robust standard errors in parentheses. *p < 0.05, **p < 0.01, ***p < 0.001

Yes
Yes
Yes
Yes
Yes

(0.007)
0.445***
(0.107)
0.0717
(0.102)
Yes
Yes
Yes
Yes
Yes

(0.006)

0.036***

0.047***

(0.005)

DV: Loss Chasing

(3)

0.043***

DV: Loss Chasing

DV: Loss Chasing

Gender
No
Age group
No
Race
No
Income
No
Education
No
First stage instrument estimates (DV: GBQ)
Perc. slot skill
− 0.954***
(0.143)
Perc. poker skill
0.357***
(0.079)
Average marginal effects of GBQ
Never
− 0.009***
Sometimes
0.003***
Most of the time
0.004***
Almost always
0.003**

Poker play freq

Slot play freq

GBQ

(2)

(1)

Table 2  Ordered probit with endogenous regressor – Loss chasing and overspending

− 0.008***
0.003***
0.003***
0.002*

− 0.950***
(0.141)
0.360***
(0.077)

No
No
No
No
No

(0.006)

0.056***

DV: Over-spending

(4)

− 0.007***
0.003***
0.002*
0.001**

− 0.955***
(0.135)
0.356***
(0.076)

Yes
Yes
Yes
Yes
Yes

(0.006)

0.056***

DV: Over-spending

(5)

− 0.004***
0.002***
0.001**
0.001**

− 0.951***
(0.137)
0.359***
(0.076)

(0.006)
0.344**
(0.113)
0.107
(0.101)
Yes
Yes
Yes
Yes
Yes

0.046***

DV: Over-spending

(6)

Journal of Gambling Studies (2023) 39:1189–1205

Journal of Gambling Studies (2023) 39:1189–1205
Table 3  Linear regression–Problem gambling severity index score

GBQ
Slot play freq

(1)

(2)

(3)

(4)

(5)

(6)

OLS

OLS

OLS

2SLS

2SLS

2SLS

0.154***

0.114***

0.045***

0.221***

0.187***

0.161**

(0.013)

(0.014)

(0.015)
1.755***
(0.415)
0.874*
(0.402)
− 3.012
(1.861)
Yes
Yes
Yes
Yes
Yes

(0.021)

(0.031)

-11.43***
(1.479)
No
No
No
No
No

− 6.817
(3.640)
Yes
Yes
Yes
Yes
Yes

(0.050)
0.701
(0.579)
0.142
(0.529)
−  7.681**
(2.824)
Yes
Yes
Yes
Yes
Yes

0.703

–0.954***
(0.145)
0.357***
(0.079)
0.412

–0.892***
(0.146)
0.227**
(0.078)
0.515

− 0.596***
(0.148)
0.123
(0.068)
0.578

Poker play freq
Constant

− 6.660***
1.025
(0.772)
(2.246)
Gender
No
Yes
Age group
No
Yes
Race
No
Yes
Income
No
Yes
Education
No
Yes
First stage instrument estimates (DV: GBQ)
Perc. slot skill
Perc. poker skill
Observations
Adjusted R 2

0.510

0.601

Robust standard errors in parentheses. *p < 0.05, **p < 0.01, ***p < 0.001

well. The F-test of weak identification = 15.740, exceeding the benchmark value of 10 (Sovey
& Green, 2011; Staiger & Stock, 1997) and the estimated 15% maximal size of 11.59 (Stock
& Yogo, 2005). The Sargan test statistic 𝜒 2 = 0.202, p = 0.653, failed to reject the assumption
that perceived skill variables do not belong in the structural equation predicting problem gambling severity index scores (Sargan, 1958), supporting the other necessary condition. In all the
linear models, higher GBQ scores are associated with higher problem gambling severity index
scores, supporting H4.
In our ordinal models estimating problem gambling severity index groups, GBQ is statistically significant and the correct sign, supporting H3a. As shown in Table 4, across all marginal effect estimates, higher cognitive distortions are associated with a lower probability of
belonging to the non-problem gambler category and a higher probability of belonging to one
of the problem gambling risk categories. The endogenous ordered probit model instruments
are all statistically significant, p < 0.001, again suggesting that the instruments are strong.

Study 2
Since the impacts of H1a, H2a, and H3a are likely to reinforce over time, we conduct
analysis on a longitudinal data set in Study 2, to further validate the cross-sectional findings from Study 1 and test H1b, H2b, AND H3b. A longitudinal data set allows us to use

Journal of Gambling Studies (2023) 39:1189–1205

Table 4  Ordered regression—Problem gambling severity index group

GBQ
Slot play freq

(1)

(2)

(3)

(4)

(5)

(6)

OLogit

OLogit

OLogit

EOProbit

EOProbit

EOProbit

0.071***

0.067***

0.044***

0.048***

0.049***

0.039***

(0.009)

(0.010)

(0.011)
0.886***
(0.234)
0.399
(0.214)
Yes
Yes
Yes
Yes
Yes

(0.004)

(0.004)

No
No
No
No
No

Yes
Yes
Yes
Yes
Yes

(0.005)
0.369***
(0.109)
0.182
(0.093)
Yes
Yes
Yes
Yes
Yes

− 0.960***
(0.143)
0.352***
(0.079)

− 0.986***
(0.131)
0.328***
(0.076)

− 0.984***
(0.134)
0.330***
(0.076)

− 0.008***
0.001***
0.001***
0.006***

− 0.006***
0.002***
0.001**
0.003***

− 0.003**
0.001***
0.001*
0.001*

Poker play freq
Gender
No
Yes
Age group
No
Yes
Race
No
Yes
Income
No
Yes
Education
No
Yes
First stage instrument estimates (DV: GBQ)
Perc. slot skill
Perc. poker skill
Average marginal effects of GBQ
Non-problem
–0.010***
Low-risk
0.001***
Moderate risk
0.001***
Problem
0.008***

–0.008***
0.002***
0.001**
0.005***

–0.005***
0.002***
0.001**
0.002***

Robust standard errors in parentheses. *p < 0.05, **p < 0.01, ***p < 0.001

methods to detect whether cognitive distortions have an impact on future gambling problems, across a multiyear period. We use secondary data from the Leisure, Lifestyle, & Lifecycle Project (LLLP, El-Guebaly et al., 2015), a high quality panel of gamblers and nongamblers. This study was found to be exempt from review by [redacted].

Methods
The LLLP was a five-year prospective longitudinal study on the factors influencing change
in gambling and problem gambling. A sample of 1808 participants from the province of
Alberta were recruited, though we limit our focus to the adult sample (n = 1,372), where
the retention rate was 76.2%. Six cases were removed that had values that were incompatible with the measures used. There were four data collection cycles in the study. Median
time between cycles was 567 days, 649 days, and then 518 days. Further information about
the sample collection procedure is available in Dataverse materials (El-Guebaly et al.,
2015).
We estimate random and fixed-effect regression models, which enables us to examine
the effects of cognitive distortions on responses to future gambling measures. Fixed-effect
models control for static individual traits, and allows us to test for changes that occur over
time, rather that cross-sectionally (Wooldridge, 2010), whereas random effects models

Journal of Gambling Studies (2023) 39:1189–1205

allow for individual heterogeneity that is constant over time and uncorrelated with the error
term. We test for the appropriate model using the Hausman test (Hausman, 1978).

Measures
The Problem Gambling Severity Index We use the same variables as in Study 1, including
items on Loss Chasing and Overspending.
Gambling Fallacies MeasureCognitive distortions are measured using the Gambling
Fallacies Measure, a 10-item summative scale assessing several gambling-related fallacies, including the illusion of control and the gambler’s fallacy (El-Guebaly et al., 2015).
Respondents receive one point for every correct response they provide to a series of questions. The scale ranges from 0–10 with lower values indicative of greater fallacies.

Results
After estimating random and fixed-effect models, the Hausman tests suggest that random
effects models are consistent and the most efficient relative to the fixed effect models using
the problem gambling severity index, χ2(3) = 1.22, p = 0.75; loss chasing, χ2(3) = 1.73,
p = 0.63; and overspending variables, χ2(3) = 1.28, p = 0.73. The regression output for both
fixed and random effects models appear in Table 5. The random effects models reinforce
our findings from Study 1 regarding H1a, H2a, and H3a and suggest that greater cognitive
distortions are related to greater contemporaneous gambling problems, loss chasing, and
overspending.
We also observe lagged effects of cognitive distortions on two outcome variables. Even
when controlling for current fallacies, we find that past fallacies are related to present gambling problems across two survey waves. The effect size of each of the past fallacy levels
is roughly half of the current fallacy levels, suggesting a meaningful effect and supporting
H3b. In addition, we also observe one period lagged effect on overspending in the more
efficient random effects model, supporting H2b. We find no evidence supporting H1b.

General Discussion
Our findings suggests that cognitive distortions may influence current and future gambling
behavior. We find that the presence of gambling-related cognitive distortions is indicative
of current loss-chasing and overspending by gamblers, even after controlling for endogenous factors such as omitted variable bias or reverse causality with an instrumental variable model. A similar relationship appears between cognitive distortions and gambling
problems. Although prior research has shown a strong correlation between these items,
our findings provide clearer evidence on the direction of relationship due to the ability of
our empirical model to control for endogenous effects. These findings complement neuropsychological research in gambling disorders, where there is growing evidence that these
processes are physiological and extend to other addictive substances and behaviors (Clark,
2014; Naqvi et al., 2014).
We further tested our findings in Study 2 using longitudinal data and a different cognitive distortion scale. We found evidence that past cognitive distortions are related to future
gambling problems and overspending. Notably, these lagged effects are linked multiple
years in the future, which is consistent with the Pathways Model that proposes a multistep

− 0.143**

(0.048)
− 0.071
(0.0466)
− 0.080*
(0.041)
2.969***
(0.719)
1,431
.04

− 0.149***

(0.036)
− 0.082**
(0.032)
− 0.075*
(0.030)
3.048***
(0.356)
1,431
.04

DV: PGSI Score

DV: PGSI Score
(0.010)
− 0.003
(0.009)
− 0.014
(0.008)
0.509***
(0.075)
1,431
.02

− 0.032***

DV: Loss Chasing

(3)
Random Effects

(0.015)
− 0.008
(0.015)
− 0.024
(0.0126)
0.653**
(0.223)
1,431
.02

− 0.037*

DV: Loss Chasing

(4)
Fixed
Effects

Standard errors in parentheses. PGSI : Problem gambling severity index. *p < 0.05, **p < 0.01, ***p < 0.001

Observations
Adjusted R2

Constant

Fallacies (2-period lag)

Fallacies (1-period lag)

Fallacies

(2)
Fixed
Effects

(1)
Random Effects

Table 5  Panel data regression models

(0.009)
− 0.015*
(0.008)
− 0.009
(0.007)
0.563***
(0.073)
1,431
.03

− 0.035***

DV: Over-spending

(5) Random Effects

(0.013)
− 0.020
(0.013)
− 0.017
(0.011)
0.665***
(0.195)
1,431
.03

− 0.037**

DV: Over-spending

(6)
Fixed
Effects

Journal of Gambling Studies (2023) 39:1189–1205

Journal of Gambling Studies (2023) 39:1189–1205

journey from conditioning effects of the gambling product to addictive behaviors. We did
not find evidence that loss chasing was indicative of future problems, which may suggest
that it’s a less consequential variable to developing long-term issues. It does appear likely
that there is a pathway that connects cognitive distortions to long-term problems with gambling products, but the relatively weak effects on overspending and loss-chasing in future
periods suggests that the mechanism may occur relatively quickly before it either resolves
or perpetuates as gambling problems. These results are consistent with the Pathways Model
and provide some direction in terms of the time period to consider for conditioning effects.

Implications for Policy and Practice
Many aspects of the consumer experience intrinsically reinforce maladaptive thoughts,
which leads to irrational consumption decisions. Given our findings, operators in gambling industries should consider how product design and marketing communications may
unnecessarily contribute to these errors in cognition. Several obvious candidates for change
include the removal of artificial near misses, losses disguised as wins, and artificial game
controls.
Interventions to educate consumers and the public of common distortions can be used
to enable resistance to the development of erroneous cognitive schemata (Shaffer et al.,
2016). If individuals can effectively rationalize likely consequences of consumption, they
may be able to resist conditioning effects and consider other tactics more aligned with their
long-term goals or wellness, such as leaving debit cards at home during casino visits. This
may be particularly important for youth and young adults who do not have firmly established beliefs.
Given the evidence of widespread commercial practices that encourage distortions, substantial regulatory intervention may be appropriate. This could include restriction of advertisements or promotions that portray or encourage myths, and restriction of product design
features. A reasonable starting point for regulation would be to minimize the presence of
features that contribute to cognitive distortions but add little novelty to the gameplay experience for most players.

Limitations and Future Research
Due to the length of time required to develop cognitive distortions and problems with gambling, this study used instrumental and longitudinal data methods to test our hypotheses,
rather than experimental data. Accordingly, typical caution should be used in interpreting
findings that may be subject to model misspecification or related errors. Notably, the estimates in Study 1 appeared stronger and more robust than in Study 2. This may be a function of the estimation strategy, time frame analyzed, or the sensitivity of the instrument
measuring gambling distortions. In addition, Study 1 relied on an internet sample recruit
through Mechanical Turk. Although we attempted to mitigate bias where possible, it may
be the case that these results are non-representative, are impacted by selection bias, and/or
are impacted by professional respondent bias. Future work in other populations would help
support our findings, ideally through natural experiments.
Funding Funding bodies have had no opportunity to influence the collection, analysis or interpretation of
data.

Journal of Gambling Studies (2023) 39:1189–1205

Declarations
Conflict of interest Kahlil Philander has received research funds from the Washington State Gaming Commission, Manitoba Gambling Research Program, UNLV International Centre for Gaming Regulation, U.S.Japan Business Council, Wynn Resorts, Board of Regents of the Nevada System of Higher Education. He has
received consulting payments from British Columbia Lottery Corporation, Eilers & Krejcik Gaming, Victor Strategies, Responsible Gambling Council of Canada, the Commonwealth of The Bahamas, Hard Rock
Seminole, and iDevelopment and Economic Association. He has received reimbursement for travel from
the Northeast Asia Economic Association Forum, National Council for Problem Gambling, International
Association of Gaming Advisors, National Centre for Responsible Gambling, North American State and
Provincial Lottery Association, Evergreen Council on Problem Gambling, Global Gaming Expo Asia, and
Alberta Gambling Research Institute. Dr. Gainsbury has worked on projects that have been received funding
and in-kind support through her institution from Australian Research Council, NSW Office of Responsible
Gambling, Svenska Spel Research Council, Responsible Wagering Australia, Australian Communication and
Media Authority, Commonwealth Bank of Australia, GameCo, ClubsNSW, Crown Resorts, Wymac Gaming.
Dr. Gainsbury is currently a member (2020–21) of the National Council on Problem Gambling International
Advisory Board (Singapore) and receives an honorarium for this role. She is a member of the Steering Committee for Remote Gambling Research and the Independent Research Oversight Panel both run by GambleAware, which provide an honorarium for reviewer reports. Dr. Gainsbury has received honorarium and travel
costs for research, presentations and advisory services from Credit Suisse, Oxford University, ClubsNSW,
Clubs4Fun, Centrecare WA, Gambling Research Exchange Ontario, Crown, Department of Social Services,
Community Clubs Victoria, Financial and Consumer Rights Council Victoria, Australian Communications
and Media Authority, Gamble Aware, Ministry of Health, Clayton Utz, Greenslade, Generation Next, KPMG,
Stibbe, QBE, Stiftelsen Nordiska Sällskapet för Upplysning om Spelberoende, Japan Medical Society for Behavioural Addiction, National Council for Problem Gambling Singapore. All professional dealings have been
conducted with the aim of understanding gambling, enhancing responsible gambling and harm minimisation
policies and practices as well as treatment interventions.
Ethical Approval All research has had human research committee ethics approval.

References
American Psychiatric Association. (2012). Diagnostic and statistical manual of mental disorder fouth edition (dsm-iv). in american psychiatric organization. https://​doi.​org/​10.​1073/​pnas.​07039​93104
Angrist, J. D., & Imbens, G. W. (1995). Two-stage least squares estimation of average causal effects in models with variable treatment intensity. Journal of the American Statistical Association. https://​doi.​org/​
10.​1080/​01621​459.​1995.​10476​535
Angrist, J. D., Imbens, G. W., & Rubin, D. B. (1996). Identification of causal effects using instrumental
variables. Journal of the American Statistical Association. https://​doi.​org/​10.​1080/​01621​459.​1996.​
10476​902
Baiocchi, M., Cheng, J., & Small, D. S. (2014). Instrumental variable methods for causal inference. Statistics in Medicine. https://​doi.​org/​10.​1002/​sim.​6128
Barton, K. R., Yazdani, Y., Ayer, N., Kalvapalle, S., Brown, S., Stapleton, J., Brown, D. G., & Harrigan, K.
A. (2017). The effect of losses disguised as wins and near misses in electronic gaming machines: a systematic review. In Journal of Gambling Studies. https://​doi.​org/​10.​1007/​s10899-​017-​9688-0
Billieux, J., Van der Linden, M., Khazaal, Y., Zullino, D., & Clark, L. (2012). Trait gambling cognitions
predict near-miss experiences and persistence in laboratory slot machine gambling. British Journal of
Psychology. https://​doi.​org/​10.​1111/j.​2044-​8295.​2011.​02083.x
Blaszczynski, A., & Nower, L. (2002). A pathways model of problem and pathological gambling. In Addiction. https://​doi.​org/​10.​1046/j.​1360-​0443.​2002.​00015.x
Clark, L. (2010). Decision-making during gambling: An integration of cognitive and psychobiological
approaches. In Philosophical Transactions of the Royal Society B. https://​doi.​org/​10.​1098/​rstb.​2009.​
Clark, L. (2014). Disordered gambling: the evolving concept of behavioral addiction. Annals of the New
York Academy of Sciences. https://​doi.​org/​10.​1111/​nyas.​12558
Clark, L., Lawrence, A. J., Astley-Jones, F., & Gray, N. (2009). Gambling near-misses enhance motivation
to gamble and recruit win-related brain circuitry. Neuron. https://​doi.​org/​10.​1016/j.​neuron.​2008.​12.​031

Journal of Gambling Studies (2023) 39:1189–1205

Croson, R., & Sundali, J. (2005). The gambler’s fallacy and the hot hand: empirical data from casinos. Journal of Risk and Uncertainty. https://​doi.​org/​10.​1007/​s11166-​005-​1153-2
Cunningham, J. A., Hodgins, D. C., & Toneatto, T. (2014). Relating severity of gambling to cognitive distortions in a representative sample of problem gamblers. Journal of Gambling Issues. https://​doi.​org/​
10.​4309/​jgi.​2014.​29.2
Currie, S. R., Hodgins, D. C., & Casey, D. M. (2013). Validity of the problem gambling severity index interpretive categories. Journal of Gambling Studies. https://​doi.​org/​10.​1007/​s10899-​012-​9300-6
Davis, D., Sundahl, I., & Lesbo, M. (2000). Illusory personal control as a determinant of bet size and type in
casino craps games. Journal of Applied Social Psychology. https://​doi.​org/​10.​1111/j.​1559-​1816.​2000.​
tb025​18.x
De Neys, W., & Goel, V. (2011). Heuristics and biases in the brain: Dual neural pathways for decision making. In Neuroscience of Decision Making. https://​doi.​org/​10.​4324/​97802​03835​920
Dixon, M. J., Harrigan, K. A., Santesso, D. L., Graydon, C., Fugelsang, J. A., & Collins, K. (2014). The
impact of sound in modern multiline video slot machine play. Journal of Gambling Studies. https://​doi.​
org/​10.​1007/​s10899-​013-​9391-8
El-Guebaly, N., Casey, D. M., Currie, S. R., Hodgins, D. C., Schopflocher, D. P., Smith, G. J., & Williams,
R. J. (2015). The Leisure, Lifestyle, & Lifecycle Project (LLLP): A Longitudinal Study of Gambling
in Alberta. Final Report for the Alberta Gambling Research Institute. https://​doi.​org/​10.​11575/​PRISM/​
Ferris, J., & Wynne, H. (2001). The Canadian Problem Gambling Index: User Manual. Ottawa, ON: Canadian Centre on Substance Abuse. https://​doi.​org/​10.​1007/​s10899-​010-​9224-y
Fortune, E. E., & Goodie, A. S. (2012). Cognitive distortions as a component and treatment focus of pathological gambling: A review. In Psychology of Addictive Behaviors. https://​doi.​org/​10.​1037/​a0026​422
Gaboury, A., & Ladouceur, R. (1989). Erroneous perceptions and gambling. Journal of Social Behavior and
Personality. https://​doi.​org/​10.​1007/​BF010​14651
Goodie, A. S., & Fortune, E. E. (2013). Measuring cognitive distortions in pathological gambling: review
and meta-analyses. Psychology of Addictive Behaviors, 27(3), 730–743. https://​doi.​org/​10.​1037/​a0031​
Goodman, J. K., Cryder, C. E., & Cheema, A. (2013). Data collection in a flat world: the strengths and
weaknesses of mechanical turk samples. Journal of Behavioral Decision Making. https://​doi.​org/​10.​
1002/​bdm.​1753
Griffiths, M. D. (1991). Psychobiology of the near-miss in fruit machine gambling. Journal of Psychology:
Interdisciplinary and Applied. https://​doi.​org/​10.​1080/​00223​980.​1991.​10543​298
Griffiths, M. D. (1994). The role of cognitive bias and skill in fruit machine gambling. British Journal of
Psychology. https://​doi.​org/​10.​1111/j.​2044-​8295.​1994.​tb025​29.x
Guryan, J., & Kearney, M. S. (2008). Gambling at lucky stores: empirical evidence from state lottery sales.
The American Economic Review, 98(1), 485–473.
Hausman, J. A. (1978). Specification Tests in Econometrics. Econometrica, 46(6), 1251. https://​doi.​org/​10.​
2307/​19138​27
Hodgins, D. C., Stea, J. N., & Grant, J. E. (2011). Gambling disorders. The Lancet, 378(9806), 1874–1884.
https://​doi.​org/​10.​1016/​S0140-​6736(10)​62185-X
Holtgraves, T. (2009). Evaluating the problem gambling severity index. Journal of Gambling Studies.
https://​doi.​org/​10.​1007/​s10899-​008-​9107-7
Inoue, H. (1998). Slot machine having two distinct sets of reels. Google Patents.
Jackson, A. C., Wynne, H., Dowling, N. A., Tomnay, J. E., & Thomas, S. A. (2010). Using the CPGI to
Determine Problem Gambling Prevalence in Australia: Measurement Issues. International Journal of
Mental Health and Addiction. https://​doi.​org/​10.​1007/​s11469-​009-​9238-9
Joukhador, J., MacCallum, F., & Blaszczynski, A. (2003). Differences in Cognitive Distortions Between
Problem and Social Gamblers. Psychological Reports, 92, 1203–1214. https://​doi.​org/​10.​2466/​pr0.​
2003.​92.​3c.​1203
Kahneman, D., Tversky, A., & Slovic, P. (1974). Judgement Under Uncertainty: Heuristics and Biases Summary. Science. https://​doi.​org/​10.​1126/​scien​ce.​185.​4157.​1124
Kessler, R. C., Hwang, I., LaBrie, R., Petukhova, M., Sampson, N. A., Winters, K. C., & Shaffer, H. J. (2008).
DSM-IV pathological gambling in the National Comorbidity Survey Replication. Psychological Medicine,
38(9), 1351–1360. https://​doi.​org/​10.​1017/​S0033​29170​80029​00
Ladouceur, R., Gaboury, A., Dumont, M., & Rochette, P. (1988). Gambling: Relationship between the frequency of wins and irrational thinking. Journal of Psychology: Interdisciplinary and Applied. https://​doi.​
org/​10.​1080/​00223​980.​1988.​99155​27

Journal of Gambling Studies (2023) 39:1189–1205

Ladouceur, R., Mayrand, M., Dussault, R., Letarte, A., & Tremblay, J. (1984). Illusion of control: Effects of
participation and involvement. Journal of Psychology: Interdisciplinary and Applied. https://​doi.​org/​10.​
1080/​00223​980.​1984.​99236​57
Langer, E. J. (1975). The illusion of control. Journal of Personality and Social Psychology. https://​doi.​org/​10.​
1037//​0022-​3514.​32.2.​311
Lee, C. K., Back, K. J., Williams, R. J., & Ahn, S. S. (2015). Comparison of telephone RDD and online panel
survey modes on CPGI scores and co-morbidities. International Gambling Studies. https://​doi.​org/​10.​
1080/​14459​795.​2015.​10683​53
Lesieur, H. R., & Blume, S. B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the
identification of Pathological gamblers. American Journal of Psychiatry. https://​doi.​org/​10.​1176/​ajp.​
144.9.​1184
Leonard, C. A., & Williams, R. J. (2016). The relationship between gambling fallacies and problem gambling. Psychology of Addictive Behaviors, 30(6), 694–704. https://​doi.​org/​10.​1037/​adb00​00189
Loo, J. M. Y., Oei, T. P. S., & Raylu, N. (2011). Psychometric Evaluation of the Problem Gambling Severity Index-Chinese Version (PGSI-C). Journal of Gambling Studies. https://​doi.​org/​10.​1007/​
s10899-​010-​9221-1
Lopez-Gonzalez, H., Estévez, A., & Griffiths, M. D. (2018). Spanish validation of the Problem Gambling Severity Index: A confirmatory factor analysis with sports bettors. Journal of Behavioral Addictions. https://​doi.​
org/​10.​1556/​2006.7.​2018.​84
Marchetti, D., Whelan, J. P., Verrocchio, M. C., Ginley, M. K., Fulcheri, M., Relyea, G. E., & Meyers, A. W.
(2016). Psychometric evaluation of the Italian translation of the Gamblers’ Beliefs Questionnaire. International Gambling Studies, 16(1), 17–30.
Miller, N. V., Currie, S. R., Hodgins, D. C., & Casey, D. (2013). Validation of the problem gambling severity
index using confirmatory factor analysis and rasch modelling. International Journal of Methods in Psychiatric Research. https://​doi.​org/​10.​1002/​mpr.​1392
Miyazaki, A. D., Brumbaugh, A. M., & Sprott, D. E. (2001). Promoting and Countering Consumer Misconceptions of Random Events: The Case of Perceived Control and State-Sponsored Lotteries. Journal of Public
Policy & Marketing. https://​doi.​org/​10.​1509/​jppm.​20.2.​254.​17364
Muir, D. H. (2009). Gaming device and method providing a near miss insurance pool or fund. Google Patents.
Naqvi, N. H., Gaznick, N., Tranel, D., & Bechara, A. (2014). The insula: A critical neural substrate for craving
and drug seeking under conflict and risk. Annals of the New York Academy of Sciences. https://​doi.​org/​10.​
1111/​nyas.​12415
Nicholson, R., Graves, C., Ellery, M., et al. (2016). The Temporal Relationship Between Faulty Gambling Cognitions and Gambling Severity in Young Adults. Journal of Gambling Studies, 32, 1215–1229. https://​doi.​
org/​10.​1007/​s10899-​016-​9605-y
Orford, J., Wardle, H., Griffiths, M., Sproston, K., & Erens, B. (2010). PGSI and DSM-IV in the 2007 British
Gambling Prevalence Survey: Reliability, item response, factor structure and inter-scale agreement. International Gambling Studies. https://​doi.​org/​10.​1080/​14459​79090​35671​32
Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as a Participant Pool. Current Directions in Psychological Science. https://​doi.​org/​10.​1177/​09637​21414​531598
Peer, E., Vosgerau, J., & Acquisti, A. (2014). Reputation as a sufficient condition for data quality on Amazon
Mechanical Turk. Behavior Research Methods. https://​doi.​org/​10.​3758/​s13428-​013-​0434-y
Petry, N. M., Ginley, M. K., & Rash, C. J. (2017). A systematic review of treatments for problem gambling.
Psychology of Addictive Behaviors, 31(8), 951–961. https://​doi.​org/​10.​1037/​adb00​00290
Philander, K. S., Gainsbury, S. M., & Grattan, G. (2019). An Assessment of the validity of the gamblers belief
questionnaire. Addictive Behaviors, 97, 104–110. https://​doi.​org/​10.​1016/j.​addbeh.​2019.​05.​029
Rassen, J. A., Brookhart, M. A., Glynn, R. J., Mittleman, M. A., & Schneeweiss, S. (2009). Instrumental variables I: Instrumental variables exploit natural variation in nonexperimental data to estimate causal relationships. Journal of Clinical Epidemiology. https://​doi.​org/​10.​1016/j.​jclin​epi.​2008.​12.​005
Raylu, N., Oei, T. P. S., Loo, J. M. Y., & Tsai, J.-S. (2016). Testing the Validity of a Cognitive Behavioral
Model for Gambling Behavior. Journal of Gambling Studies, 32(2), 773–788. https://​doi.​org/​10.​1007/​
s10899-​015-​9567-5
Sargan, J. D. (1958). The Estimation of Economic Relationships using Instrumental Variables. Econometrica.
https://​doi.​org/​10.​2307/​19076​19
Schulte, L., Orlowski, S., Brandt, D., Bischof, A., Besser, B., Bischof, G., & Rumpf, H.-J. (2021). The role
of migration background and cognitive distortions in the trajectories of gambling problems—A longitudinal study. International Gambling Studies, 21(1), 17–37. https://​doi.​org/​10.​1080/​14459​795.​2020.​
17912​20
Shaffer, H. J., Ladouceur, R., Blaszczynski, A., & Whyte, K. (2016). Extending the RENO Model: Clinical and
Ethical Applications. American Journal of Orthopsychiatry. https://​doi.​org/​10.​1037/​ort00​00123

Journal of Gambling Studies (2023) 39:1189–1205

Sharp, C., Steinberg, L., Yaroslavsky, I., Hofmeyr, A., Dellis, A., Ross, D., & Kincaid, H. (2012). An Item
Response Theory Analysis of the Problem Gambling Severity Index. Assessment. https://​doi.​org/​10.​1177/​
10731​91111​418296
Sharpe, L. (2002). A reformulated cognitive–behavioral model of problem gambling: A biopsychosocial perspective. Clinical Psychology Review, 22(1), 1–25.
So, R., Matsushita, S., Kishimoto, S., & Furukawa, T. A. (2019). Development and validation of the Japanese
version of the problem gambling severity index. Addictive Behaviors, 98, 105987. https://​doi.​org/​10.​
1016/j.​addbeh.​2019.​05.​011
Sovey, A. J., & Green, D. P. (2011). Instrumental variables estimation in political science: a readers’ guide.
American Journal of Political Science. https://​doi.​org/​10.​1111/j.​1540-​5907.​2010.​00477.x
Staiger, D., & Stock, J. H. (1997). Instrumental variables regression with weak instruments. Econometrica,
65(3), 557–586. https://​doi.​org/​10.​2307/​21717​53
Stange, M., Brown, D. G., Harrigan, K., & Dixon, M. (2017a). Built-in bad luck: Evidence of near-miss outcomes by design in scratch cards. Journal of Gambling Issues, 36.
Stange, M., Grau, M., Osazuwa, S., Graydon, C., & Dixon, M. J. (2017b). Reinforcing Small Wins and Frustrating Near-Misses: Further Investigation Into Scratch Card Gambling. Journal of Gambling Studies. https://​
doi.​org/​10.​1007/​s10899-​016-​9611-0
Stanovich, K. E. (2000). Individual Differences in Reasoning: Implications for the Rationality Debate. Behavioral and Brain Sciences. https://​doi.​org/​10.​1017/​S0140​525X0​00034​35
StataCorp LLC. (2017). Stata (No. 15).
Steenbergh, T. A., Meyers, A. W., May, R. K., & Whelan, J. P. (2002). Development and validation of the Gamblers’ Beliefs Questionnaire. Psychology of Addictive Behaviors, 16(2), 143–149. https://​doi.​org/​10.​1037/​
0893-​164X.​16.2.​143
Stock, J. H., & Yogo, M. (2010). Testing for Weak Instruments in Linear IV Regression. In D. W. K. Andrews
& J. H. Stock (Eds.), Identification and Inference for Econometric Models: Essays in Honor of Thomas
Rothenberg (pp. 80–108). Cambridge University Press. https://​doi.​org/​10.​1017/​CBO97​80511​614491.​006
Toneatto, T. (1999). Cognitive psychopathology of problem gambling. Substance Use and Misuse. https://​doi.​
org/​10.​3109/​10826​08990​90394​17
Tversky, A., & Kahneman, D. (1971). Belief in the law of small numbers. Psychological Bulletin, 76(2), 105.
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive
Psychology. https://​doi.​org/​10.​1016/​0010-​0285(73)​90033-9
Welte, J. W., Barnes, G. M., Tidwell, M.-C.O., Hoffman, J. H., & Wieczorek, W. F. (2015). Gambling and problem gambling in the united states: changes between 1999 and 2013. Journal of Gambling Studies, 31(3),
695–715. https://​doi.​org/​10.​1007/​s10899-​014-​9471-4
Winfree, W. R., Ginley, M. K., Whelan, J. P., & Meyers, A. W. (2015). Psychometric evaluation of the Gamblers’ Beliefs Questionnaire with treatment-seeking disordered gamblers. Addictive Behaviors, 43,
97–102. https://​doi.​org/​10.​1016/j.​addbeh.​2014.​12.​016
Winfree, W. R., Meyers, A. W., & Whelan, J. P. (2013). Validation of a Spanish translation of the Gamblers’
Beliefs Questionnaire. Psychology of Addictive Behaviors. https://​doi.​org/​10.​1037/​a0030​824
Wong, S. S. K., & Tsang, S. K. M. (2012). Validation of the Chinese Version of the Gamblers’ Belief Questionnaire (GBQ-C). Journal of Gambling Studies. https://​doi.​org/​10.​1007/​s10899-​011-​9286-5
Wooldridge, J. M. (2010). Econometric Analysis of Cross Section and Panel Data. MIT Press.
Xian, H., Shah, K. R., Phillips, S. M., Scherrer, J. F., Volberg, R., & Eisen, S. A. (2008). Association of cognitive distortions with problem and pathological gambling in adult male twins. Psychiatry Research. https://​
doi.​org/​10.​1016/j.​psych​res.​2007.​08.​007
Xu, J., & Harvey, N. (2014). Carry on winning: The gamblers’ fallacy creates hot hand effects in online gambling. Cognition. https://​doi.​org/​10.​1016/j.​cogni​tion.​2014.​01.​002
Yau, Y. H. C., & Potenza, M. N. (2015). Gambling Disorder and Other Behavioral Addictions: Recognition
and Treatment. Harvard Review of Psychiatry, 23(2), 134–146. https://​doi.​org/​10.​1097/​HRP.​00000​00000​
000051
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such publishing agreement and applicable
law.