THE SOCIAL ENVIRONMENT OF SPORT AND ATHLETES’ DOPING-RELATED COGNITIONS 

By 

Tyler Scott Harris 

 

 

 

 

 

 

 

 

 

 

 

A DISSERTATION 

Submitted to 

Michigan State University 

in partial fulfillment of the requirements 

for the degree of 

Kinesiology – Doctor of Philosophy 

2020 

 

 

 

 
 

THE SOCIAL ENVIRONMENT OF SPORT AND ATHLETES’ DOPING-RELATED COGNITIONS 

PUBLIC ABSTRACT 

By 

Tyler Scott Harris 

Despite the best efforts of organizations like the World Anti-Doping Agency, athletes 

still make the conscious decision to use illegal performance enhancing drugs or methods, a 

behavior referred to as doping. Researchers have examined many factors an athlete might think 

about when making such a decision, even in the face of punishment for athletes who are caught 

doping. One thought process to consider is how athletes may make justifications to themselves 

for their behavior, endorsing what would otherwise be against the rules. Although these are 

internal processes, athletes also may take into consideration their perceptions of the social 

environment when developing these justifications. A greater understanding of the interplay 

between internal justifications and perceived social norms might allow for the development of 

more effective anti-doping educational programs for coaches and athletes. The goal of this two-

study dissertation was to contribute to this understanding. 

Study 1 involved a descriptive exploration of individuals’ thought processes within the 

social environment of athletes with physical disabilities. Disability sport athletes’ thought 

processes were expected to show a discrete number of patterns. Our hypothesis was supported, as 

we found four unique groups that differed in how prevalent they thought doping was, how much 

they thought the people around them would support/pressure them in doping, and how well they 

could justify to themselves that doping was acceptable. One group, for example, saw doping as 

more prevalent than the other groups, felt more pressure or support to dope than the other groups, 

and personally justified doping more than the other groups. Therefore, this group was considered 

 
 

the most at-risk for doping. Indeed, athletes in this group expected to feel less guilty if they were 

to use performance enhancing drugs, and reported greater doping intentions.  

Study 2 explored these social-cognitive variables in high-level able-bodied athletes using 

a different data analysis strategy. An interesting result in this study was how athletes’ personal 

justifications in regard to doping seemed to play a larger role in their doping decisions than their 

perceived prevalence or approval of doping. The level of guilt the athletes anticipated from a 

hypothetical doping scenario also played a role in these decisions. Even when taking into 

consideration the chance that athletes may respond deceptively to paint themselves in a better 

light, their personal thought processes seemed to be more influential to their doping decisions 

than perceptions of the social environment.  

As a whole, these two studies suggest athletes’ willingness to endorse the legitimacy of 

doping may be a dominant risk factor to actually engaging in the behavior, even if the social 

environment does not promote it. It is possible that because these were adult athletes, they had 

already internalized the norms of their environment, and therefore depended more on their own 

moral reasoning (or personal justifications, referred to as moral disengagement) to make doping 

decisions. Organizations such as the World Anti-Doping Agency have endorsed anti-doping 

educational programs and other campaigns to minimize the behavior, and may benefit from 

taking these thought processes into consideration.  

 

 

 
 

THE SOCIAL ENVIRONMENT OF SPORT AND ATHLETES’ DOPING-RELATED COGNITIONS 

ABSTRACT 

By 

Tyler Scott Harris 

 

Research on doping moral disengagement (MD; temporary endorsement of a 

transgressive behavior) suggests this cognition is influential in athletes’ doping-related decisions. 

This line of research should also include social variables such as subjective norms (perceived 

approval of or pressure to perform a behavior from significant others) and descriptive norms 

(perceived prevalence of a behavior), because these normative perceptions reflect the social 

context within which cognitions like doping MD take place. All of these variables have been 

found to independently predict outcomes such as anticipated guilt from doping (Ring & 

Kavussanu, 2017), doping intentions (Ntoumanis et al., 2014) and doping 

susceptibility/consideration (Boardley et al., 2019). Examining these variables together, 

especially in different populations (i.e., disability and able-bodied sport), could offer a fuller 

picture of doping-related cognitions. This holds long-term potential to develop more effective, 

theory-driven anti-doping interventions and maintain the spirit of fair play in sport. The purpose 

of this dissertation was to explore how morality-based cognitions and the social environment of 

sport (as reflected in norms-based cognitions) associate with doping-related cognitions. 

 

Study 1 was a cross-sectional examination of how doping MD, descriptive norms, and 

subjective norms responses of disability sport athletes are patterned. Cluster analysis revealed 

four distinct groups. One-way MANOVAs determined mean differences across groups on the 

variables of anticipated guilt and doping intentions. One cluster, characterized by high levels of 

all focal variables, was one of two clusters to score significantly lower on anticipated guilt, and 

 
 

the only cluster to score significantly higher on doping intentions than the other athlete groups. 

This at-risk pattern of cognitions, which was expressed in the smallest cluster (n = 15), warrants 

particular attention in future anti-doping campaigns.  

 

Study 2 used latent variable analysis to examine athletes’ doping MD, descriptive norms, 

and subjective norms as predictors of doping consideration, directly and indirectly via 

anticipated guilt. A model was tested with the expected relationships and exhibited adequate to 

excellent fit to the data. Doping MD had significant direct and indirect effects on doping 

consideration, and the direct effect persisted when controlling for social desirability. Such effects 

were not present for the descriptive norms and subjective norms constructs. Findings suggest the 

strength of the relationship between doping MD and doping-related outcomes is stronger than 

that tied to social norms. Such results may be applied to anti-doping educational programs, 

which have not often deliberately targeted doping MD.  

In sum, this dissertation extends the literature supporting the role of doping MD in 

athletes’ doping-related outcomes. This role persists across sport type and when controlling for 

social desirability. It also appears to be more salient than normative influences in predicting 

doping-related outcomes. Researchers and anti-doping interventionists may benefit from 

incorporating doping MD as a target of future research and education efforts, which could lead to 

more effective doping prevention. 

 

 

 
 

Copyright by 
TYLER SCOTT HARRIS 
2020 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 
 
 
 
 
 
 
 
 
 
 
 

To my family – 

biological and beyond. 

v 
 

ACKNOWLEDGEMENTS 

 

 

First and foremost, my advisor Dr. Alan Lyle Smith, for seeing a lot of potential in a 

master’s student from a little school with little sport psychology background and little research 

experience. You carved out the steep learning curve that will launch me into an impactful 

academic career, and molded me into a scholar despite the rough edges. In addition to your 

personal mentorship, you helped to surround me with fantastic faculty who could provide 

invaluable insight throughout my doctoral pursuits.  

 

My comprehensive exam committee was first to provide this insight. My knowledgeable 

committee – Drs. Florian Kagerer, Matthew Pontifex, and Wei Peng – challenged me to be 

informed on many aspects in the field of motivation and behavior, forcing me to motivate myself 

in my own background research. Our discussions were enlightening and enjoyable. 

 

Dr. Peng continued on and was joined by Drs. Leapetswe Malete and Nicholas Myers to 

sit on my dissertation committee, and I couldn’t have been more fortunate to work with this star-

studded group. You know that when I say, “go easy on me,” I am asking for quite the opposite, 

and will be all the better for it. 

 

My academic siblings are as valuable as my academic predecessors. You let me join you 

for walks to get coffee, even though I do not drink coffee and just came to hang out. Our shared 

experience as members of the SiMPL squad will be the foundation for lifelong friendship and 

collaboration, and I’m excited for the future ahead of us.  

 

Earning a Ph.D. required an undeniable work ethic, and it is undeniable that my work 

ethic is the product of what I saw in my parents, Meena and Keith. You worked long and late 

hours, yet made time to attend my choir concerts, basketball games, and graduations. Every 

vi 
 

success of mine is a reflection of your success as a mother and father. You have raised not just 

one, but two amazing kids, and I am forever grateful. 

 

My sister, Syreeta, set the example for me from a young age. For as far apart as we were 

in age, I am glad we are as close as we are today. I have gone to you in times of need and you’ve 

offered pivotal advice, and that is unlikely to change. You have always been an inspiration to 

me, and I hope to make you proud. 

 

Ashley, my fiancée, has been the strongest support system as of late. For every time I 

said, “This is going to be the most difficult semester of all, I don’t know how I’m going to get 

through it,” you were there to say, “You said that last semester, and you made it. Everything will 

be okay.” You make everything okay, and I find comfort in knowing you will be there to ensure 

it for years to come. 

 

This list is long, but certainly not exhaustive. It takes a village to raise a child, but much 

more to raise a Ph.D. graduate. You have all helped me to pursue my passions, and I’m a better 

person for it. “Thank you” could never be enough to represent the gratitude I have for the part 

you have played in my journey. 

 

 

 

vii 
 

TABLE OF CONTENTS 

 
 
LIST OF TABLES .......................................................................................................................... x 
 
LIST OF FIGURES ....................................................................................................................... xi 
 
CHAPTER ONE: GENERAL INTRODUCTION ......................................................................... 1 
 
CHAPTER TWO: STUDY ONE DISABILITY SPORT ATHLETES' DOPING COGNITIONS 
AND THE IMPACT ON ANTICIPATED GUILT AND INTENTIONS ..................................... 9 
Preface......................................................................................................................................... 9 
Abstract ..................................................................................................................................... 10 
Introduction ............................................................................................................................... 11 
Method ...................................................................................................................................... 15 
Participants ............................................................................................................................ 15 
Instruments ............................................................................................................................ 16 
Procedure .............................................................................................................................. 19 
Data analysis ......................................................................................................................... 19 
Results ....................................................................................................................................... 20 
Descriptive statistics ............................................................................................................. 20 
Cluster analysis ..................................................................................................................... 21 
Group difference analyses ..................................................................................................... 23 
Discussion ................................................................................................................................. 24 
Implications ........................................................................................................................... 28 
Limitations ............................................................................................................................ 29 
Conclusion ............................................................................................................................ 31 

 
CHAPTER THREE: STUDY TWO DOPING MORAL DISENGAGEMENT PREDICTS 
ANTICIPATED GUILT AND DOPING CONSIDERATION WITHIN THE CONTEXT OF 
SOCIAL NORMS ......................................................................................................................... 32 
Abstract ..................................................................................................................................... 32 
Introduction ............................................................................................................................... 33 
Method ...................................................................................................................................... 37 
Participants ............................................................................................................................ 37 
Instruments ............................................................................................................................ 38 
Procedure .............................................................................................................................. 42 
Data Analysis ........................................................................................................................ 43 
Results ....................................................................................................................................... 43 
Descriptive Statistics ............................................................................................................. 43 
Model Testing ....................................................................................................................... 45 
Supplemental Analyses ......................................................................................................... 48 
Discussion ................................................................................................................................. 49 
Implications ........................................................................................................................... 54 
Limitations ............................................................................................................................ 55 

viii 
 

Conclusion ............................................................................................................................ 57 

 
CHAPTER FOUR: GENERAL DISCUSSION ........................................................................... 58 
 
APPENDICES .............................................................................................................................. 71 
APPENDIX A Study One - Human Research Protection Program Approval Letter ............... 72 
APPENDIX B Study One - Questionnaire Packet  ................................................................... 74 
APPENDIX C Study Two - Human Research Protection Program Approval Letter  .............. 82 
APPENDIX D Study Two - Questionnaire Packet ................................................................... 84 

 
REFERENCES ............................................................................................................................. 93 
 
 

 

ix 
 

 

LIST OF TABLES 

Table 2.1. Descriptive Statistics and Correlations for Study Variables of Final Sample (n = 185) 
....................................................................................................................................................... 21 
 
Table 2.2. Participant Numbers, Means, Standard Deviations, and Standardized Scores for 
Cluster Characteristics .................................................................................................................. 23 
 
Table 2.3. MANOVA Results for Cluster Differences ................................................................. 24 
 
Table 3.1. Descriptive Statistics and Correlations for Study Variables of Final Sample (n = 238)
....................................................................................................................................................... 45 
 
Table 3.2. Items, Standardized Factor Loadings, and Error Variances in Study 2 ....................... 46 
 
 
 

 

 

 

 

x 
 

 

LIST OF FIGURES 

Figure 2.1. Results of k-means cluster analysis (N = 185) ........................................................... 22 
 
Figure 3.1. Hypothesized model for proposed Study 2 ................................................................ 37 

Figure 3.2. Model testing results................................................................................................... 47 

 

 

xi 
 

CHAPTER ONE: GENERAL INTRODUCTION 

 

The International Olympic Committee (IOC) includes “social responsibility and respect 

for universal fundamental ethical principles” in the first fundamental principle of Olympism 

(IOC, 2015, pg. 13). Doping, the use of prohibited substances or methods with the potential to 

artificially improve performance through changes in physical or mental condition (IPC, 2018; 

Lauré, 1997; WADA, 2021), is a direct contravention of this principle. Not only is doping 

against the spirit of sport (IOC, 2015), it introduces risks to physical and psychological well-

being (Pope et al., 2014). As such, doping “poses serious public health hazards and calls for 

preventive action” (European Union, 2011; p. 6). 

 

The most prominent method used by anti-doping organizations to prevent doping is the 

detection-deterrence method (i.e., sanctions for athletes who commit anti-doping rule violations; 

Morente-Sánchez & Zabala, 2013). The idea is that the risk of getting caught and sanctioned will 

deter athletes from doping. Unfortunately, this method has found minimal success in deterring 

doping in sport (Strelan & Boeckman, 2006; Westmattelmann, Dreiskämper, Strauß, Schewe, & 

Plass, 2018). Despite the risk of being caught and sanctioned, some athletes still use 

performance-enhancing substances or methods, meaning there must be other factors contributing 

to doping decisions. Among these are moral and social cognitions. Understanding these 

cognitions is of interest in order to develop theory-based anti-doping initiatives. 

 

Personal moral beliefs have been argued to be a stronger influence on an individual’s 

decision to dope than drug testing, at least in hypothetical situations (Strelan & Boeckman, 

2006). According to the social-cognitive theory of moral thought and action (Bandura, 1991), 

people develop an understanding of morality based on the social sanctions received for acting 

immorally. Over time, these sanctions are internalized, and people act in line with their 

1 
 

individual moral beliefs to avoid self-sanctions (i.e., negative emotions such as guilt or shame). 

Punishments for doping (e.g., fines or suspensions for anti-doping rule violations) represent the 

social sanctions in the context of sport. For those athletes who internalize these sanctions, they 

would choose not to use performance-enhancing substances because they would feel guilty for 

violating the morals of their sport. However, there are certain cognitive mechanisms that can 

help an athlete circumvent the anticipated guilt brought on by doping behavior. One of these 

mechanisms is moral disengagement (MD; Bandura, 1990), which is the conditional 

endorsement of an otherwise transgressive or reprehensible behavior.  

 

Of the eight mechanisms of MD originally specified by Bandura (1990), six have been 

shown to be used for endorsing doping behavior (Boardley, Grix, & Dewar, 2014). Moral 

justification involves making the behavior seem morally or socially acceptable by offering a way 

in which it can do some good. Euphemistic labeling uses benign language to distort the perceived 

reprehensibility of the behavior. Advantageous comparison is when the transgressor contrasts the 

behavior with a more outrageous flagrant offense. Displacement of responsibility is putting the 

responsibility on some other authoritative decision-maker. Diffusion of responsibility is 

spreading the responsibility of the decision to act to a group in order to minimize individual 

responsibility. Lastly, distortion of consequences involves minimizing or ignoring the negative 

ramifications of the behavior. The two original mechanisms of dehumanization (i.e., removing 

the victim’s human qualities) and attribution of blame (i.e., considering the victim’s actions to be 

a provocation of the transgressor) are relevant in other anti-social behaviors in sport, but not 

doping specifically. Recent research has suggested that MD, specifically in relation to doping 

behavior, can allow an athlete to dope while minimizing anticipated feelings of guilt (Boardley, 

2 
 

Smith, Mills, Grix, & Wynne, 2017; Ring & Kavussanu, 2017, Ring, Kavussanu, & Gürpinar, 

2020). 

Self-sanctions such as guilt materialize based on internalization of the negative 

consequences (i.e., social sanctions) of a behavior (Bandura, 1991). Though individuals are 

believed to act in accordance with their own morality once internalization of social sanctions has 

occurred, doping-related cognitions do not occur in isolation. The athletic environment provides 

context for doping-related cognitions, regardless of whether sanctions have been internalized by 

the athlete. Therefore, environmental social factors (i.e., normative influences) should be 

considered when studying the relationship between doping MD and doping-related outcomes.  

 

There are two distinct types of normative influence: subjective norms and descriptive 

norms (Cialdini, Reno, & Kallgren, 1990). Subjective norms (Ajzen, 1991) are the perceived 

pressures significant others put on an individual to perform or not perform a behavior. The 

construct of descriptive norms originates in research on littering (Cialdini et al., 1990), and refers 

to one’s perceptions about the actual behavior of others. In doping research, descriptive norms 

are typically measured through perceived prevalence (Barkoukis, Lazuras, & Tsorbatzoudis, 

2014; Lazuras et al., 2010). Studies have supported the association of both subjective norms and 

descriptive norms with self-involvement in doping behavior (Petróczi, Mazanov, Nepusz, 

Backhouse, & Naughton, 2008; Uvacsek et al., 2011). A seminal meta-analysis supported both 

subjective and descriptive norms as predictors of doping behavior, both directly and indirectly 

through doping intentions (Ntoumanis, Ng, Barkoukis, & Backhouse, 2014). Further, a test of a 

more recent integrative model of doping use supported both types of norms as proximal 

predictors of doping intentions (Lazuras, Barkoukis, & Tsorbatzoudis, 2015). 

3 
 

Though doping MD, subjective norms, and descriptive norms have all been shown to be 

independent predictors of doping outcomes, the simultaneous influence of these variables has not 

typically been considered. Some studies emphasizing descriptive norms include measures of 

sportspersonship or personal morality rather than doping MD (Barkoukis, Lazuras, 

Tsorbatzoudis, & Rodafinos, 2013; Gucciardi, Jalleh, & Donovan, 2011), whereas studies 

emphasizing doping MD examine it alongside subjective norms exclusively (Lucidi et al., 2008). 

In effect, the athletic environment contains various potential influences on an athlete’s cognitions 

and decisions regarding doping (Barkoukis, Brooke, Ntoumanis, Smith, & Gucciardi, 2019). 

Research on these variables in combination can capture a broader picture of the athletic 

environment as it relates to doping. 

Yet another opportunity to broaden the understanding of doping-related cognitions is to 

explore it in special populations. A particularly underrepresented population in doping research 

is athletes with physical disabilities. In 2019, the United States Olympic Committee changed its 

name to the United States Olympic and Paralympic Committee (USOPC, 2019), and the board 

passed a vote that would increase monetary rewards for Paralympic medalists to the same 

amount as their Olympic counterparts. With the continued success of the Paralympic movement, 

researchers have voiced concern about the growth of participation, competition, and 

compensation of parasport athletes creating a “win-at-all-costs” attitude toward these sports (Van 

de Vliet, 2012). This changing perspective may increase doping behavior in disability sport. For 

this reason, the International Paralympic Committee has also published an Anti-Doping Code 

(IPC, 2018). Doping methods such as boosting (artificial induction of autonomic dysreflexia; 

Legg & Mason, 1998; Mazzeo, Santamaria, & Iavarone, 2015) are unique to the disability sport 

4 
 

environment, and research in this context may provide greater insight into doping issues as a 

whole.  

A greater knowledge of the psychosocial factors affecting doping outcomes could have 

many implications for future research and practice. Beyond the detection-deterrence method, 

anti-doping educational programs such as ATLAS and ATHENA (Barkoukis, Kartali, Lazuras, 

& Tsorbatzoudis, 2016) work to decrease adolescents’ positive attitudes toward prohibited 

performance-enhancing substances. However, there is an absence of strong support for a causal 

relationship between doping attitudes and doping outcomes (Backhouse, McKenna, Robinson, 

and Atkin, 2007). More recent interventions have targeted variables such as beliefs about 

supplement use (Hurst, Ring, & Kavussanu, 2020), coach communication style, and motivational 

climate (Ntoumanis et al., 2018; Ntoumanis et al., 2020). These are recent examples of practical 

outcomes following psychosocial doping research. Deeper understanding of other psychosocial 

contributors to doping, such as those mentioned above, can inform the curriculum of anti-doping 

educational programs and help to develop theory-informed interventions, thus contributing to 

athlete well-being and upholding the spirit of sport. 

The overarching goal of this dissertation was to explore how the social environment of 

sport—as reflected in morality-based and norms-based cognitions, uniquely and in 

combination—associates with doping-related outcomes. Previous research on moral cognitions, 

normative influences, and other behavioral outcomes provided a foundation to design the studies 

herein. The two studies explored social influences of doping decisions in unique ways, varying in 

methodology, sample population, and focal variables. The studies in this document extend 

previous research by simultaneously examining social norms and doping MD in two distinct 

populations (e.g., disability sport and able-bodied sport) as predictors of various doping 

5 
 

outcomes (e.g., anticipated guilt, doping intentions, and doping susceptibility/consideration). 

Altogether, these studies offer insight regarding the morality- and norms-related factors tied to 

doping decisions in the athletic environment. 

 

Study 1 was a cross-sectional, survey-based examination of how doping MD, subjective 

norms, and descriptive norms may cluster together to evince distinct groups of disability sport 

athletes. These variables have rarely been examined simultaneously in either disability sport 

athletes or their able-bodied counterparts, and Study 1 represents an initial venture into the 

combination of these variables. The exclusive recruitment of disability sport athletes, in contrast 

to their able-bodied counterparts, was a novel contribution to this line of research. Doping 

practices have been discovered in a number of disability sport events (Van de Vliet, 2012), and 

although there are differences between the social contexts of disability and able-bodied sport, 

research on disability sport athletes’ doping-related cognitions can help to uphold the Anti-

Doping Code published by the IPC. Due to the novelty of this research being conducted in a 

disability sport context, an exploratory statistical approach was employed to describe 

relationships among the variables. From this, a more descriptive foundation could be set in order 

to springboard future research on doping-related cognitions in disability sport athletes. Therefore, 

the purpose of Study 1 was two-fold: a) to examine profiles of disability sport athletes’ in terms 

of patterns of responding to doping cognitions measures and b) to determine whether 

associations between these cognitions and doping outcomes of interest vary by profile 

membership.  

 

In Study 2, subjective norms, descriptive norms, and doping MD were examined in a 

sample of able-bodied adult athletes. Additionally, an outcome measure of doping consideration 

(Gucciardi, Jalleh, & Donovan, 2010) was used in place of doping intentions. Doping 

6 
 

consideration is distinct from doping intentions in that it is based on the amount of consideration 

one would give to a hypothetical opportunity to use a performance-enhancing substance without 

detection. It is important to note here that this measure has previously been called doping 

susceptibility; however, its use in this study is better described as a measure of doping 

consideration than doping susceptibility. Therefore, the operationalization used in Study 2 was 

referred to as doping consideration. Examining doping MD, perceived norms, anticipated guilt, 

and doping consideration in an able-bodied athletic population can provide more robust support 

for the associations between these psychosocial variables. Furthermore, use of a more vigorous 

statistical analysis would have theoretical implications for the associations found. For example, 

the construct of doping MD stems from the social-cognitive theory of moral thought and action 

(Bandura, 1991), whereas the construct of subjective norms stems from the theory of planned 

behavior (Ajzen, 1991) and descriptive norms from research on littering (Cialdini et al., 1990). 

An integrated model informed by these various theoretical foundations would allow for a 

comparison of the theories. Therefore, the purpose of Study 2 was to test a theoretically-

informed, integrated conceptual model in which doping MD, subjective norms, and descriptive 

norms predicted doping consideration, both directly and indirectly through anticipated guilt 

(Figure 3.1).  

 

Together, this series of studies can increase the knowledge base of psychosocial 

correlates of an athlete’s doping-related outcomes. Research involving both disability sport 

athletes and their able-bodied counterparts contributes to the literature by exploring the potential 

generalizability of previously established constructs and relationships. In addition, research on 

the role of normative influences broadens the literature to more agents in the athlete entourage 

(Barkoukis et al., 2019). With this further knowledge, organizations such as the IOC and the 

7 
 

World Anti-Doping Agency (WADA) can establish best practices when creating educational 

programs and other anti-doping campaigns. Developing a successful, theory-informed 

intervention to reduce the prevalence of doping in sport would be a large step forward in the 

fight to maintain the “spirit of friendship, solidarity and fair play” (IOC, 2015, pg. 13), and 

would protect the well-being of the athletes involved. 

 

 

 

8 
 

CHAPTER TWO: STUDY ONE 

DISABILITY SPORT ATHLETES’ DOPING COGNITIONS AND THE IMPACT ON 

ANTICIPATED GUILT AND INTENTIONS 

Preface 

 

The initial proposal for this study was presented in March 2019 at the Eastern Canada 

Sport and Exercise Psychology Symposium (ECSEPS) in York, Ontario, Canada, and results of 

the final study were presented in October 2019 in Vancouver, British Columbia, Canada at the 

Canadian Society for Psychomotor Learning and Sport Psychology (SCAPPS) conference. 

 

 

 

 

9 
 

Abstract 

 

Research suggests doping moral disengagement (MD; conditional endorsement of a 

transgressive behavior) is associated with reduced anticipated guilt (Kavussanu & Stanger, 2017) 

and increased doping intentions (Ntoumanis et al., 2014). However, MD occurs within the 

greater social context. Subjective (perceived support from significant others) and descriptive 

(perceived doping prevalence) norms should be considered, as moral behavior is a product of 

social circumstances and moral reasoning (Bandura, 1991). The purpose of this study was to 

examine profiles of disability sport athletes’ doping cognitions, and whether these profiles differ 

in anticipated guilt or intentions to dope. A survey was completed by 186 athletes (Mage = 36.5 

years, 77.4% male, 53% wheelchair basketball) assessing norms (Barkoukis et al., 2014; Lazuras 

et al., 2015), doping MD (Boardley et al., 2018), anticipated guilt (Boardley et al., 2017), and 

doping intentions (Lazuras et al., 2010). Two step-cluster analysis revealed four distinct clusters, 

including two mixed profiles. One-way MANOVAs revealed lower anticipated guilt in the two 

clusters characterized by higher doping MD. Higher doping intentions were reported in the 

cluster characterized by relatively higher doping MD and subjective norms. Some findings may 

be due to distinct characteristics of the disability sport context, but most results were in line with 

previous research. Future work should examine these variables and others in different 

populations for a more robust understanding of their role in disability and able-bodied sport 

doping. Regardless, this study suggests there is a group of athletes most susceptible to doping 

without guilt who may be a key target for anti-doping campaigns. 

 

 

 

10 
 

Introduction 

In 2019, the United States Olympic Committee changed its name to the United States 

Olympic and Paralympic Committee (USOPC, 2019), and the board passed a vote that would 

increase monetary compensation for Paralympic medalists to the same amount as their Olympic 

counterparts. These are prominent signs of the recent success of the Paralympic Movement, 

which may catalyze participation opportunities for individuals with disabilities (de Jong, 

Vandreusel, & van Driel, 2010). Though this inclusivity is positive in many respects, the growth 

of participation, competition, and compensation has the potential to create a “win-at-all-costs” 

attitude toward disability sport (Van de Vliet, 2012). Athletes looking to gain an advantage may 

be persuaded to behave in ways that go against the spirit of sport; particularly, doping. Doping is 

the use of prohibited substances or methods by an athlete with the potential to artificially 

improve performance through changes in a physical and/or mental condition (Lauré, 1997). 

Multiple instances of anti-doping rule violations exist in past parasport events (Van de Vliet, 

2012). With these concerns in mind, the International Paralympic Committee published a World 

Anti-Doping Code in 2004 (IPC, 2018).  

Doping is not only a contravention of the spirit of sport (IOC, 2015), but also introduces 

risks to physical and psychological well-being (Pope et al., 2014). As such, doping “poses 

serious public health hazards and calls for preventive action” (European Union, 2011; p. 6). 

Common strategies to combat this behavior have been detection-deterrence and education 

programs (Morente-Sánchez & Zabala, 2013). Detection-deterrence involves sanctions as 

punishment for athletes who commit anti-doping rule violations, with the hopes that the cost of 

being caught outweighs the perceived potential benefits of doping. These strategies have found 

minimal success, because these sanctions have little influence on athletes’ cognitions 

11 
 

surrounding doping (Strelan & Boeckman, 2006). For this reason, researchers have attempted to 

better understand psychosocial contributors to doping intentions and behavior. 

Because doping is a contravention of sport ethic, athletes may use moral reasoning to 

make their decisions regarding doping. The social-cognitive theory of moral thought and action 

(Bandura, 1990; 1991) suggests that people anticipate the affective consequences of their actions 

and behave in a way that will bring about desirable affective outcomes. In terms of social and 

moral conduct, internal values set the standard for behavior. Acting in opposition to this standard 

can create a negative self-reaction. A person may therefore act morally in order to avoid 

anticipated self-reactions such as feelings of guilt or shame.  

When faced with the decision to dope, an athlete may be drawn to the idea of improved 

performance yet anticipate the possible guilt they may feel from conducting such behavior. This 

tension may require certain mechanisms before deciding whether or not to dope, such as moral 

activation or disengagement. Moral disengagement (MD) provides an opportunity to relieve this 

tension, as it is the conditional endorsement of an otherwise transgressive behavior (Bandura, 

1991). Research suggests athletes with spinal cord injury report less antisocial behavior in sport 

than their able-bodied counterparts, but these group differences are mediated by the use of MD 

(Kavussanu, Ring, & Kavanagh, 2015). Thus, more research is needed on MD in this population 

when examining a behavior such as doping. 

Of the eight original MD mechanisms, six have been shown to potentially facilitate 

doping behavior (Boardley, Grix, & Harkin, 2014). These are moral justification (i.e., making a 

transgression appear acceptable due to perceived social or moral benefits), euphemistic labeling 

(i.e., using alternative language to reduce the reprehensibility of the transgression), advantageous 

comparison (i.e., contrasting against a more harmful behavior so the behavior in question appears 

12 
 

less damaging in comparison), displacement of responsibility (i.e., placing responsibility onto 

implicit or explicit social pressures), diffusion of responsibility (i.e., dispersing responsibility 

amongst a wider protagonist), and distortion of consequences (i.e., minimizing or ignoring the 

adverse effects of the behavior). Recent research has suggested that doping MD can allow an 

athlete to dope while minimizing anticipated feelings of guilt (Boardley, Smith, Mills, Grix, & 

Wynne, 2017; Ring & Kavussanu, 2017).  

Though a cognition like doping MD is a positive predictor of doping outcomes, morality-

based cognitions do not occur in isolation. In fact, two mechanisms of doping MD refer to the 

social norms surrounding doping: diffusion of responsibility and displacement of responsibility. 

There are two types of norms: subjective norms (sometimes referred to as injunctive) and 

descriptive norms (Cialdini, Reno, & Kallgren, 1990). Descriptive norms may refer to diffusion 

of responsibility, as they are based on the perceived behavior of others. In the doping context, 

these norms are measured based on the athlete’s perceived prevalence of doping among their 

peers (Lazuras, Barkoukis, Rodafinos, & Tzorbatzoudis, 2010). Subjective norms may be 

relevant to displacement of responsibility, as they are perceived expectations of significant others 

regarding the consequences of a behavior (Ajzen, 1985). In the case of doping, subjective norms 

are the athlete’s perceptions of the level of support/approval they would receive from significant 

others in their life in their decision to dope.  

Support for normative influences as predictors of doping intentions and behavior has 

grown over the last decade. Uvacsek et al. (2011) found that users of performance enhancing 

drugs overestimated the prevalence of the behavior in their sport. A test of a more recent 

integrative model of doping use (Lazuras, Barkoukis, & Tsorbatzoudis, 2015) supported both 

subjective norms and descriptive norms as proximal predictors of doping intentions. In a seminal 

13 
 

meta-analysis, norms (both subjective and descriptive) were positive predictors of doping 

intentions and doping behavior (Ntoumanis, Ng, Barkoukis, & Backhouse, 2014).  

Considering the similarities between mechanisms of doping MD and normative 

influences surrounding the behavior, athletes may use the social context in addition to their 

individual moral views when making decisions about doping. For instance, an athlete high in 

doping MD, who also perceives environmental factors to create social pressures to dope, may be 

particularly at risk for doping. In contrast, lower levels of any of these variables may minimize 

the association between the other variables and doping-related outcomes. Moral conduct is 

reciprocally regulated by the interaction between self-sanctions and social influences (Bandura, 

1991), suggesting norms-based cognitions should be simultaneously considered when studying 

doping MD and anticipated guilt.  

Though doping MD, descriptive norms, and subjective norms are independent predictors 

of doping intentions, these cognitions have not often been examined simultaneously. Some 

studies emphasizing descriptive norms include measures of sportspersonship or personal 

morality but not doping MD (Barkoukis, Lazuras, Tsorbatzoudis, & Rodafinos, 2013; Gucciardi, 

Jalleh, & Donovan, 2011), whereas studies emphasizing doping MD examine it alongside 

subjective norms exclusively (Lucidi et al., 2008). This gap in the literature should be filled, as 

this may help contextualize the socialization process. Socialization is the process by which social 

sanctions are internalized by an individual (Bandura, 1991). A study of this nature can help to 

understand the relationship between perceived social sanctions (in the form of subjective and 

descriptive norms) and self-sanctions (in the form of MD and anticipated guilt) and how 

combinations of these variables predict doping outcomes. 

14 
 

The purpose of the current study was two-fold. The first purpose was to describe the 

patterns of relationships between doping MD, subjective norms, and descriptive norms within a 

sample of disability sport athletes. The current study may be one of the first to examine these 

particular cognitions together within this population. It was hypothesized that there are a finite 

number of groups, or clusters, which can be characterized by athletes’ unique responses to the 

variables of doping MD, subjective norms, and descriptive norms. The second purpose was to 

assess potential cluster differences in doping outcomes such as anticipated guilt and doping 

intentions. Understanding how group differences predict these doping-related outcomes may help 

anti-doping campaigns to tailor their messages to the most at-risk athletes. Based on the current 

literature, it was hypothesized that groups characterized by higher doping MD, subjective norms, 

and descriptive norms would report lower levels of anticipated guilt and higher levels of 

intentions than those characterized by lower scores on any or all of these variables. 

Participants 

Method 

Athletes with physical disabilities (N = 186) ranging in age from 17 to 69 years (M = 

37.6, SD = 12.47) were recruited at regional, national, and international events and through 

disability sports organizations between March and August of 2019. The sample was 78.0% male 

and 22.0% female, and was ethnically diverse (13.4% affirming Hispanic or Latino ethnicity; 

races reported were 54.5% White, 20.5% Black or African-American, 10.9% unknown/other, 

6.4% more than one race, 4.5% American Indian/Alaskan Native, 3.2% Asian). Sports in which 

the athletes participated were wheelchair basketball (45.1%), para track and field/cycling 

(14.3%), wheelchair tennis (12.6%), other (11.4%), wheelchair baseball/softball (7.4%), 

wheelchair rugby (6.9%), shooting/air rifle (1.7%), and powerlifting/weightlifting (0.6%). The 

15 
 

participants’ disability classifications were spinal cord injury (37.8%), amputation (13.0%), spina 

bifida (10.3%), cerebral palsy (7.6%), and other/multiple physical disabilities (31.3%). 

Instruments 

Included in the survey was the aforementioned definition of doping, explicitly stating 

how prohibited methods are included alongside prohibited substances used to enhance athletic 

performance. Following questions collecting demographic information (i.e., age, sex, 

ethnicity/race, sport, level of competition, years in the sport, training hours per week, training 

weeks per year, type of disability), the survey included a series of relevant validated instruments 

to measure the core study variables. Further details for each instrument follow.  

Doping moral disengagement. 

Doping MD was measured using the Doping Moral Disengagement Scale (DMDS; 

Boardley et al., 2018). The DMDS was selected because of its versatility to measure the 

construct of doping MD in both individual and team sports. Additionally, development of this 

measure evinced six dimensions relating to the six MD mechanisms, collectively representing 

one higher-order factor: doping MD. The measure consists of 18 items (e.g., “Compared to most 

lifestyles in the general public, doping isn’t that bad”), with three items for each of the six 

dimensions of doping MD. Participants responded with their level of agreement with each 

statement on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). Internal 

consistency reliability of the scores was strong in the developmental stage of the measure, with 

Cronbach’s alpha values of .95 in one sample and .96 in another (Boardley et al., 2018). Internal 

consistency reliability of the scores in the present study was also strong, with a Cronbach’s alpha 

coefficient for the DMDS of .91. 

16 
 

Descriptive norms. 

Descriptive norms were measured using three items (Barkoukis, Lazuras, & 

Tsorbatzoudis, 2014) from previous research with able-bodied athletes. All items were measured 

on a percentage scale, as they related to perceived prevalence (e.g., “Out of 100%, how many 

athletes at your competitive level do you believe engage in doping to enhance their 

performance?”). The first item was in relation to athletes at the competitive level of the 

respondent and the latter two asked for perceived prevalence in relation to elite athletes in the 

United States. Internal consistency reliability information on scores from this measure has not 

been reported in previous research (Lazuras et al., 2010; Lazuras et al., 2015); however, internal 

consistency reliability of scores in the current study was good, with a Cronbach’s alpha value 

of .85. 

Subjective norms. 

Subjective norms were measured using three items (e.g., “Most people who are important 

to me would want me to use prohibited substances or methods to enhance my performance.”) 

scored on a 1 (strongly disagree) to 7 (strongly agree) scale (Lazuras et al., 2010; Lazuras et al., 

2015). The phrase “or methods” was added to the original items to reflect the provided definition 

of doping in the current study. Greater scores on this measure represented greater perceived 

social pressure to use performance enhancing substances. Previous research with these items has 

shown scores to possess good internal consistency reliability, with Cronbach’s alpha coefficients 

of .84 (Lazuras et al., 2010) and .81 (Lazuras et al., 2015). Internal consistency reliability of 

scores in the present study was comparable, with a Cronbach’s alpha coefficient of .77. 

17 
 

Anticipated guilt. 

Anticipated guilt was measured using five items from the State Shame and Guilt Scale 

(Marschall, Saftner, & Tangney, 1994) adapted to a hypothetical doping situation (Boardley, 

Smith, Mills, Grix, & Wynne, 2017). Participants were asked to read the following scenario:  

Having returned to training following a period of injury, you are feeling very out of 

shape. As such, you feel the need to get back in shape as soon as possible. A friend who you train 

with has been taking a training supplement that he/she says really helped him/her get back in 

shape quickly following a similar injury. He/she offers to give you some and you decide to take it. 

Subsequently you get back in shape much quicker than expected, but then discover the 

supplement you have been taking is a banned performance-enhancing substance. However, due 

to the improvements you have experienced, you decide to continue taking the substance. 

After reading this scenario, the participant was asked to respond to five items rating the 

level at which they would feel particular aspects of guilt (e.g., “I would feel remorse, regret”). 

Responses were on a five-point scale ranging from 1 (not at all) to 5 (extremely), with greater 

scores reflecting greater anticipated guilt. Previous research using this measure showed scores to 

exhibit good internal consistency reliability, with a Cronbach’s alpha coefficient of .95 (Boardley 

et al., 2017). Internal consistency reliability of scores in the present study was strong, with a 

Cronbach’s alpha value of .95. 

Doping intentions. 

Intentions were measured using three items (e.g., “I intend to use prohibited substances or 

methods to enhance my performance during this season”) from a study by Lazuras et al., (2010). 

The phrase “or methods” was added to each item to better reflect the current study’s definition. 

Responses were on a seven-point scale ranging from 1 (Definitely not) to 7 (Definitely yes), with 

18 
 

greater scores reflecting higher intentions to use prohibited substances or methods. Previous 

research showed scores to exhibit strong internal consistency reliability, with a Cronbach’s alpha 

coefficient of .97. Internal consistency reliability of scores in the present study was strong as 

well, with a Cronbach’s alpha value of .92. 

Procedure 

After obtaining IRB approval, participants were recruited to take an in-person, pencil-

and-paper survey. The primary investigator initially contacted disability sport athletes, coaches, 

and program coordinators by phone or email using publicly available contact information to 

describe the study. In addition, recruitment of participants occurred in person at regional, 

national, and international disability sport events in the United States and Canada. If allowed, the 

survey was described to athletes verbally as well as on the first page of the survey. Whenever 

possible, surveys were distributed to entire teams simultaneously. Participants were asked not to 

include their name anywhere on the survey, as completing the survey after receiving verbal and 

written information about the study risks and benefits constituted informed consent. Participants 

completed the survey individually or with assistance from either the investigator or an assistant 

of their choice. Examples of help offered included reading the questions to participants with 

visual impairments or marking answers for participants who had difficulty writing. 

Data analysis 

Two-step cluster analysis was conducted using the IBM Statistical Package for the Social 

Sciences v22.0 (SPSS Inc., Chicago, IL) to address the first purpose of this study. Variables used 

to create clusters were doping MD, subjective norms, and descriptive norms. First, a hierarchical 

cluster analysis using Ward’s linkage method and squared Euclidean distance as the similarity 

19 
 

measure was conducted. This step provided guidance on how many clusters should be retained 

for the analysis. The second step involved k-means (non-hierarchical) cluster analysis. 

Following cluster analysis, the second purpose of the study was addressed by comparing 

the clusters that emerged based on differences in the dependent variables of anticipated guilt and 

doping intentions. To test these differences, one-way multivariate analyses of variance 

(MANOVAs) were conducted using cluster identity as the grouped variable. 

Descriptive statistics 

Results 

Means and standard deviations were calculated for all variables. Overall, athletes 

reported moderately low levels of doping MD (M = 2.23, SD = 0.94), perceived prevalence of 

24.15% (SD = 20.27), low levels of subjective norms (M = 1.39, SD = .72), moderately high 

levels of anticipated guilt (M = 3.82, SD = 1.20), and low intentions (M = 1.19, SD = 0.75). 

Means, standard deviations, and correlations of all variables can be found in Table 1. 

After scale scores were calculated using the data for each measure, standardized z-scores 

were used to search for outliers. Cases with scores greater than 3.29 SD from the mean for any 

variable were considered outliers. In the complete data set, seven outliers were found. However, 

removal of outliers should be carefully considered when conducting cluster analysis, as they 

might represent undersampling of a bona fide group within the population (Hair et al., 1998). For 

this reason, all outliers were retained for the first step of the cluster analysis to determine 

whether these outliers might properly represent the population sampled. 

 

 

 

20 
 

Variable 
1. Doping MD 
2. Descriptive Norms 
3. Subjective Norms 
4. Anticipated Guilt 
5. Intentions 

 

Mean 
2.23 
24.15 
1.39 
3.82 
1.19 

SD 
.94 
20.27 
0.72 
1.20 
0.75 

1 
.91 
.05 
.42* 
-.37* 
.25* 

2 
 
.85 
.010 
-.01 
.15* 

3 
 
 
.77 
-.23* 
.53* 

4 
 
 
 
.95 
-.04 

5 
 
 
 
 
.92 

Table 2.1. Descriptive Statistics and Correlations for Study Variables of Final Sample (n = 185) 

Note:  *p <.05; Cronbach’s alpha coefficients along diagonal. 

 

Cluster analysis 

Standardized scores for doping MD, descriptive norms, and subjective norms were used 

to conduct the cluster analysis. Examination of the agglomeration coefficients resulting from the 

hierarchical analysis showed a notable increase when moving from five clusters to four clusters, 

from four clusters to three clusters, and from three clusters to two clusters. This suggested the 

appropriateness of a five, four, or three cluster solution. Interestingly, the five-cluster solution 

produced a cluster with a single outlier case. This case was removed—creating a final sample of 

185—and the resulting four-cluster solution was retained. 

In the k-means (non-hierarchical) cluster analysis, four non-redundant clusters with 

distinct characteristics were produced. Means, standard deviations, and standardized scores for 

cluster variables (doping MD, descriptive norms, and subjective norms) can be found in Table 2, 

with visual representation in Figure 1. A z-score at least 0.5 above or below the mean represented 

relatively high or low scores, respectively. These scores would be considered relatively high or 

low, regardless of whether they correspond to low or high scale scores. For example, cluster 1 

was characterized by higher scores in doping MD despite its mean falling near the middle of the 

DMDS (M = 3.35). 

21 
 

 

Figure 2.1. Results of k-means cluster analysis (N = 185) 

 

Cluster 1 (n = 34) was characterized by relatively high scores on doping MD, with 

descriptive norms and subjective norms values around the mean. Individuals in this group may 

be described as using mechanisms of moral disengagement to a greater extent, but not perceiving 

a greater or lower acceptance or prevalence of doping than the other clusters. Cluster 2 (n = 44) 

was characterized by relatively high descriptive norms, with doping MD and subjective norms 

values around the mean. Individuals in this group may be described as perceiving doping to be 

more prevalent, but do not see it as accepted and do not disengage their moral reasoning any 

more or less than other groups. Cluster 3 (n = 15) was characterized by relatively high scores on 

all three variables. Individuals in this group may be described as perceiving doping to be 

accepted and prevalent, and having greater use of doping MD mechanisms. Cluster 4 (n = 92) 

was characterized by relatively low doping MD and descriptive norms, with subjective norms 

clustering near the mean. Individuals in this group would be described as perceiving low 

prevalence of doping, as well as little use of doping MD mechanisms. 

22 
 

 
n 

34 

44 

15 

92 

 
Cluster 

1 

2 

3 

4 

 

Doping MD 
M 
z 
(SD) 
3.41 
(0.63) 
1.97 
(.67) 
3.24 
(1.20 
1.75 
(0.51 

-0.29 

1.01 

1.19 

-0.51 

-.40 

Descriptive Norms 
M 
z 
(SD) 
16.16 
(12.36) 
49.66 
(16.54) 
37.41 
(20.23 
12.74 
(9.17) 

-0.56 

1.19 

0.61 

Subjective Norms 

M 
(SD) 
1.51 
(0.51) 
1.14 
(.299) 
3.27 
(0.76) 
1.16 
(0.38) 

z 

0.11 

-0.35 

2.34 

-0.32 

Table 2.2. Participant Numbers, Means, Standard Deviations, and Standardized Scores for 

Cluster Characteristics  

 

Group difference analyses 

We conducted a one-way MANOVA to assess potential cluster differences in anticipated 

guilt and doping intentions. The analysis yielded a significant multivariate effect, Pillai’s trace 

= .355, F (6, 352) = 12.662, p < .001, ηp

2 = .178. Follow-up ANOVAs yielded cluster differences 

in both guilt and doping intentions. Scheffe post hoc comparisons were conducted to assess the 

nature of these differences. In regard to anticipated guilt, clusters 2 (M = 4.08) and 4 (M = 4.11) 

scored significantly higher than clusters 1 (M = 3.10) and 3 (M = 2.80), with cluster 4 scoring the 

highest and cluster 3 scoring the lowest. There was no significant difference between cluster 1 

and 3, nor was there a significant difference between clusters 2 and 4. Concerning doping 

intentions, only cluster 3 (M = 2.19) showed a significant difference from the other three clusters. 

This cluster showed significantly higher scores than clusters 1, 2, and 4 (M = 1.24, 1.10, and 

1.02, respectively). All these differences held when using Dunnett’s T3 and Dunnett’s C tests, 

which adjusted for unequal variances and unequal sample sizes across groups. 

23 
 

Cluster 

 
 
Variable 

Anticipated 
Guilt 
Doping 
Intentions 

 
 
F 

(3, 176) 
11.46* 

14.77* 

 

1 

2 

 
 
M 
 
(SD) 
ηp
.16  3.10a 
(0.19) 
.20  1.24a 
(.11) 

 
z 

-0.61 

0.11 

2 

M 
(SD) 
4.08b 
(0.17) 
1.10a 
(0.10) 

 
z 

0.24 

-0.10 

M 
(SD) 
2.80a 
(0.30) 
2.19b 
(0.17) 

3 

 
z 

-0.87 

1.48 

4 

M 
(SD) 
4.11b 
(0.12) 
1.02a 
(0.07) 

 
z 

0.25 

-0.22 

Table 2.3. MANOVA Results for Cluster Differences  

Note: *p < .05; Cluster differences based on pairwise comparison indicated by distinct 

superscripts (a represents lower value, b represents higher value). 

 

Discussion 

The first purpose of this study was to examine patterns of three cognitive antecedents of 

doping outcomes in disability sport athletes: doping MD, descriptive norms, and subjective 

norms. Though these variables have been studied in the literature, they are not often examined 

together, and typically only in able-bodied athletes. Thus, the current study contributes to the 

literature on doping in multiple ways. As hypothesized, there were a finite number (i.e., four) of 

distinct profiles grouped by relative scores on the variables of interest. These profiles emerged 

across multiple approaches to cluster analysis. One high-risk profile (i.e., Cluster 3) was 

observed, and was characterized by relatively high scores on doping MD, descriptive norms, and 

subjective norms. One low-risk profile (i.e., Cluster 4) was observed, and was characterized by 

relatively low scores on all these variables. Two mixed profiles (i.e., Clusters 1 and 2) were also 

observed that were characterized by relatively high scores on one variable (i.e., doping MD and 

descriptive norms, respectively), but not the others. 

24 
 

The second purpose of this study was to assess meaningful distinctions between clusters 

of cognitions on two outcomes: anticipated guilt and doping intentions. Clusters 1 and 3 showed 

significantly lower scores on the outcome of anticipated guilt than clusters 2 and 4. Because 

clusters 1 and 3 were characterized by higher levels of doping MD, results support previous 

research suggesting the negative association between doping MD and anticipated guilt from 

doping (Boardley et al., 2017; Ring & Kavussanu, 2017). Although anticipated guilt was 

significantly lower in the clusters characterized by higher doping MD, these clusters were not 

necessarily characterized by higher descriptive and subjective norms. These findings suggest 

anticipated guilt from doping is related to internal moral cognitions regarding the act, 

independent of the perceived social norms surrounding the behavior. Cluster differences in 

intentions could provide further context to the influence of social norms. Cluster 3 was 

characterized by relatively higher subjective norms than the other three clusters. This same 

cluster reported significantly higher intentions to dope than the other three clusters. This supports 

previous research suggesting the prediction of doping intentions by perceived social norms 

(Ntoumanis et al., 2014). Group differences in line with theory supported the predictive value of 

the clusters, providing robust evidence of the profiles representing actual group structures (Hair 

et al., 1998). 

Aside from the aforementioned support for the hypotheses, this study could provide some 

context for the social-cognitive theory of moral thought and action (Bandura, 1991). Bandura 

suggested internalization of social sanctions can occur over time, and these self-sanctions are 

what influence behavior. Because subjective norms did not appear to be associated with 

anticipated guilt (r = -.23, p < .05), this suggested a lack of an internalization in these athletes. 

Furthermore, anticipated guilt and intentions were not related presently. This initially seemed to 

25 
 

refute the social-cognitive theory of moral thought and action, but qualitative research with 

athletes who have a history of doping suggests there is a limited group of people these athletes 

will be honest with about their PED use (Boardley et al., 2014). In other words, athletes may 

choose to avoid sharing their doping decisions with those who would stimulate the negative self-

emotions brought on by the behavior. Therefore, they may surround themselves with friends or 

family who are supportive of the behavior, completing a self-fulfilling prophecy and increasing 

the perceived doping norms. Moreover, surrounding oneself with colleagues who use 

performance enhancing substances or methods may expose the athlete to mechanisms of doping 

MD. This may best explain the findings in the current study, and could describe a possible 

dynamic based on the combination of variables observed in cluster 3, the most at-risk group. 

Cluster 3, the smallest cluster (n = 15), was characterized by relatively high scores on all 

three variables. Considering the association of these variables with doping-related outcomes, the 

small number of cases in this cluster should be viewed optimistically. According to previous 

research, higher doping MD is associated with lower anticipated guilt (Boardley et al., 2017) and 

greater doping susceptibility (Boardley, Smith, Ntoumanis, Gucciardi, & Harris, 2019), and 

higher perceived norms are associated with higher doping intentions and behavior (Ntoumanis et 

al., 2014). It may be good to know that the cluster with high levels on all three of these variables 

was the smallest cluster, at least in a disability sport setting. Furthermore, even if the intentions 

of the members of this cluster are relatively high, there is still an intention-behavior gap 

(Sheeran, 2002) that would need to be overcome before these athletes engage in doping behavior. 

The exclusive recruitment of disability sport athletes offered a broader picture of the 

doping environment. Differences in antisocial behavior between athletes with spinal cord injury 

and able-bodied athletes have previously been explained by use of MD mechanisms, despite 

26 
 

similarities in empathy and negative emotion (Kavussanu et al., 2015). Emotional responses may 

therefore not be enough to promote or deter doping. In the current study, the most at-risk group 

was characterized by high doping MD as well as high perceived social norms. Among athletes 

with physical disabilities, social support is a significant predictor not only of physical activity 

levels (Martin Ginis et al., 2011) but also of competitive level of participation (Stapleton, Perrier, 

Campbell, Tawse, & Martin Ginis, 2016). Subjective norms are a proxy for social support of a 

particular behavior, and thus may help to understand its association with doping intentions in 

disability sport athletes. Athletes, both disabled and able-bodied, appear to be willing to put more 

effort into doping if they have the support of the people around them. 

Although the current study exclusively recruited athletes with physical disabilities, the 

measures in this study were originally developed for use with able-bodied athletes. This may 

have led to issues with interpretation of some items—particularly, descriptive norms. Two of the 

three descriptive norms items ask the participant about their perceived prevalence of doping in 

“elite” athletes. Research by Tasiemski, Kennedy, Gardner, & Blaikly (2004) has demonstrated 

that individuals with an acquired physical disability (i.e., spinal cord injury) have lower levels of 

athletic identity than what is reported by the general population. It is difficult to decipher 

whether the word “elite” evoked an image of high-level disability sport athletes or their able-

bodied counterparts. How the athletes interpreted these items could have confounded responses 

when measuring perceived prevalence of doping. Therefore, reliability across athletic contexts 

should be taken into consideration when interpreting results in the descriptive norms measure. 

Despite the use of measures originally developed for able-bodied athletes, the athletes in 

this study reported scores similar to previous research. Mean doping MD in this study (M = 2.23) 

was comparable to a study conducted with Olympic-style sport athletes (M = 2.14; Boardley et 

27 
 

al., 2019). The perceived prevalence of doping in disability sport reported in this study (M = 

24.15) fell within the range reported in studies with student-athletes (Petróczi, Mazanov, Nepusz, 

Backhouse, & Naughton, 2008) and Hungarian competitive athletes (Uvacsek et al., 2011). 

Lastly, athletes in this study reported low levels of subjective norms (M = 1.39), which was 

comparable to what was reported by elite athletes in previous research (M= 1.45) by Lazuras et 

al. (2010).  

Implications 

Athletes in cluster 3 reported significantly higher doping intentions than any other group, 

and were also in one of the two groups with the lowest reported anticipated guilt from doping. 

This puts these athletes in a pivotal position in anti-doping efforts. Not only are these athletes at 

risk for intentional doping, but they are also inclined to feel less guilty if they do end up using 

prohibited substances or methods to enhance performance. A multi-functional view of moral 

disengagement (Tillman, Gonzalez, Whitman, Crawford, & Hood, 2018) considers MD to occur 

at multiple stages, both before and after a behavior is performed. Whether this multi-stage, multi-

functional view of MD describes the members in cluster 3, thus providing theoretical insight into 

the MD process, needs to be evinced through further investigation. 

Regardless of the exact relationship between doping MD, perceived norms, anticipated 

guilt, and doping intentions, there are practical implications to take from the current study. Anti-

doping education programs such as the ATLAS and ATHENA programs (Barkoukis et al., 2016) 

have support in that they reduce adolescents’ positive attitudes toward prohibited substances, but 

research has failed to establish a causal relationship between doping attitudes and doping 

behavior (Backhouse et al., 2007). Due to the implications doping MD can have on doping 

related outcomes, anti-doping education programs may benefit from making the athletes aware 

28 
 

of possible mechanisms that could be used to endorse a transgressive behavior like doping, thus 

decreasing the likelihood they would use these mechanisms to justify their own future, present, 

and past doping behavior. Another option might be to address antecedents of doping MD, such as 

moral identity (Kavussanu & Ring, 2017) or basic values (Ring, Kavussanu, & Gürpinar, 2020), 

with an expectation this may decrease the use of doping MD in these athletes. 

Identifying which cluster an athlete belongs to may also help with anti-doping messaging 

campaigns. Tailored health communication (Rimer & Kreuter, 2006), in which health-related 

messaging is specific to the characteristics or needs of the individual or group, has found 

moderate success in improving multiple health-related behaviors and in various populations 

(Lustria et al., 2013). This style of health communication may be beneficial to anti-doping 

interventions that implement smartphone applications rather than in-person educational programs 

(Nicholls et al., 2020). Using web-based or smartphone media would allow anti-doping 

organizations to efficiently tailor messages to those athletes who need these messages most, such 

as those in cluster 3 of the current study. 

Limitations 

There were limitations in this study, which may have stemmed from using measures 

developed for able-bodied athletes and applying them to a sample of athletes with physical 

disabilities. Besides the considerations previously discussed, our definition of “doping” may not 

have covered the nuanced context of disability sport. Although the definition of doping included 

in the survey contained the phrase “prohibited substances or methods,” the perception of what 

constituted a “method” may have been ambiguous in our sample. Possible methods unique to the 

disability sport context are falsifying ability/disability level during sport classification and 

boosting (i.e., artificial induction of autonomic dysreflexia, which has been suggested to enhance 

29 
 

performance) in athletes with high-level (T6 or above) spinal cord injury (Legg & Mason, 1998; 

Mazzeo, Santamaria, & Iavarone, 2015). It is difficult to know whether these methods were 

acknowledged by athletes when completing the survey. 

What is “prohibited” may also be subject to interpretation. Athletes with a high-level 

spinal cord injury exhibit dysautonomic disorders (e.g., low blood pressure and heart rate during 

physical exertion) associated with poor sport performance (Bhambhani, 2002). Boosting is 

therefore perceived by some athletes as necessary simply to “normalize” or remain competitively 

“relevant” (Sparkes & Brighton, 2019). This perception, in combination with the belief that 

boosting often occurs by accident, provides possible justification for the practice. At a policy 

level, boosting is banned by the International Paralympic Committee (IPC) but is not listed as a 

doping method by the World Anti-Doping Agency (WADA; Mazzeo et al., 2015). Such 

ambiguity may have affected how athletes in this study perceived whether a method is 

“prohibited.” These issues are more relevant to the disability sport context than traditional 

definitions of doping, and future researchers may benefit from making distinctions between 

doping substances or methods that are exclusive to the disability sport context. In the meantime, 

caution should be taken when making any generalizations with the findings of the current study. 

Further research should also investigate other doping-related outcome variables, such as 

doping behavior or susceptibility to inadvertent doping. Self-reported measurement of doping 

behavior comes with limitations of its own, but studying more distal outcomes can provide better 

understanding of the phenomenon and minimize the intention-behavior gap that is often seen in 

sport and exercise research (Sheeran, 2002).  

30 
 

Conclusion 

Disability sport athletes’ cognitions surrounding doping are similar to those of their able-

bodied counterparts. The current study suggests there may be a small group of athletes whose 

normative and moral cognitions may put them at risk of doping and minimizing self-sanctions 

that typically follow transgressive behavior. Researchers are advised to be aware of the unique 

doping context in disability sport, and a deeper understanding of the distinctions between the 

different contexts of sport can offer insight into the best anti-doping strategies for that particular 

group of athletes. 

 

 

 

 

 

 

31 
 

CHAPTER THREE: STUDY TWO 

DOPING MORAL DISENGAGEMENT PREDICTS ANTICIPATED GUILT AND DOPING 

CONSIDERATION WITHIN THE CONTEXT OF SOCIAL NORMS 

Abstract 

Doping Moral Disengagement (MD) is suggested to decrease feelings of anticipated guilt 

from doping (Boardley et al., 2017) and increase the consideration one would give to doping 

(Boardley et al., 2019). These cognitions do not occur in isolation and must be understood within 

the social context. Subjective norms (perceived social approval or support for doping) and 

descriptive norms (perceived doping prevalence) are also suggested to influence doping-related 

outcomes (Ntoumanis et al., 2014). The purpose of this study was to test a conceptual model that 

specifies these three constructs (doping MD, subjective norms, and descriptive norms) to predict 

doping consideration, directly and indirectly through anticipated guilt. Athletes from various 

sports (N = 238) responded to a survey measuring doping MD (Boardley et al., 2018), subjective 

norms (Lazuras et al., 2010), descriptive norms (Barkoukis et al., 2014), anticipated guilt 

(Boardley et al., 2017), and consideration of doping in a hypothetical situation (Gucciardi et al., 

2010). Structural equation modeling supported the direct and mediated relationship from doping 

MD to doping consideration, but no such relationships were found for subjective or descriptive 

norms. The direct effects of MD were robust against socially desirable responding and removal 

of outliers. Results therefore suggest doping MD to be a dominant predictor of doping-related 

cognitions. Anti-doping educational programs, which typically do not address doping MD and 

are sub-optimally effective (Backhouse et al., 2007), may benefit from deliberately targeting this 

construct.  

 

 

32 
 

Introduction 

 

Sport can be thought of as a context distinct from everyday life in both people’s moral 

reasoning (Bredemeier & Shields, 1984) and moral behavior (Kavussanu, Boardley, Sagar, & 

Ring, 2013). This is evidenced by athletes being more likely to perform antisocial behavior (e.g., 

intimidating or injuring an opponent) in the sport context than other contexts. Antisocial 

behavior is defined as behavior that is intended to harm or disadvantage another (Sage, 

Kavussanu, & Duda, 2006). One form of such behavior is doping, defined as the use of 

prohibited substances or methods with the potential to artificially improve performance through 

changes in physical or mental condition (IPC, 2018; Lauré, 1997; WADA, 2021). Despite some 

authors having argued that doping is not inherently immoral (Arandjelović, 2015), it contravenes 

the rules of sport and athletes use morality-related cognitions when making decisions whether or 

not to use performance enhancing substances or methods. For example, athletes may use 

cognitive mechanisms to conditionally endorse their transgressive behavior, a process referred to 

as moral disengagement (MD; Bandura, 1991). 

 

There are eight mechanisms used in MD. Moral justification involves making the 

behavior seem morally or socially acceptable by offering a way in which it can do some good. 

Euphemistic labeling uses benign language to distort the perceived reprehensibility of the 

behavior. Advantageous comparison is when the transgressor contrasts the behavior with a more 

outrageous flagrant offense. Displacement of responsibility is putting the responsibility on some 

other authoritative decision-maker. Diffusion of responsibility is spreading the responsibility of 

the decision to act to a group in order to minimize individual responsibility. Distortion of 

consequences involves minimizing or ignoring the negative ramifications of the behavior. 

Dehumanization removes the victim’s human qualities in order to minimize empathetic feelings 

33 
 

toward them. Finally, attribution of blame places the fault with the victim, as if their actions 

provoked the transgressor. Research in the sport area suggests six of these mechanisms of MD 

(the exceptions being dehumanization and attribution of blame) are relevant to doping (Boardley 

et al., 2018). 

 

Athletes may anticipate feeling guilty for deciding to use performance enhancing 

substances or methods. The social-cognitive theory of moral thought and action (Bandura, 1991) 

suggests individuals may have negative affective responses when behaving immorally, and 

therefore will act morally in order to avoid these negative emotions. MD acts as a method to 

circumvent the anticipation of guilt in response to the behavior. In terms of doping, an athlete 

may use doping MD to circumvent the anticipated feelings of guilt they expect from engaging in 

performance enhancing drug use. Empirical research has found doping MD to inversely associate 

with anticipated guilt, which itself is inversely associated with outcomes such as doping 

likelihood (Kavussanu & Ring, 2017). Thus, there is support in the literature that a moral 

cognition such as doping MD factors into athletes’ doping decisions. 

 

Individual moral cognitions occur within, and are likely to be shaped by, social 

environmental factors. In fact, the self-sanctions one feels after committing a transgression are 

suggested to be a consequence of having observed social sanctions for the transgression over 

time (Bandura, 1991). It is therefore important to consider the social environment, particularly 

normative influences, when examining cognitions such as doping MD. The two most relevant 

normative influences in terms of doping are subjective norms and descriptive norms. Subjective 

norms are the perceived pressure by significant others to perform, or to not perform, a particular 

behavior (Ajzen, 1991). In terms of doping, this is the athlete’s perception of whether people 

they care about would support or approve of their doping behavior.  

34 
 

 

Descriptive norms are distinct from subjective norms in that they are the perception of the 

actual behavior of others. In doping terms, this is the athlete’s perception of the prevalence of 

doping in their sport. A meta-analysis on the theory of planned behavior suggested that 

descriptive norms contribute additional explanation of the variance in various social and health 

behaviors beyond the original variables of the theory of planned behavior, which included 

subjective norms (Rivis & Sheeran, 2003). Furthermore, a doping-specific meta-analysis 

suggested that both subjective norms and descriptive norms significantly predict outcomes such 

as doping intentions and doping behavior (Ntoumanis, Ng, Barkoukis, & Backhouse, 2014).  

 

Moral conduct is reciprocally regulated by the interaction of self-sanctions and social 

influences (Bandura, 1991), suggesting that norms-based cognitions should be simultaneously 

considered when studying a morality-based cognition like doping MD. Despite the need for 

research that considers socially-driven cognitions alongside internal cognitions, normative 

influences are typically not measured in combination with doping MD. Some studies 

emphasizing doping MD have examined it alongside subjective norms exclusively (Lucidi et al., 

2008), whereas studies emphasizing descriptive norms have included measures of personal 

morality or sportspersonship but not doping MD (Barkoukis, Lazuras, Tsorbatzoudis, & 

Rodafinos, 2013; Gucciardi, Jalleh, & Donovan, 2011). A simultaneous test of doping MD, 

subjective norms, and descriptive norms (which have all independently been shown to predict 

doping outcomes) would offer a fuller picture of the extent to which internalization of social 

doping sanctions has occurred in athletes. This picture could provide a deeper understanding of 

which variables are most salient to athlete doping outcomes. 

 

Doping outcomes of interest have plagued doping researchers, because objective and 

subjective measures both present setbacks. Objective measures, which more or less are the same 

35 
 

as drug-testing procedures to detect doping athletes, are not always successful at detecting the 

presence or absence of prohibited substances (Kayser, Mauron, & Miah, 2007; Strelan & 

Boeckmann, 2006). Furthermore, survey-based research is susceptible to false reporting of actual 

doping behavior. The most common solution to this problem has been to instead ask respondents 

to imagine a hypothetical scenario in which they have the opportunity to use a performance 

enhancing substance. Such hypothetical scenarios are popular because they factor in 

environmental variables from other theories such as availability (Donovan, Egger, Kapernick, & 

Mendoza, 2002; Petróczi & Aidman, 2008) or situational temptation (Lazuras, Barkoukis, 

Rodafinos, & Tzorbatzoudis, 2010), and because they promote honest responding by avoiding 

self-incrimination. Outcome variables measured with this method have included doping 

likelihood (Kavussanu & Ring, 2017) and doping susceptibility/consideration (Boardley, Smith, 

Ntoumanis, Gucciardi, & Harris, 2019). Whereas doping susceptibility was originally 

operationalized as any level of doping consideration (Gucciardi, Jalleh, & Donovan, 2010), later 

research used the same measure in a Likert-type format (Boardley et al., 2019). For the sake of 

clarity, the measure is referred to as doping consideration from here forward. Doping 

consideration is particularly important in the current study, because doping MD, descriptive 

norms, and subjective norms may all be integrated into the decision-making process and be 

associated with the amount of consideration the athlete might give to an opportunity to illegally 

enhance performance.  

The purpose of the current study was to examine athlete doping-related cognitions within 

the greater context of the social environment of sport. Specifically, this study tested a conceptual 

model in which the three variables of doping MD, subjective norms, and descriptive norms 

predict doping consideration, directly and indirectly through anticipated guilt (see Figure 3.1). 

36 
 

Based on previous doping literature, it was hypothesized that doping MD, subjective norms, and 

descriptive norms would be positively associated with each other, and all have independent 

negative associations with anticipated guilt from doping; that doping MD, subjective norms, and 

descriptive norms would have independent positive associations with doping consideration; and 

that anticipated guilt would be negatively associated with doping consideration. Finally, the 

associations between the three predictor variables and doping consideration were hypothesized to 

be explained by significant mediation through anticipated guilt. 

 

 

Figure 3.1. Hypothesized model for proposed Study 2 

Note: MD = moral disengagement 

 

Participants 

Method 

 

Participants included a convenience sample of high-level athletes from the USA (N = 

238) who were recruited between April and October of 2020. The sample was 59.2% female and 

40.8% male, with ages ranging from 17 to 82 years (M = 27.86, SD = 14.72). Ethnic makeup of 

37 
 

the sample included 5.0% affirming Hispanic or Latino ethnicity, with races reported as 85.7% 

White, 6.8% more than one race, 3.4% Black or African-American, 2.5% Asian, 1.3% 

unknown/other, and 0.4% American Indian/Alaskan Native. Athletes were currently competing 

at the regional (n = 89), national (n = 104), or international (n = 45) level in endurance (n = 144) 

and non-endurance (n = 94) sports. Sports in which the athletes participated were swimming 

(26.6%), triathlon/multisport (22.4%), lacrosse (11.4%), cross country (9.7%), volleyball (5.9%), 

track and field (5.5%), football (2.5%), tennis (2.5%), baseball/softball (2.1%), cycling (2.1%), 

diving (2.1%), bobsled/skeleton (1.7%), basketball (1.3%), golf (1.3%), other (1.3%), 

weightlifting (0.8%), curling (0.4%), and wrestling (0.4%). On average, athletes had been 

competing in their sport for 11.09 years (SD = 14.72), had been members of their current team 

for 3.82 years (SD = 4.87), and trained for 17.17 hours (SD = 6.55) a week for 42.99 weeks (SD 

= 11.25) out of the year. 

Instruments 

The participants completed surveys assessing demographic information (i.e., age, sex, 

ethnicity/race, sport, team/club/school, level of competition, years in the sport, training hours per 

week, training weeks per year) and the core study variables (see Appendix D). A definition of 

doping was provided to athletes prior to the appearance of doping-related instruments: “Doping 

is defined as use of prohibited substances or methods by an athlete with the potential to 

artificially improve performance through changes in physical and/or mental condition.” Further 

details follow for each instrument used to assess the core study variables.  

Doping moral disengagement. 

Doping MD was measured by administering the Doping Moral Disengagement Scale 

(DMDS; Boardley et al., 2018) and retaining the items from the short version of the instrument 

38 
 

(DMDS-S) for analysis. The DMDS-S was selected because of its versatility to measure the 

construct of doping MD in both team and individual sports. Additionally, development of the full 

measure evinced six dimensions relating to the six MD mechanisms observed in sport doping, 

collectively representing one higher-order factor: doping MD. The full measure consists of 18 

items (e.g., “Doping doesn’t really harm anyone else.”), with three items for each of the six 

dimensions of doping MD. The short version includes one item for each of the six dimensions, 

making for a six-item measure. The short version of the scale was chosen for this study based on 

its robustness and utility in previous research as well as to allow for a simpler model to be 

estimated. Participants responded with their level of agreement with each statement on a scale 

ranging from 1 (strongly disagree) to 7 (strongly agree). Internal consistency reliability of scores 

for the DMDS-S was strong in the developmental stage of the measure, with Cronbach’s alpha 

values of .86 in one sample and .89 in another (Boardley et al., 2018). Internal consistency 

reliability of scores in the present study was acceptable, with a Cronbach’s alpha coefficient 

of .72. 

Subjective norms. 

Subjective norms were measured using three items (e.g., “Most people who are close to 

me would like me to use doping substances or methods to enhance my performance.”) scored on 

a 1 (strongly disagree) to 7 (strongly agree) scale (Lazuras et al., 2010; Lazuras, Barkoukis, & 

Tsorbatzoudis, 2015). The phrase “or methods” was added to the original items to fully map to 

the definition of doping provided to participants in the current study. Greater scores on this 

measure represented greater perceived social pressure to use performance enhancing substances 

or methods. Previous research has shown scores on these items to exhibit good internal 

consistency reliability, with Chronbach’s alpha coefficients of .84 (Lazuras et al., 2010) and .81 

39 
 

(Lazuras et al., 2015). Internal consistency reliability of scores in the present study was 

comparable, with a Cronbach’s alpha coefficient of .85. 

Descriptive norms. 

Descriptive norms were measured using three items (Barkoukis, Lazuras, & 

Tsorbatzoudis, 2014). All items were measured on a 0-100 scale, as they relate to perceived 

prevalence (e.g., “Out of 100%, how many athletes at your competitive level do you believe 

engage in doping to enhance their performance?”). The first item was in relation to athletes at the 

competitive level of the respondent and the latter two asked for perceived prevalence in relation 

to elite athletes in the United States. Internal consistency reliability information on scores from 

this measure have not been reported in previous research (Lazuras et al., 2010; Lazuras et al., 

2015) using these specific items. Internal consistency reliability of scores in the present study 

was good, with a Cronbach’s alpha value of .86. 

Anticipated guilt. 

Anticipated guilt was measured using five items from the State Shame and Guilt Scale 

(Marschall, Saftner, & Tangney, 1994) adapted to a hypothetical doping situation (Boardley et 

al., 2017). Participants were asked to read the following scenario:  

Having returned to training following a period of injury, you are feeling very out of 

shape. As such, you feel the need to get back in shape as soon as possible. A friend who you train 

with has been taking a training supplement that he/she says really helped him/her get back in 

shape quickly following a similar injury. Your friend offers to give you some and you decide to 

take it. Subsequently you get back in shape much quicker than expected, but then discover the 

supplement you have been taking is a banned performance-enhancing substance. However, due 

to the improvements you have experienced, you decide to continue taking the substance. 

40 
 

After reading this scenario, the participant was asked to respond to five items rating the 

level at which they would feel particular aspects of guilt (e.g., “I would feel bad about what I 

was doing.”). Responses were on a five-point scale ranging from 1 (not at all) to 5 (extremely), 

with greater scores reflecting greater anticipated guilt. Previous research using this measure 

showed scores to exhibit good internal consistency reliability, with a Cronbach’s alpha 

coefficient of .95 (Boardley et al., 2017). Internal consistency reliability of scores in the present 

study was good, with a Cronbach’s alpha value of .89. 

Doping consideration. 

A single item was used to measure doping consideration. First, participants were asked to 

read the following scenario in which they are offered an opportunity to dope: 

 If you were offered a banned performance‐enhancing substance under medical 

supervision at low or no financial cost and the banned performance‐enhancing substance could 

make a significant difference to your performance and was currently not detectable. 

Following the scenario, athletes were asked to respond to the item “How much 

consideration would you give to the offer?” on a seven-point scale ranging from 7 (a lot of 

consideration) to 1 (none at all). Higher scores indicated greater consideration of doping. 

Previous research has supported the validity of this measure for examining doping susceptibility 

(Gucciardi, Jalleh, & Donovan, 2010). No internal consistency reliability statistic was calculated 

for this measure because it consists of a single item. 

Social desirability. 

 

In line with previous doping research studies (Barkoukis et al., 2014; Gucciardi et al., 

2010; Kavussanu et al., 2016; Lazuras et al., 2015), we included a measure of social desirability. 

While not essential to address the focal purpose of the current study, it was deemed something to 

41 
 

consider through supplemental analysis. Therefore, the seventeen-item Social Desirability Scale 

(SDS-17; Stöber, 2001) was administered after demographic information and prior to doping-

relevant instruments. This scale presented True/False items that may elicit socially desirable 

responding (e.g., “In traffic I am always polite and considerate of others.”). True is scored as 1 

and False is scored as 0, with the exception of six reverse-coded items (items 1, 4, 6, 7, 11, 15, 

and 17). Scale total was used for supplemental analysis, with higher scores indicating more 

socially desirable responding. Previous research has shown scores on these items to exhibit 

acceptable internal consistency reliability, with Chronbach’s alpha coefficients ranging from .61 

to .84 (Stöber, 2001). Internal consistency reliability of scores in the present study was 

suboptimal, with a Cronbach’s alpha value of .64. 

Procedure 

After obtaining IRB approval, participants were recruited to take an online survey using 

Qualtrics software. An online method was employed because of restrictions to in-person research 

during the COVID-19 pandemic. Coaches, directors, and administrators were contacted by phone 

or email using publicly available contact information or through personal contact with the 

primary investigator to describe the purpose of the study. If the contact was amenable to their 

athletes completing the survey, the investigator shared a link to the online survey for the coach, 

director, or administrator to distribute to their athletes (see Appendix D). When the link was 

followed, the first page of the site presented study information to the athletes before choosing to 

continue. Completing the survey after receiving this written information constituted informed 

consent to participate. 

42 
 

Data Analysis 

 

After data screening was completed, descriptive statistics, bivariate correlations, and 

reliabilities were calculated for all variables. To address the purpose of the study, latent variable 

modeling was employed. Specifically, structural equation modeling (SEM) was conducted using 

Mplus (Muthén & Muthén, 1998-2015) to assess the fit of the hypothesized model to the data 

and to test the specified direct and indirect relationships. The analysis used a maximum 

likelihood robust (MLR) estimator (Yuan & Bentler, 2000). Chi-square (χ2), comparative fit 

index (CFI), Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA), and 

standardized root mean square residual (SRMR) assessed model-data fit. Values greater than 

0.95 and 0.90 for the CFI and TLI and less than 0.06 and 0.08 for RMSEA indicated excellent 

and adequate model fit, respectively (Hu & Bentler, 1999). An SRMR value less than .08 was 

also considered good fit. In line-with the two-step approach to SEM (Anderson & Gerbing, 

1998), the measurement model was examined first, followed by testing of the structural model. 

Descriptive Statistics 

Results 

 

Qualtrics software received a total of 406 recorded responses. Of these, 121 completed 

zero items (i.e., followed the link to the survey, but did not proceed beyond the information and 

consent page), and another 47 did not complete any items in the core study variables. The 

remaining 238 responses were retained for the analysis of this study. Initial screening examined 

the data for missing values, violations of assumptions, and outliers (Tabachnick & Fidell, 2013). 

In the missing data analysis, it was found that 1.15% of all items were missing. Furthermore, 

data were considered to be missing completely at random as evidenced by Little’s MCAR test, χ2 

= 698.181, p = .908. When this is the case, there are a number of viable missing data analysis 

43 
 

techniques. The default techniques of the statistical software were used in both the correlation 

(pairwise deletion in SPSS) and latent variable (full information maximum likelihood in Mplus) 

analyses. 

Criteria for deviations from normality were based on levels provided by Finney and 

DiStefano (2013), in which absolute values of skew > 2 and kurtosis > 7 may present issues with 

maximum likelihood (ML) analyses. Based on these criteria, only subjective norms presented 

deviations from normality, with a skewness of 4.025 and kurtosis of 21.489. Leptokurtic 

distributions (positive kurtosis) pose the risk of rejecting a correctly specified model and 

overestimating the significance of model parameters (Finney & DiStefano, 2013). In light of 

these risks, a robust analysis was used. As will be described in the results, neither of these issues 

arose in the analysis—the model fit the data well and subjective norms were not significant 

predictors of the endogenous variables. 

 

Outliers were determined to be any score that fell outside of 3.29 standard deviations 

from the mean of the variable. A total of eight univariate outliers were found. However, these 

values may have offered a true representation of a small number of participants (as seen in Study 

1). Therefore, these outliers were not removed for further analysis. For the sake of being 

thorough, an analysis was conducted for correlations and the hypothesized model with these 

outliers removed. Results of these analyses can be found in the supplemental analysis.  

Means, standard deviations, and correlations were calculated for all variables (see Table 

3.1). Overall, relative to response set options the athletes reported moderately low levels of 

doping MD (M = 2.27, SD = 0.95), low levels of subjective norms (M = 1.33, SD = 0.81), 

moderately high levels of anticipated guilt (M = 4.24, SD = 0.80), and low doping consideration 

(M = 2.24, SD = 1.62). They perceived doping on average to be 22.28% (SD = 18.00). Doping 

44 
 

MD showed significant positive correlations with subjective norms (r = .423) and doping 

consideration (r = .389), and a significant negative correlation with anticipated guilt (r = -.491). 

Additionally, Subjective norms showed a significant positive correlation with consideration (r 

= .223) and a significant negative correlation with anticipated guilt (r = -.263). A significant 

negative correlation was found between anticipated guilt and doping consideration (r = -.413). 

There were no significant correlations with descriptive norms. 

 

Variable 
1. Doping MD 
2. Descriptive Norms 
3. Subjective Norms 
4. Anticipated Guilt 
5. Consideration 
Table 3.1. Descriptive Statistics and Correlations for Study Variables of Final Sample (n = 238) 

1 
.72 
.004 
.423* 
-.491* 
.389* 

Mean 
2.27 
22.28 
1.33 
4.24 
2.24 

SD 
0.95 
18.00 
0.81 
0.80 
1.62 

3 

.558* 
.004 
.85 
-.263* 
.223* 

5 

.425* 
.067 
.231* 
-.412* 

- 

2 
.079 
.86 
-.026 
-.081 
.073 

4 

-.593* 
-.064 
-.303* 
.89 
-.413* 

Note:  *p <.05; Cronbach’s alpha coefficients along diagonal, with zero-order Pearson 

correlations below and measurement model latent variable correlations above. 

 

Model Testing 

 

The measurement model was examined first, in line with the two-step approach. In this 

model, covariations between the latent variables of doping MD, descriptive norms, subjective 

norms, anticipated guilt, and the doping consideration item were freely estimated. The number of 

estimated parameters was the same in the measurement model and the structural model, therefore 

the goodness of fit indices were equal in both models. Thus, the fit indices will be reported once. 

Overall, the indices of these models indicated adequate to excellent fit, χ2 (126) = 193.055, p < 

.01, CFI = .946, TLI = .935, RMSEA = .047 (.033, .060), SRMR = .049. Factor loading values 

were generally acceptable, with a primary exception being from the DMDS-S item associated 

45 
 

with euphemistic labelling. In line with previous research that has supported and maintained the 

inclusion of this item, it was retained in later analyses. Factor loadings and error variances are 

shown in Table 3.2. 

Factor 

Item 

Factor 
Loading 

Error 

Variance 

.31 

.92 

.58 

.76 

.68 

.54 

.60 

.23 

.06 

.31 

.49 

.14 
.35 
.32 
.44 
.46 
.17 

Doping is okay if it helps an athlete advise others on how 
to do it right (moral justification). 
Using terms such as “gear” or “juice” makes doping sound 
less harmful (euphemistic labelling). 
Compared to most lifestyles in the general public, doping 
isn't that bad (advantageous comparison). 
Athletes shouldn't be blamed for doping if training 
partners/teammates pressure them to do it (displacement of 
responsibility). 
It's not right to condemn individuals who dope when many 
in their sport are doing the same (diffusion of 
responsibility). 
Risks associated with doping are exaggerated (distortion of 
consequences). 
Out of 100%, how many athletes at your competitive level 
do you believe engage in doping to enhance their 
performance?   
Out of 100%, how many elite athletes in the USA do you 
think engage in doping to enhance their performance?   
Out of 100%, how many elite athletes do you believe will 
be engaged in doping during the next 3 years to enhance 
their performance? 
Most people who are important to me would want me to 
use doping substances or methods to enhance my 
performance. 
Most people I know would agree if I used doping 
substances or methods to enhance my performance. 
Most people who are close to me would like me to use 
doping substances or methods to enhance my performance. 
I would feel remorse, regret. 
I would feel tension about what I was doing. 
I would not be able to stop thinking about the bad thing I 
was doing. 
I would feel like apologizing, confessing. 
I would feel bad about what I was doing. 

DMDS-S 

Descriptive 
Norms 

Subjective 
Norms 

Anticipated 
Guilt 

.83 

.28 

.65 

.49 

.56 

.68 

.63 

.88 

.97 

.83 

.72 

.93 
.81 
.82 
.75 
.74 
.91 

Table 3.2. Items, Standardized Factor Loadings, and Error Variances in Study 2 

Note:  DMDS-S = Doping Moral Disengagement Scale-Short.  

46 
 

 

In step two, the structural model was tested with the hypothesized relationships. In line 

with hypotheses, doping MD showed an inverse association with anticipated guilt and a positive 

association with doping consideration. Anticipated guilt also had an inverse association with 

doping consideration. Furthermore, there was a strong positive covariance between doping MD 

and subjective norms. No other estimated covariances or direct effects were found to be 

statistically significant (see Figure 3.2). Overall, the model accounted for 35.3% of the variance 

in anticipated guilt and 22.1% of the variance in doping consideration. 

 

The MODEL = INDIRECT command was used to examine the significance of 

mediational paths specified within the model. Doping MD indirectly predicted doping 

consideration via anticipated guilt (β = .151, p < .05, 95% CI = .018, .283). No other significant 

indirect effects were found in this analysis.  

 

Figure 3.2. Model testing results 

Note: MD = moral disengagement; significant associations denoted by asterisk and solid arrows, 

 

non-significant by dotted arrows 

 

47 
 

Supplemental Analyses 

Social desirability. 

 

Supplemental analyses were conducted to assess if social desirability was salient to our 

observations. Correlation analyses revealed a significant positive correlation between social 

desirability and anticipated guilt (r = .240), as well as a significant negative correlation between 

social desirability and doping consideration (r = -.279). 

Accordingly, the hypothesized structural model was tested again, this time with social 

desirability included. The sum of the 17 items in this scale was used to create a single variable 

representing social desirability. A model was estimated in which social desirability was included 

as an exogenous variable alongside the doping MD and norms variables. In this model, social 

desirability was free to covary with the other exogenous variables and to have direct effects on 

the endogenous variables of anticipated guilt and doping consideration.  

The model with social desirability showed adequate to excellent fit, χ 2 (139) = 204.053, p 

< .01, RMSEA = .044 (.031, .057), CFI = .950, TLI = .939, SRMR = .048. While it is difficult to 

directly compare this model to the focal model of this study, a model with social desirability 

could be estimated in which all its covariances and direct effects were fixed at zero. In this case, 

the simpler model exhibited adequate fit, χ2 (144) = 230.311, p < .01, RMSEA = .050 (.038, 

.062), CFI = .934, TLI = .922, SRMR = .063. A Chi-square difference test was conducted to 

determine whether this model fit significantly worse than the model with social desirability 

saturated. According to this test, there was evidence to suggest the simpler model fit significantly 

worse than the model that included social desirability as a saturated variable, Δχ2 (df = 5, N = 

238) = 29.642, p < .05. Based on these results, the variable of social desirability should not be 

ignored in the model.  

48 
 

In the model that included this variable, social desirability positively associated with 

anticipated guilt (.180) and inversely associated with doping consideration (-.200). Additionally, 

there was a significant negative covariance with subjective norms (-.130). Most interesting was 

the effect of anticipated guilt on doping consideration, which was not significant in this model. 

This might have suggested that the relationship between anticipated guilt and doping 

consideration was better explained by socially desirable responding. Similarly, the mediation 

analysis found an insignificant indirect effect from doping MD to doping consideration through 

anticipated guilt. However, the direct effects of doping MD on anticipated guilt and doping 

consideration remained significant.  

Outliers removed. 

 

After removing the eight outlier scores from the dataset, no significant differences were 

found in the correlation analysis. Additionally, the hypothesized model exhibited adequate fit to 

the data, χ2 (df = 126) = 194.846, p < .01, RMSEA = .048 (.034, .061), CFI = .941, TLI = .929, 

SRMR = .049. Notably, there was an insignificant direct effect from anticipated guilt to doping 

consideration, which was different from the model that included the outlier cases. The mediation 

analysis also found an insignificant indirect effect from doping MD to doping consideration 

through anticipated guilt, despite doping MD having a significant direct effect on both variables.  

Discussion 

 

The purpose of this study was to examine the role of the social environment in shaping 

athletes’ doping-related cognitions—specifically, anticipated guilt and doping consideration. The 

core predictor variables of doping MD, descriptive norms, and subjective norms, while often 

studied in the doping literature, have rarely been combined in a single model. After examining 

49 
 

how these relationships functioned within the model of the current study, meaningful knowledge 

is added to the doping literature. 

 

Recent research has argued the existence of a relationship between doping MD and 

doping susceptibility (Boardley et al., 2019), which we call doping consideration in the current 

study. Some of these studies have also examined the role of anticipated guilt as a mediator of this 

relationship, showing mixed findings (Boardley et al., 2017; Kavussanu & Ring, 2017; Ring & 

Kavussanu, 2017). The presence of these relationships was supported in the current study. 

Doping MD showed a significant effect on doping consideration, directly and indirectly through 

anticipated guilt. This finding provided support to previous literature for the role of moral 

disengagement in athletes’ doping decisions. In other words, an athlete may justify the behavior, 

despite it being a transgression, when considering whether or not to use performance enhancing 

drugs. An athlete who more readily uses such mechanisms of doping MD may be more likely to 

consider doping if presented the opportunity. 

 

The finding of a non-significant effect from anticipated guilt to doping consideration 

when including social desirability in the model should not be ignored. Social desirability 

captures how an individual may respond differently or deceptively to meet perceived social 

expectations. In light of the moral connotations of doping, it is understandable that social 

desirability may have been salient to the relationships among the constructs examined in the 

present work. Higher social desirability was associated with a greater reported anticipation of 

guilt and lower reported doping consideration. Considering this, along with the disappearance of 

the direct effect from anticipated guilt to doping consideration, a spurious relationship may have 

existed. The relationship between anticipated guilt and doping consideration may have been 

better explained by socially desirable responding. Moreover, this could be the reason for some 

50 
 

studies finding anticipated guilt to have a significant association with doping susceptibility 

(Boardley et al., 2017) or doping likelihood (Kavussanu & Ring, 2017; Ring, Kavussanu, & 

Gürpinar, 2020) and other studies do not (Ring & Kavussanu, 2017). In the studies where 

anticipated guilt predicted doping outcome variables, researchers should consider the possibility 

that athletes may be responding in socially desirable ways on the doping measures.  

 

Previous doping research has addressed social desirability in many ways. Many authors 

(Barkoukis et al., 2014; Lazuras et al., 2010; Lazuras et al., 2015) examined the construct with 

correlation and moderation analyses, finding small correlations between social desirability and 

doping-related variables but no confounding effect on the relationship between predictors and 

outcomes. Kavussanu et al. (2016) did not find social desirability to correlate with their MD 

measure, nor did they find it to influence the relationship between MD and doping likelihood. 

Gucciardi et al. (2010) found the relationship between doping attitudes and doping susceptibility 

to be partially mediated by social desirability using structural equation modeling. While these 

methods are preferred over the assumption that social desirability is mitigated by use of an 

anonymous survey (Kavussanu & Ring, 2017) or hypothetical scenarios (Ring, Kavussanu, 

Simms, & Mazanov, 2018), a consensus has not been reached as to the optimal way to address 

this variable. The inclusion of social desirability in the supplemental analysis of this study was a 

relatively robust strategy, and provided great insight when interpreting results. Future research 

should examine this variable in a similar way, considering the implications of social desirability 

on self-report of doping. 

 

Despite the decreased significance of an effect from anticipated guilt to doping 

consideration when social desirability is included in the model, doping MD still showed 

significant effects on both anticipated guilt and doping consideration. It appeared that doping 

51 
 

MD predicted doping-related outcomes, even when controlling for social desirability. This is an 

interesting result, considering there have been mixed results regarding social desirability in 

doping research (Backhouse et al., 2007; Gucciardi et al., 2010; Kavussanu et al., 2016). In the 

current study, social desirability was included to provide a fuller picture of the relationship 

among the core variables. Specifically, doping MD turned out to be a robust predictor when 

considered alongside the social contextual variables of subjective norms and descriptive norms.  

 

Subjective norms are a measure of how an individual considers social expectations in 

regard to a particular behavior—doping, in the case of this study. An interesting finding was that 

subjective norms did not have a significant association with anticipated guilt and doping 

consideration. The lack of association of subjective norms with these variables is contradictory to 

previous research (Lazuras et al., 2015; Ntoumanis et al., 2014), but might provide insight into 

the role of social dynamics in doping decisions. Subjective norms did exhibit a moderately 

strong relationship with doping MD in the hypothesized model. Such a relationship might reflect 

a close conceptual link between these two variables. Considering this, in addition to the 

correlations found between subjective norms and the outcome variables of interest, some insight 

might be drawn. Items from the DMDS (Boardley et al, 2018) targeting the doping MD 

dimension of displacement of responsibility involve the role of significant others (e.g., 

teammates, training groups, and coaches) in one’s endorsement of doping. It is possible that 

doping MD better explains anticipated guilt and doping consideration because it already includes 

the cognitive appraisal of social support for doping. Therefore, further examination of subjective 

norms simultaneously with doping MD may require development of a more distinctive subjective 

norms measure.  

52 
 

 

Notably, the three subjective norms items used in this study ask about athlete perceptions 

of approval or support for doping. What is not included in these items is the level of disapproval 

for the behavior, or level of support for behaviors that avoid doping. These kinds of items may 

be more relevant in reference to doping, because use of performance enhancing drugs is a 

transgressive behavior. Moreover, this might allow for subjective norms measures to be more 

distinct from the doping MD measure, which captures the process of justifying transgressive 

behavior. This could affect the pattern of associations in the present structural model such that 

subjective norms would directly predict the doping outcomes of interest in addition to the direct 

effect from doping MD. 

 

Similar to subjective norms, descriptive norms did not show any significant predictive 

effects in the model. This reflects the general lack of correlation between descriptive norms and 

the other variables, and is a finding that contrasts with previous research (Ntoumanis et al., 

2014). To make sense of this contradictory finding, consideration was given to whether the three 

items used for measuring descriptive norms would actually be salient to athletes. Two items 

asked about their perception of the behavior of elite athletes, but it was difficult to know whether 

respondents perceived themselves as elite athletes. Research on social norms theory (Perkins & 

Berkowitz, 1986) suggests an individual’s perceptions of others’ behavior differs from 

perceptions of their own. Such self-other discrepancies have been found in the doping literature 

(Ring, Kavussanu, Mazanov, 2019). Furthermore, Borsari and Carey (2003) suggest a reference 

group that is further in proximity away from the participant will lead to more inaccurate or 

inflated perceptions of the reference group’s behavior. An athlete’s perception of elite athlete 

doping may be dependent on their perceived proximity to elite athletics. In addition to the elite 

athlete items, all three descriptive norms items were restricted to perceived doping behavior. Past 

53 
 

research has used hypothetical scenarios similar to the anticipated guilt and doping consideration 

measures in the current study (Petróczi, Mazanov, Nepusz, Backhouse, & Naughton, 2008), and 

asked participants what percentage of athletes would dope in these particular situations. Such 

scenario-based measures of descriptive norms may better capture the nuance of potential doping 

scenarios in future research. 

Implications 

Overall, the implications of this study are that the construct of doping MD may be a 

better indicator of downstream doping-related cognitions than normative influences. This makes 

sense when viewed through the lens of the social-cognitive theory of moral thought and action 

(Bandura, 1991), which argues that social sanctions are internalized over time. It is possible that 

social sanctions for doping had already been internalized by these athletes, therefore leading to 

greater salience of the cognitive appraisals to doping behavior than the perceived appraisals of 

significant others. Because athletes in this study were adults, it would make sense for this 

internalization to have already occurred. Future research may be able to explore this concept 

further by recruiting youth athletes, a population with great potential for intervention (Barkoukis 

Kartali, Lazuras, & Tsorbatzoudis, 2016, Hurst, Ring, & Kavussanu, 2020). Regardless of age, 

researchers should consider the nature of the relationship between doping MD and normative 

influence when examining how these predict doping outcomes.  

Practical implications would be seen in the development of anti-doping campaigns, which 

are continuing to see modifications for use with different age groups (Hurst et al., 2020), with 

various members of the athletic support personnel (Ntoumanis et al., 2020), and through various 

forms of media (Nicholls et al., in press). The content of these interventions is in a constant state 

of development. Whereas early interventions targeted athletes’ attitudes toward doping 

54 
 

(Backhouse, McKenna, Robinson, & Atkin, 2007), more recent interventions have targeted 

variables such as beliefs about supplement use (Hurst et al., 2020), coach communication style, 

and coach motivational climate (Ntoumanis et al., 2018; Ntoumanis et al., 2020). The present 

results suggest that doping MD may be an important consideration when developing anti-doping 

campaigns. Hurst et al. (2020) incorporated a measure of doping MD in their intervention study, 

but did not seem to describe how this construct was addressed in the clean sport educational 

program. Despite this, the intervention was successful in decreasing doping MD in the short-

term, but the authors suggested stronger interventions or regularly-scheduled booster sessions to 

maintain the effects in the long-term. It is possible that deliberately impacting doping MD 

constructs (e.g., targeting distortion of consequences by reminding participants that doping 

presents risks to physical and mental health, and describing how downplaying these adverse 

effects would lead an athlete to falsely convince themselves that doping is okay) may be more 

effective in minimizing doping MD use for the short- and long-term. 

Limitations 

 

Beyond the measurement selection issues addressed above, a particularly important 

limitation was the timing of data collection. The entire distribution and collection period was 

April to October 2020, in the midst of the COVID-19 pandemic. During this time, competitions 

and practices were modified or canceled completely as a means of minimizing the spread of the 

virus. Such restrictions could have interfered with typical athlete training habits, especially with 

sports that require simultaneous, team-based training. Moreover, the restrictions on social 

gatherings and traveling could have impacted the non-training social environment. Such 

interactions might otherwise make social norms more salient, or reinforce the strength of doping-

55 
 

based cognitions. While it cannot be fully understood how this historical period could have 

influenced athlete survey responses, the possibility should not be ignored. 

 

A possible method of examining the influence of the worldwide pandemic on doping 

cognitions would be to consider differences between endurance and non-endurance athletes. 

Endurance sports such as swimming, running, or cycling provide the opportunity to continue 

training independently, whereas non-endurance sports (particularly team sports such as lacrosse, 

volleyball, and football) require a certain level of codependent training with the team. Thus, 

social restrictions may have had a greater influence on the social component of codependent 

sports than independent sports. Even in the unrestricted setting, some researchers argue that 

endurance and non-endurance athletes should be studied separately, not only because of the 

distinct nature of training, but also because endurance sports have been argued to be at higher 

risk for doping practices (Aubel, Lefevre, Le Goff, Taverna, 2019). For this reason, a post-hoc 

analysis of endurance and non-endurance athlete data was conducted, running independent t-tests 

to find any significant mean differences on the variables of interest. In this analysis, the only 

significant difference in means between endurance and non-endurance athletes was in doping 

consideration (F = 4.623, p = .03). Surprisingly, non-endurance athletes reported higher doping 

consideration (M = 2.39, SD = 1.55) than endurance athletes (M = 2.14, SD = 1.72), which was in 

contrast to expectations based on previous authors’ speculation (Aubel et al., 2019). Whether this 

difference was an effect of the groups in the sample or the environmental conditions can only be 

discussed as speculation until further research is completed.  

 

An interesting finding tied to endurance/non-endurance correlations was in regard to 

subjective norms, which had significant correlations with doping MD (r = .346, p < .05), 

anticipated guilt (r = -.485, p < .05) and doping consideration (r = .360, p < .05) in the non-

56 
 

endurance athletes but not in the endurance athletes. These results would suggest that athletes in 

co-dependent sports may take social support/pressures into consideration when making doping 

decisions to a greater extent than their endurance/individual sport counterparts. Unfortunately, 

this relationship was not retained in the tested structural model for non-endurance athletes, but 

this may have had to do with the small sample size (n = 94) for running the statistical model with 

a subgroup. Regardless, it may be important to consider social factors or group dynamics in 

future research, even when examining individual sports (Evans, Eys, & Bruner, 2012).  

Conclusion 

 

This study tested a model in which doping MD and normative influences predicted 

anticipated guilt from, and consideration of, doping. Doping MD was the dominant construct in 

the model, predicting doping consideration directly and indirectly via anticipated guilt. Thus, 

doping MD may explain downstream doping cognitions better than social contextual norms, and 

should be strongly considered by researchers and anti-doping advocates. 

 

 

57 
 

CHAPTER FOUR: GENERAL DISCUSSION 

 

Since the World Anti-Doping Agency formed in 1999, its multifaceted mission to “lead a 

collaborative worldwide movement for doping-free sport” (WADA, 2019) has been met with 

numerous obstacles. In a time when the cost of anti-doping amounts to about $69,300 for every 

anti-doping rule violation (Maennig, 2014), the efficiency and legitimacy of anti-doping 

campaigns have been brought into question. Specifically, a recent mapping review argues there 

is a perception of anti-doping to be “doing the right thing,” but seemingly not “doing it the right 

way” (Woolway, Lazuras, Barkoukis, & Petróczi, 2020). In light of this perception, efforts to 

improve the anti-doping regime have combined the traditional detection-deterrence technique 

with programs targeting psychosocial predictors of doping decisions.  

 

Possibly the most promising psychosocial predictor of doping decisions is moral 

disengagement (MD). Though this conditional endorsement of transgressive behavior has been 

considered in previous research on doping, the recent development of the Doping Moral 

Disengagement Scale (DMDS; Boardley et al., 2018) has presented great value for understanding 

the nuance of MD in the context of doping in team and individual sport. Research on this 

variable has supported a predictive effect of doping MD on various doping outcomes (Boardley 

et al., 2019; Ring & Hurst, 2019). A potential mechanism of this effect is through the level of 

guilt the athlete anticipates feeling if they were to use performance enhancing drugs or methods, 

although not every study has found support for this relationship (Ring & Kavussanu, 2017). 

 

In addition to doping MD, two socially salient variables may similarly predict doping 

decisions. Specifically, the perceived prevalence of doping (i.e., descriptive norms) or the extent 

to which significant others support the behavior (i.e., subjective norms) may be taken into 

account. These normative influences provide a context within which cognitions such as doping 

58 
 

MD take place. Therefore, possibly the best method to examine the role of doping MD in athlete 

decision making is to consider the simultaneous role of subjective and descriptive norms. 

Simultaneous examination of these variables provides a more comprehensive picture of the 

factors playing a role in athletes’ doping cognitions. 

 

The purpose of this dissertation was to simultaneously examine three variables which 

have previously been found to independently predict such outcome variables as doping 

susceptibility/consideration (Boardley et al., 2019), doping intentions (Lazuras, Barkoukis, 

Mallia, Lucidi, & Brand, 2017), or doping likelihood (Kavussanu & Ring, 2017). While these 

variables have been studied separately, this dissertation combined doping MD, subjective norms, 

and descriptive norms in a number of different ways. The contexts of the two studies differed in 

population recruited, outcome measures used, and statistical method. 

 

A novel contribution of Study 1 was the exclusive recruitment of disability sport athletes. 

Concerns for doping are not unique to the able-bodied sports context. Therefore, an examination 

of the three variables of interest in disability sport athletes was conducted. This study was 

expected to springboard further research with this underrepresented population. Due to the 

exploratory nature of the study in recruiting from this population, a descriptive statistical 

approach was used. Specifically, cluster analysis was used to reveal four unique groups of 

athletes based on the three variables of doping MD, descriptive norms, and subjective norms. 

Athletes in clusters 1 (n = 34) and 2 (n = 44) exhibited a mixture of high and low levels on the 

three cluster variables. Specifically, cluster 1 was characterized by relatively high scores on 

doping MD, with descriptive norms and subjective norms values around the mean. Cluster 2 was 

characterized by relatively high descriptive norms, with doping MD and subjective norms values 

59 
 

around the mean. Athletes in clusters 3 (n = 15) and 4 (n = 92) were characterized by relatively 

high and low levels on all the cluster variables, respectively.  

 

Athletes in clusters 1 and 3 reported significantly lower levels of anticipated guilt in a 

hypothetical doping situation. Furthermore, athletes in cluster 3 reported significantly higher 

doping intentions than the other three clusters. The good news for anti-doping stakeholders was 

that cluster 4, characterized by low levels of all three cluster variables, was the largest group and 

cluster 3 (the high-risk group), characterized by high levels of all three cluster variables, 

represented a small number of disability sport athletes. An added benefit of finding this group of 

athletes was the potential to develop targeted, tailored anti-doping messages or educational 

programs, as these athletes may be the most at risk for doping.  

 

Although the exclusive recruitment of disability sport athletes in Study 1 offered a novel 

contribution to the literature, it also presented a tradeoff in the form of study limitations. For 

example, the use of definitions (i.e., doping) and items (i.e., descriptive norms) that have 

typically been used in research with able-bodied athletes made it difficult to determine whether 

they covered the nuances of disability sport, and whether they were interpreted in the way 

expected. Actions were taken to minimize the chance of these variables being misinterpreted, but 

the success of these measures can only be strengthened through further research with disability 

sport athletes as well as their able-bodied counterparts.  

 

To address some limitations of Study 1, and to expand on the findings therein, Study 2 

was conducted with three distinctions. First, able-bodied athletes were recruited rather than 

athletes with physical disabilities in order to simultaneously examine the core variables from 

Study 1 in a different population. Second, a measure of doping consideration (previously referred 

to as doping susceptibility) was used in place of doping intentions. Third, structural equation 

60 
 

modeling was performed to further align with previous research in the able-bodied population. 

Whereas Study 1 used exploratory, descriptive analyses in an understudied population, Study 2 

used a sample more heavily represented in the literature to test a model in which the three focal 

variables of doping MD, descriptive norms, and subjective norms predicted doping 

consideration, directly and indirectly via anticipated guilt. This method offered a more robust 

simultaneous examination of the variables, providing a fuller representation of the social context 

surrounding doping cognitions. 

The tested model exhibited adequate to excellent fit with the data, explaining 35.3% and 

22.1% of the variance in anticipated guilt and doping consideration, respectively. Doping MD 

emerged as a significant direct and indirect predictor of doping consideration, and the direct 

effect was maintained even when a measure for social desirability was included in the model or 

outliers were removed. The predictive effect of anticipated guilt on doping consideration 

disappeared when outliers were removed as well as when social desirability was included, which 

may have provided a greater picture regarding the relationship between these two variables. 

The outlier analysis found three outlier cases on the variable of anticipated guilt, all of 

which were at the low end of the scale. Though these values were removed for the outlier 

analysis, it has been argued that these values could represent a true, small subset of the sample 

(see Study1). In the context of doping, it is particularly possible that while most athletes 

anticipate some guilt if they were to use performance enhancing drugs, a small group of athletes 

anticipate little or no guilt. Thus, it may make more sense to retain the outliers who are low on 

anticipated guilt, and similarly to retain outliers who are high in doping MD, perceived norms, or 

doping consideration. The relationship between anticipated guilt and doping outcomes may be 

more prominent when the small group of athletes with low anticipated guilt happen to be 

61 
 

included in the sample. It is possible the presence of outliers provides a more accurate picture of 

the relationship. 

Furthermore, the contradictory findings regarding anticipated guilt may not be a complete 

refutation of its role in doping decisions. For instance, activation or disengagement of moral 

reasoning is argued to occur after an individual anticipates self-sanctions like guilt or shame for 

performing a behavior (Bandura, 1991). If this is the case, anticipated guilt may act as more of a 

genesis or catalyst for the doping decision-making process, and less so as an integral part of the 

decision. Future investigators may be interested in placing anticipated guilt earlier in the model, 

if they choose to include it at all. After all, doping MD maintained its role under various 

circumstances, and may therefore be more important to include than anticipated guilt in future 

models of doping decisions. 

A surprising finding was the lack of direct or indirect effects on doping consideration 

coming from subjective and descriptive norms. Furthermore, whereas subjective norms showed a 

strong covariance with doping MD in the model, descriptive norms did not. From a theoretical 

standpoint, the interplay between doping MD and the norms measures may be explained by the 

social-cognitive theory of moral thought and action (Bandura, 1991). According to this theory, 

individuals internalize the social sanctions they observe over time. Both studies included adult 

athletes, which may indicate this internalization had already occurred. For this reason, the 

perceived social sanctions may not have been as salient to the athletes as their own internal 

sanctions or justifications for doping. Continued examination with different populations (i.e., 

youth athletes) may help to establish whether doping MD is salient across different stages of 

internalization. Regardless, a number of considerations should be taken to ensure future 

examination is theoretically sound. Firstly, it is possible the DMDS better predicts the outcome 

62 
 

variables because it includes components similar to the measure of subjective norms. Hence, it 

may be necessary to modify the subjective norms items in order to make a proper distinction 

between this variable and doping MD. Such modifications may involve items asking to what 

extent significant others disapprove of doping or approve of doping avoidance behaviors. 

Secondly, the descriptive norms items may need to be revisited for their utility in doping 

research. Petroczi et al. (2008) used hypothetical doping scenarios to measure athletes’ estimates 

of doping use in others, which may fall more in line with the measures for anticipated guilt and 

doping consideration in this dissertation. This method may be a starting point for using 

alternative measures of descriptive norms. 

Viewing the results of the two studies collectively, there are a number of similarities 

between responses from the two study populations. Athletes in disability sport and able-bodied 

sport scored similarly in doping MD (M = 2.21 and M = 2.27, respectively), descriptive norms 

(M = 24.15 and M = 22.28, respectively), subjective norms (M = 1.39 and M = 1.33, 

respectively), and anticipated guilt (M = 3.82 and M = 4.24, respectively). From this it can be 

concluded that athletes from both of these populations have similar cognitions regarding the 

prevalence and approval of doping, as well as similar use of MD mechanisms. It is possible, 

then, that the social environment surrounding doping is comparable across disability and able-

bodied sport, at least as perceived and appraised by the athletes.  

Beyond this, strong insight can be garnered and applied in future research and anti-

doping campaigns. Most prominent across the two studies was the role of doping MD. Expected 

relationships were present in application to distinct populations (i.e., disability sport athletes in 

Study 1 and able-bodied athletes in Study 2), under varying degrees of robustness in statistical 

analysis (i.e., cluster analysis in Study 1 and latent variable analysis in Study 2), across different 

63 
 

distal dependent variables (i.e., doping intentions in Study 1 and doping consideration in Study 

2), and when taking social desirability into consideration (Study 2 only). Therefore, mechanisms 

of doping MD are apparently dominant and robust in predicting doping-related decision-making 

outcomes. 

Insight can also be gained on the role of anticipated guilt from this research. In Study 1, 

groups characterized by lower doping MD reported higher anticipated guilt. This was in line with 

previous research (Boardley et al., 2017; Ring et al., 2020). Additionally, anticipated guilt was a 

proximal predictor of doping consideration in the full sample in Study 2. However, this 

predictive effect became non-significant after including social desirability in the model or 

ignoring outliers. Study 2 was not the first to find this contradictory result (Ring & Kavussanu, 

2017). It is possible that social desirability would better explain the relationship between 

anticipated guilt and doping consideration than anticipated guilt alone. The relationship between 

anticipated guilt and doping outcomes may not be as promising as previous research has 

suggested. With social desirability predicting scores of anticipated guilt and doping consideration 

better than subjective norms, it might be argued that social expectations could influence how 

athletes respond to questions about doping, but not necessarily influence the decision-making 

outcomes of such cognitions. Future research may want to take heed to the effect of social 

desirability when examining anticipated guilt and doping consideration. Protecting participants’ 

anonymity can reduce likelihood of socially desirable responding, but greater measures may 

need to be explored beyond what was done in this dissertation. Importantly, doping MD 

exhibited a significant effect on doping consideration regardless of whether or not social 

desirability was included. 

64 
 

Anti-doping organizations could draw practical implications from this dissertation, 

specifically in regard to doping MD as a target variable. Early anti-doping interventions targeted 

attitudes toward doping (Backhouse et al., 2007) with the assumption that attitudes were the 

primary predictor of doping behavior. Other psychosocial variables have been examined as 

outcomes of these education programs, and these include doping MD. Hurst et al. (2020), for 

example, found athletes in the clean sport program to have lower doping MD immediately 

following the program. Unfortunately, these effects did not continue long-term. This issue may 

stem from the five parts of the 60-minute program (i.e., anti-doping governance, anti-doping rule 

violations, drug testing procedures, banned medications, and contaminated supplements) not 

deliberately addressing possible mechanisms of doping MD. A similar issue can be found in an 

intervention by Barkoukis and colleagues (2016), in which normative beliefs were measured but 

were not targeted other than to describe historical and modern instances of doping.  

While measuring variables like doping MD as outcomes of interventions is a step in the 

right direction, the development of interventions to address these variables directly is necessary 

to continue strengthening these campaigns. For example, an intervention could address the 

doping MD dimension of displacement of responsibility by acknowledging how athletes tend to 

justify use of performance enhancing drugs via an argument that they do not have a choice if the 

coach tells them to do so. Following this, the program could include strategies or resources for 

minimizing the chances an athlete would use this mechanism (e.g., independent moral reasoning 

or whistleblowing hotlines). Incorporating sessions like these would potentially be more 

effective in reducing doping MD use by athletes, thus creating more effective anti-doping 

programs. However, more research on such strategies is necessary before proponents of anti-

doping can know the best intervention practices. 

65 
 

 

Although the two studies in this dissertation complement one another and offer important 

contributions to the extant knowledge base, there are important limitations of the research to 

consider. First, non-experimental survey research with a single timepoint for each participant is 

not the gold standard for finding causal relationships among variables. Latent variable analysis, 

such as the structural equation modeling from Study 2, can at best imply causal relationships 

based on previous theory and researcher expertise. Experimental interventions can be difficult to 

implement, but are necessary to find robust support for a causal relationship between predictors 

and outcomes of interest. While intervention studies come with their own difficulties and 

limitations, some have been undertaken with a degree of success—including one with doping 

MD as an outcome variable of interest (Hurst et al., 2020). 

Second, measurement of MD has evolved since its first application to doping research, 

but measurement of descriptive and subjective norms has remained relatively stagnant. This may 

help to explain why doping MD has shown considerable promise in predicting various doping-

related outcomes. Many studies have examined normative variables alongside MD variables, but 

only a dearth of studies (if any) have examined both subjective and descriptive norms 

simultaneously with the DMDS. This may explain why doping MD came out as a predictor in 

Study 2, while the norms measures did not. At this point in the research, it may be a good idea to 

revisit the commonly-used measures for descriptive and subjective norms and determine whether 

they can maintain their predictive strength when held to the same standard of measure 

development as the DMDS.  

Third, this dissertation fell in line with numerous studies by examining doping intentions 

(Lazuras., 2017; Ntoumanis et al., 2014) and doping susceptibility/consideration (Boardley et al., 

2019; Gucciardi et al., 2010). Although these are common outcome variables in doping research, 

66 
 

the practical outcome of interest is actual doping behavior. Unfortunately, objective measures 

typically do not offer accurate assessment of doping, and self-reported doping is suggested to be 

highly susceptible to false reporting. Despite these reservations, assessment of doping behavior 

could offer more ecologically valid measurement of doping outcomes. 

Finally, the data collection period of Study 2 occurred in the midst of the COVID-19 

pandemic. This may have been a limitation, but may also provide greater insight considering the 

adjusted training environment for athletes during this time. Whether or not such a historical 

period has a positive or negative effect on doping research can only be speculated as of the 

publication of this dissertation. Regardless, this is important to keep in mind if future research 

determines that restrictions on social contact affect athletes’ responses to doping-related 

questionnaires, perspectives on doping, or actual doping behavior. 

Also relevant to the COVID-19 pandemic is the issue of collecting data in person or 

online. Neither of these options has been put forward as the preferred method of data collection. 

Most doping research studies use either online (e.g., Petroczi et al., 2008; Ring et al., 2020) or in-

person (e.g., Boardley et al., 2019; Lazuras et al., 2015) surveys, but not both. The few studies 

implementing both online and in-person surveys provide mixed results in comparing the two 

methods. Bhambhani et al. (2010) used both online and in-person methods to increase sample 

size in a study of exclusively athletes with spinal cord injury, but did not make any comparisons 

across the two methods. Pitsch and Emrich (2011) argued that their online survey failed to 

produce reliable results, and chose to retain the in-person data. In contrast, Connor, Woolf, and 

Mazanov (2012) found no significant difference between their interview-based and online 

collection methods. In-person surveys may be beneficial for monitoring the environment in 

which the survey is completed, whereas online surveys may be particularly beneficial in 

67 
 

reinforcing perceived anonymity of responses and in research with disability sport athletes who 

may require accommodating technologies. Future research should determine whether one 

method is more reliable and valid for use in the context of doping, or if both can be considered 

viable options. 

 

Future research should also aim to further explore athletes’ doping decisions based on the 

results of this dissertation. The promising line of research on doping MD has led to a number of 

studies exploring possible antecedents, such as moral identity (Kavussanu & Ring, 2017), basic 

moral values (Ring et al., 2020), and doping confrontation efficacy (Sullivan, Feltz, LaForge-

MacKenzie, & Hwang, 2015). Of these antecedents, doping confrontation efficacy is unique 

because it involves members of the athlete support personnel as responsible members of the fight 

against doping. Early research on doping considered the behavior to be an individualized 

decision, ignoring the role of the social environment (Erickson, Backhouse, & Carless, 2017; 

Whitaker, Backhouse, & Long, 2017). Such narrow focus may explain why previous 

interventions targeted attitudes while ignoring other, more salient constructs. An athlete’s 

perception of their coach’s, parent’s, or teammate’s ability to confront them about their doping 

behavior may be just as important as their own attitudes, normative beliefs, or moral 

disengagement regarding doping. Previous research on doping confrontation efficacy has 

focused on coaches, and has found that athlete perceptions of their coaches’ doping confrontation 

efficacy has downstream effects on their doping attitudes (Sullivan & Razavi, 2017), doping 

MD, and doping susceptibility (Boardley et al., 2019). Educational programs and anti-doping 

campaigns can benefit from not only targeting individual psychological constructs such as 

attitudes, perceived norms, or doping MD, but by also considering the social, environmental 

agents in athletes’ doping decisions. Interventions do exist that target coach knowledge and 

68 
 

attitudes of doping (Nicholls et al, 2020) as well as coach motivational climate (Ntoumanis et al., 

2018). Considering these interventions, variables related to the process of doping MD have been 

largely ignored. One study includes doping MD as an outcome variable (Hurst et al., 2020), but 

targeting antecedents of doping MD (either in athletes or in members of their entourage) could 

improve the multifaceted anti-doping approach taken by such organizations as WADA and the 

IOC. 

 

Generalization of the findings put forth may also be explored more in future research. 

Disability sport athletes in Study 1 reported levels of the focal variables similar to those of able-

bodied athletes in Study 2 and previous research. However, the unique circumstances and doping 

methods may require a more in-depth look. Boosting (i.e., artificial induction of autonomic 

dysreflexia, which has been suggested to enhance performance) is an example of a behavior that 

can be considered doping in athletes with spinal cord injury (Legg & Mason, 1998; Mazzeo, 

Santamaria, & Iavarone, 2015). Knowledge regarding this practice is primarily anecdotal, so 

future researchers may benefit from exploring the social environment regarding this behavior. 

The exploration of doping in the disability sport context in this dissertation is hopefully the 

beginning of a long line of research that will help anti-doping efforts in disability and able-

bodied sport alike.  

 

Altogether, the current dissertation addressed how athletes in disability and able-bodied 

sport integrate social norms with moral cognitions when making doping-related decisions. In the 

short-term, this dissertation hopefully offers enough insight to guide further examination of these 

variables in the doping context. Long-term, this research hopefully lays the foundation for 

theory-driven interventions to fight doping in sport. Currently, elite athletes hold mixed 

perceptions about the effectiveness of anti-doping efforts (Westmattelmann, Dreiskämper, 

69 
 

Strauß, Schewe, & Plass, 2018), bearing the question of whether the noble goal of doping-free 

sport is being sought out in the right way (Woolway et al., 2020). A stronger theoretical base 

could improve anti-doping campaigns as well as their perceived effectiveness, which might 

create an adaptive, self-fulfilling anti-doping cycle. 

 

Although doping is a popular topic of discussion among the media, scientific researchers, 

and the general public, it is not the only morality-based behavior of interest in the context of 

sport. There are numerous prosocial and antisocial behaviors to be considered in sport, with a 

multitude of antecedents and consequences for each given behavior (Kavussanu & Stanger, 

2017). However, doping may be the preeminent behavioral threat to the sanctity of sport. 

Developing a successful, theory-driven intervention to reduce the prevalence of doping in sport 

would be a large step forward in the fight to maintain a “spirit of friendship, solidarity and fair 

play” (IOC, 2015, pg. 13). A small difference made on the doping front could be the harbinger of 

change in other prosocial and antisocial behaviors in sport. The current dissertation advances 

knowledge in a way that hopefully will offer some contribution to this change. 

 

 

 

70 
 

 

 

 

 

 

 

 

 

 

 

 

 

APPENDICES 

 

71 
 

APPENDIX A 

Study One – Human Research Protection Program Approval Letter 

 

 

72 
 

73 
 

 

 

 

APPENDIX B 

Study One – Questionnaire Packet 

 

 

74 
 

Please provide some information about yourself. 
 

1. Age (years): _____________        2. Sex:   Male □    Female □    Prefer not to say 
□ 

3. What is your main sport: ___________________________    

4. Years competing in this sport: ______________________ 

5. Highest level you compete at in your main sport:  
Regional □     National □     International □ 
6. Average training/competition hours per week for your main sport: 
__________________ 
7. Weeks out of the year spent training/competing in your main sport: 
_________________ 
 

The following questions are for athletes competing in adaptive (para)-sports.  

If these questions do not apply to you, please write “N/A” 

 
8. Type of disability: ________________________    Acquired □    Congenital □     

9. Classification:  ________________________ 

 
 
 
 
 

 

75 
 

Below you will find a list of statements. Please read each statement carefully and 
decide if that statement describes you or not. If it describes you, circle the word 
“true”; if not, check the word “false.” 
  

Please indicate whether each of the following statements is true or 
false for you. 
1.  I sometimes litter 
2.  I always admit my mistakes openly and face the potential negative 

consequences. 

 

 

True  False 

True  False 

True  False 
True  False 

True  False 

3.  In traffic I am always polite and considerate of others. 
4.  I have tried illegal drugs (for example, marijuana, cocaine, etc.). 
5.  I always accept others’ opinions, even when they don’t agree with my 

own. 

6.  I take out my bad moods on others now and then. 
True  False 
7.  There has been an occasion when I took advantage of someone else.  True  False 
8.  In conversations I always listen attentively and let others finish their 

True  False 

sentences. 

9.  I never hesitate to help someone in case of emergency. 
10. When I have made a promise, I keep it – no ifs, ands or buts. 
11. I occasionally speak badly of others behind their back. 
12. I would never live off other people. 
13. I always stay friendly and courteous with other people, even when I 

am stressed out. 

14. During arguments I always stay objective and matter-of-fact. 
15. There has been at least one occasion when I failed to return an item 

that I borrowed. 

16. I always eat a healthy diet. 
17. Sometimes I only help because I expect something in return. 

True  False 
True  False 
True  False 
True  False 

True  False 

True  False 

True  False 

True  False 
True  False 

Doping 

 
 

Doping is defined as use of prohibited substances or methods by an athlete with the 
potential to artificially improve performance through changes in physical and/or mental 
condition. Please keep this definition in mind when answering questions about doping. 

 

 

76 
 

A number of statements describing thoughts that athletes might have about 
doping are listed below. Please read these statements carefully and indicate your 
level of agreement with each one by circling the appropriate number. Please 
respond honestly. 
 

What  is  your  level  of  agreement  with  the 
following statements? 

1.  It is okay to dope if it helps an athlete to 

provide for his/her family. 

2.  Saying you "take steroids" feels worse than 

saying you "use some gear". 

3.  Compared to most lifestyles in the general 

public, doping isn't that bad. 

4.  Athletes shouldn't be blamed for doping if 

training partners/teammates pressure them 
to do it. 

5.  If most athletes in a sport dope, no one 

athlete should be held responsible for doing 
it. 

6.  Risks associated with doping are 

exaggerated. 

7.  Doping is okay if it helps an athlete advise 

others on how to do it right. 

8.  Using words like "roids", "gear" and 

"pinning" makes doping feel more 
acceptable. 

9.  Compared to smoking, doping is pretty safe. 
10. An athlete shouldn’t be blamed for doping if 

a member of his/her training group has 
encouraged it. 

11. It’s not right to condemn individuals who 

dope when many in their sport are doing the 
same. 

12. Doping doesn’t really harm anyone else. 
13. It is acceptable to dope if knowledge gained 

helps an athlete advise others on safe 
doping. 

77 
 

 

 

l

y
g
n
o
r
t
S

e
e
r
g
a
s
D

i

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

 

e
e
r
g
a
s
D

i

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

 

 

y

e
e
r
g
a
s
D

i

l
t
h
g

i
l

S

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

 
l

a
r
t
u
e
N

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

 

y

 

e
e
r
g
A

l
t
h
g

i
l

S

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

 

e
e
r
g
A

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

 

 

l

y
g
n
o
r
t
S

e
e
r
g
A

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

1 

1 

1 

1 

1 

2 

2 

2 

3 

3 

3 

4 

4 

4 

5 

5 

5 

6 

6 

6 

2 

3 

4 

5 

6 

2 

3 

4 

5 

6 

7 

7 

7 

7 

7 

14. Using terms such as "gear" or "juice" makes 

doping sound less harmful. 

15. Compared to physical violence, doping isn't 

that serious. 

16. An athlete shouldn't be held responsible for 
doping if his/her coach encouraged him/her 
to do it. 

17. If an athlete trains/competes in an 

environment in which doping is the norm, 
he/she shouldn't be held accountable for 
doing it. 

18. The negative aspects of doping are 

exaggerated by the media. 

 
 

 

78 
 

Now, we would like to assess your perceptions about the prevalence of doping in 
sport. 

 

1.  Out of 100%, how many athletes at your competitive level do 
you believe engage in doping to enhance their performance?   

____________% 

2.  Out of 100%, how many elite athletes in the USA do you 
think engage in doping to enhance their performance?   

3.  Out of 100%, how many elite athletes do you believe will be 
engaged in doping during the next 3 years to enhance their 
performance? 

____________% 

____________% 

 
 
 
Now, we would like to assess your perceptions of others’ thoughts on doping in 
sport. 

What  is  your  level  of  agreement  with  the 
following statements? 

 
y
l
g
n
o
r
t
S

 
e
e
r
g
a
s
i
D

 
e
e
r
g
a
s
i
D

 
y
l
t
h
g

i
l

S

 
e
e
r
g
a
s
i
D

1.  Most people who are important to me would 

want me to use doping substances or 
methods to enhance my performance. 

2.  Most people I know would agree if I used 

doping substances or methods to enhance 
my performance. 

3.  Most people who are close to me would like 
me to use doping substances or methods to 
enhance my performance. 

1 

1 

1 

2 

2 

2 

3 

3 

3 

 
l
a
r
t
u
e
N

4 

4 

4 

 
y
l
t
h
g

i
l

S

 
e
e
r
g
A

 
e
e
r
g
A

 
y
l
g
n
o
r
t
S

 
e
e
r
g
A

5 

5 

5 

6 

6 

6 

7 

7 

7 

 

 
 
 
 
 

 

 
 

 
 
 

79 
 

Now, please imagine being in the following situation:  
 

“Having returned to training following a period of injury, you are feeling very out of 

shape. As such, you feel the need to get back in shape as soon as possible. A friend 
who you train with has been taking a training supplement that he/she says really helped 

him/her get back in shape quickly following a similar injury. He/she offers to give you 

some and you decide to take it. Subsequently you get back in shape much quicker than 

expected, but then discover the supplement you have been taking is a banned 
performance-enhancing substance. However, due to the improvements you have 

experienced, you decide to continue taking the substance.” 

Now, using the following scale, please rate the extent to which you anticipate you 
would feel in the ways described below if you decided to continue taking the 
substance. Please answer honestly. 
 

 

 

 
l
l

a
 
t
a
 
t
o
N

1 

1 

1 

1 

1 

 

y

l

e
t
a
r
e
d
o
M

3 

3 

3 

3 

3 

 

e
l
t
t
i
l
 

A

2 

2 

2 

2 

2 

 

h
c
u
M
y
r
e
V

 

 

l

y
e
m
e
r
t
x
E

4 

4 

4 

4 

4 

5 

5 

5 

5 

5 

I would anticipate feeling the following 
about continuing to take the substance… 

1.  I would feel remorse, regret. 

2.  I would feel tension about what I was doing. 
 
3.  I would not be able to stop thinking about 

the bad thing I was doing. 

4.  I would feel like apologizing, confessing. 

5.  I would feel bad about what I was doing. 

 
 
 
 
 

 

 
 
 
 
 

80 
 

Next, we would like to assess your intentions towards doping in sport. For each of 
the questions listed below, please circle the number that best corresponds to your 
level of intention. Please respond honestly. 

 

Please indicate your level of intention for the 
following: 

4.  I intend to use prohibited substances or methods to 

enhance my performance during this season 
5.  At some point this season, I intend to use a 

prohibited substance or method to help improve my 
performance 

 

l

y
e
t
i
n
i
f
e
D

 
t
o
N

   

     

 

l

y
e
t
i
n
i
f
e
D

 

s
e
Y

1  2  3  4  5  6  7 

1  2  3  4  5  6  7 

6.  I will use a prohibited substance or method this 
season to help improve my athletic performance 

1  2  3  4  5  6  7 

 
 
 
 

 

THANK YOU FOR PARTICIPATING IN THIS STUDY! 

 

 

81 
 

APPENDIX C 

Study Two - Human Research Protection Program Approval Letter  

 

 

 

82 
 

 

 

 

83 
 

APPENDIX D 

Study Two – Questionnaire Packet 

 

 

 

84 
 

Please provide some information about yourself. 

1. Age (years): _____________        2. Sex:   Male □    Female □    Prefer not to say 
□ 

3. What is your main sport: ___________________________ 

4. Years competing in this sport: ______________________ 

5. Highest level you compete at in your main sport: 
Regional □     National □     International □ 
6. Highest level you have ever competed at in your main sport: 
Regional □     National □     International □ 
7. How many years have you been a member of your current team?  
_________________  
8. Average training/competition hours per week for your main sport: 
__________________ 
9. Weeks out of the year spent training/competing in your main sport: 
_________________ 
10. Please indicate your ethnicity:      Hispanic or Latino □       Not Hispanic or Latino 
□ 

11. Please indicate your race: 
          American Indian/Alaskan Native □ 
          Asian □           
          Black or African-American □ 
          White □ 
          More than one race □ 
          Unknown/Other □ 

 

 

 

12.  What is the name of your team/club/school? 

______________________________________ 

 

 

85 
 

Below you will find a list of statements. Please read each statement carefully and 
decide if that statement describes you or not. If it describes you, circle the word 
“true”; if not, check the word “false.” 
  

Please indicate whether each of the following statements is true or 
false for you. 
1.  I sometimes litter 
2.  I always admit my mistakes openly and face the potential negative 

consequences. 

3.  In traffic I am always polite and considerate of others. 

4.  I have tried illegal drugs (for example, marijuana, cocaine, etc.). 
5.  I always accept others’ opinions, even when they don’t agree with my 

own. 

 

 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

True  False 

6.  I take out my bad moods on others now and then. 
7.  There has been an occasion when I took advantage of someone else.  True  False 
8.  In conversations I always listen attentively and let others finish their 
True  False 

sentences. 

9.  I never hesitate to help someone in case of emergency. 

10. When I have made a promise, I keep it – no ifs, ands or buts. 

11. I occasionally speak badly of others behind their back. 

12. I would never live off other people. 
13. I always stay friendly and courteous with other people, even when I 

am stressed out. 

14. During arguments I always stay objective and matter-of-fact. 
15. There has been at least one occasion when I failed to return an item 

that I borrowed. 

16. I always eat a healthy diet. 

17. Sometimes I only help because I expect something in return. 

 

 

86 
 

Doping 

Doping is defined as use of prohibited substances or methods by an athlete with the 
potential to artificially improve performance through changes in physical and/or mental 
condition. Please keep this definition in mind when answering questions about doping. 

 

 

 

 

87 
 

A number of statements describing thoughts that athletes might have about 
doping are listed below. Please read these statements carefully and indicate your 
level of agreement with each one by circling the appropriate number. Please 
respond honestly. 
 

What  is  your  level  of  agreement  with  the 
following statements? 

1.  It is okay to dope if it helps an athlete to 

provide for his/her family. 

2.  Saying you "take steroids" feels worse than 

saying you "use some gear". 

3.  Compared to most lifestyles in the general 

public, doping isn't that bad. 

4.  Athletes shouldn't be blamed for doping if 

training partners/teammates pressure them 
to do it. 

5.  If most athletes in a sport dope, no one 
athlete should be held responsible for 
doing it. 

6.  Risks associated with doping are 

exaggerated. 

7.  Doping is okay if it helps an athlete advise 

others on how to do it right. 

8.  Using words like "roids", "gear" and 

"pinning" makes doping feel more 
acceptable. 

9.  Compared to smoking, doping is pretty 

safe. 

10. An athlete shouldn’t be blamed for doping 
if a member of his/her training group has 
encouraged it. 

11. It’s not right to condemn individuals who 
dope when many in their sport are doing 
the same. 

12. Doping doesn’t really harm anyone else. 
13. It is acceptable to dope if knowledge 

gained helps an athlete advise others on 
safe doping. 

88 
 

 

 

l

y
g
n
o
r
t
S

e
e
r
g
a
s
D

i

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

1 

 

e
e
r
g
a
s
D

i

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

2 

 

 

y

e
e
r
g
a
s
D

i

l
t
h
g

i
l

S

 
l

a
r
t
u
e
N

 

y

 

 

e
e
r
g
A

e
e
r
g
A

l
t
h
g

i
l

S

 

 

l

y
g
n
o
r
t
S

e
e
r
g
A

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

3 

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

4 

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

5 

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

6 

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

7 

1 

1 

1 

1 

1 

2 

2 

2 

2 

2 

3 

3 

3 

3 

3 

4 

4 

4 

4 

4 

5 

5 

5 

5 

5 

6 

6 

6 

6 

6 

7 

7 

7 

7 

7 

14. Using terms such as "gear" or "juice" 

makes doping sound less harmful. 

15. Compared to physical violence, doping 

isn't that serious. 

16. An athlete shouldn't be held responsible for 

doping if his/her coach encouraged 
him/her to do it. 

17. If an athlete trains/competes in an 

environment in which doping is the norm, 
he/she shouldn't be held accountable for 
doing it. 

18. The negative aspects of doping are 

exaggerated by the media. 

 
 
 

 

89 
 

Now, we would like to assess your perceptions about the prevalence of doping in 
sport. 

 

1.  Out of 100%, how many athletes at your competitive level do 
you believe engage in doping to enhance their performance?   

____________% 

2.  Out of 100%, how many elite athletes in your country do you 

think engage in doping to enhance their performance?   

____________% 

3.  Out of 100%, how many elite athletes do you believe will be 
engaged in doping during the next 3 years to enhance their 
performance? 

 
 
 
Now, we would like to assess your perceptions of others’ thoughts on doping in 
sport. 

____________% 

 
y
l
g
n
o
r
t
S

 
e
e
r
g
a
s
i
D

 
e
e
r
g
a
s
i
D

 
y
l
t
h
g

i
l

S

 
e
e
r
g
a
s
i
D

 
l
a
r
t
u
e
N

 
y
l
t
h
g

i
l

S

 
e
e
r
g
A

 
e
e
r
g
A

 
y
l
g
n
o
r
t
S

 
e
e
r
g
A

1 

2 

3 

4 

5 

6 

7 

1 

2 

3 

4 

5 

6 

7 

1 

2 

3 

4 

5 

6 

7 

 

What  is  your  level  of  agreement  with  the 

following statements? 
1.  Most people who are important to me 

would want me to use doping substances 
or methods to enhance my performance. 
2.  Most people I know would agree if I used 

doping 
substances or methods to enhance my 

performance. 

3.  Most people who are close to me would 

like me to use doping substances or 
methods to enhance my performance. 

 
 
 
 
 
 
 

 
 

90 
 

Now, please imagine being in the following situation:  
 

“Having returned to training following a period of injury, you are feeling very out of 

shape. As such, you feel the need to get back in shape as soon as possible. A friend 
who you train with has been taking a training supplement that really helped them get 
back in shape quickly following a similar injury. They offer to give you some and you 
decide to take it. Subsequently you get back in shape much quicker than expected, but 

then discover the supplement you have been taking is a banned performance-

enhancing substance. However, due to the improvements you have experienced, you 

decide to continue taking the substance.” 

Now, using the following scale, please rate the extent to which you anticipate you 
would feel in the ways described below if you decided to continue taking the 
substance. Please answer honestly. 
 

 

 

 
l
l

a
 
t
a
 
t
o
N

1 

1 

1 

1 

1 

 

y

l

e
t
a
r
e
d
o
M

3 

3 

3 

3 

3 

 

e
l
t
t
i
l
 

A

2 

2 

2 

2 

2 

 

h
c
u
M
y
r
e
V

 

 

l

y
e
m
e
r
t
x
E

4 

4 

4 

4 

4 

5 

5 

5 

5 

5 

I would anticipate feeling the following 
about continuing to take the substance… 

1.  I would feel remorse, regret. 

2.  I would feel tension about what I was doing. 
 

3.  I would not be able to stop thinking about 

the bad thing I was doing. 

4.  I would feel like apologizing, confessing. 

5.  I would feel bad about what I was doing. 

 
 
 
 
 

 

 
 
 
 
 

91 
 

i.  Now, please imagine being in the following situation:  
 

“You have been offered a banned performance-enhancing substance under medical 

supervision at low or no financial cost and the banned performance-enhancing 

substance could make a significant difference to your performance and was currently 

not detectable” 

Now, please indicate: 

How much consideration would you give to the offer? 

 

 

 

 

 

None 
at all 

1 

A lot of 
considerati

on 
7 

 
 
 

 

 

3 

2 

6 

5 

4 

 

THANK YOU FOR PARTICIPATING IN THIS STUDY! 

 

92 
 

 

 

 

 

 

 

 

 

 

 

 

 

REFERENCES 

 

 

 

 

93 
 

REFERENCES 

 

Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In Action control (pp. 

11-39). Springer, Berlin, Heidelberg. 

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision 

Processes, 50(2), 179–211. doi:10.1016/0749-5978(91)90020-t 

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review 

and recommended two-step approach. Psychological Bulletin, 103(3), 411–423. 
doi:10.1037/0033-2909.103.3.411 

Arandjelović, O. (2015). Doping use meta-analysis: Science seasoned with moralistic prejudice. 

Sports Medicine, 45(3), 443–444. doi:10.1007/s40279-015-0312-0 

Aubel, O., Lefèvre, B., Le Goff, J. M., & Taverna, N. (2019). The team effect on doping in 
professional male road cycling (2005‐2016). Scandinavian Journal of Medicine & 
Science in Sports, 29(4), 615-622. doi:10.1111/sms.13384 

Backhouse, S., McKenna, J., Robinson, S., & Atkin, A. (2007). International literature review: 
attitudes, behaviours, knowledge and education–drugs in sport: Past, present and future. 
Report to the World Anti-Doping Agency (WADA). 

Bandura, A. (1990). Selective activation and disengagement of moral control. Journal of Social 

Issues, 46(1), 27–46. doi:10.1111/j.1540-4560.1990.tb00270.x 

Bandura, A. (1991). Social cognitive theory of moral thought and action. In Handbook of moral 

behavior and development (pp. 69-128). Psychology Press. 

Bandura, A. (1991). Social cognitive theory of self-regulation. Organizational Behavior and 

Human Decision Processes, 50(2), 248–287. doi:10.1016/0749-5978(91)90022-l 

Barkoukis, V., Brooke, L., Ntoumanis, N., Smith, B., & Gucciardi, D. F. (2019). The role of the 

athletes’ entourage on attitudes to doping. Journal of Sports Sciences, 37(21), 2483–
2491. doi:10.1080/02640414.2019.1643648 

Barkoukis, V., Kartali, K., Lazuras, L., & Tsorbatzoudis, H. (2016). Evaluation of an anti-doping 

intervention for adolescents: Findings from a school-based study. Sport Management 
Review, 19(1), 23–34. doi:10.1016/j.smr.2015.12.003 

Barkoukis, V., Lazuras, L., & Tsorbatzoudis, H. (2014). Beliefs about the causes of success in 

sports and susceptibility for doping use in adolescent athletes. Journal of Sports Sciences, 
32(3), 212–219. doi:10.1080/02640414.2013.819521 

Barkoukis, V., Lazuras, L., Tsorbatzoudis, H., & Rodafinos, A. (2013). Motivational and social 

cognitive predictors of doping intentions in elite sports: An integrated approach. 

94 
 

Scandinavian Journal of Medicine & Science in Sports, 23, e330-e340. 
doi:10.1111/sms.12068 

Bhambhani, Y. (2002). Physiology of wheelchair racing in athletes with spinal cord injury. 

Sports Medicine, 32(1), 23–51. doi:10.2165/00007256-200232010-00002 

Bhambhani, Y., Mactavish, J., Warren, S., Thompson, W. R., Webborn, A., Bressan, E., … 

Vanlandewijck, Y. (2010). Boosting in athletes with high-level spinal cord injury: 
knowledge, incidence and attitudes of athletes in paralympic sport. Disability and 
Rehabilitation, 32(26), 2172–2190. doi:10.3109/09638288.2010.505678 

Boardley, I. D., Grix, J., & Dewar, A. J. (2014). Moral disengagement and associated processes 

in performance-enhancing drug use: A national qualitative investigation. Journal of 
Sports Sciences, 32(9), 836–844. doi:10.1080/02640414.2013.862842 

Boardley, I. D., Grix, J., & Harkin, J. (2014). Doping in team and individual sports: A qualitative 

investigation of moral disengagement and associated processes. Qualitative Research in 
Sport, Exercise and Health, 7(5), 698–717. doi:10.1080/2159676x.2014.992039 

Boardley, I., Smith, A. L., Mills, J. P., Grix, J., & Wynne, C. (2017). Empathic and self-

regulatory processes governing doping behavior. Frontiers in Psychology, 8(1495). 
doi:10.31236/osf.io/dxqng 

Boardley, I., Smith, A. L., Mills, J. P., Grix, J., Wynne, C., & Wilkins, L. (2018). Development 
of moral disengagement and self-regulatory efficacy assessments relevant to doping in 
sport and exercise. Psychology of Sport and Exercise, 36, 57-70. 
doi:10.31236/osf.io/y8v4q 

Boardley, I. D., Smith, A. L., Ntoumanis, N., Gucciardi, D. F., & Harris, T. S. (2019). 
Perceptions of coach doping confrontation efficacy and athlete susceptibility to 
intentional and inadvertent doping. Scandinavian Journal of Medicine & Science in 
Sports, 29(10), 1647–1654. doi:10.1111/sms.13489 

Borsari, B., & Carey, K. B. (2003). Descriptive and injunctive norms in college drinking: A 

meta-analytic integration. Journal of Studies on Alcohol, 64(3), 331–341. 
doi:10.15288/jsa.2003.64.331 

Bredemeier, B. J., & Shields, D. L. (1984). Divergence in moral reasoning about sport and 
everyday life. Sociology of Sport Journal, 1(4), 348–357. doi:10.1123/ssj.1.4.348 

Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative conduct: 

Recycling the concept of norms to reduce littering in public places. Journal of 
Personality and Social Psychology, 58(6), 1015–1026. doi:10.1037/0022-3514.58.6.1015 

Connor, J., Woolf, J., & Mazanov, J. (2013). Would they dope? Revisiting the Goldman 

dilemma. British Journal of Sports Medicine, 47(11), 697–700. doi:10.1136/bjsports-
2012-091826 

95 
 

De Jong, R., Vanreusel, B., & Driel, R. van. (2010). Relationships between mainstream 
participation rates and elite sport succes in disability sports. European Journal of 
Adapted Physical Activity, 3(1), 18–29. doi:10.5507/euj.2010.002. 

Donovan, R. J., Egger, G., Kapernick, V., & Mendoza, J. (2002). A conceptual framework for 

achieving performance enhancing drug compliance in sport. Sports Medicine, 32(4), 269–
284. doi:10.2165/00007256-200232040-00005 

European Union (2011). Communication on sport: Developing the European dimension in sport. 

Belgium: Publications Office of the European Union. doi:10.2766/16859 

Erickson, K., Backhouse, S. H., & Carless, D. (2017). Doping in sport: Do parents matter? Sport, 

Exercise, and Performance Psychology, 6(2), 115–128. doi:10.1037/spy0000081 

Evans, M. B., Eys, M. A., & Bruner, M. W. (2012). Seeing the “we” in “me” sports: The need to 

consider individual sport team environments. Canadian Psychology/Psychologie 
Canadienne, 53(4), 301–308. doi:10.1037/a0030202 

Finney, S. J., & DiStefano, C. (2013). Non-normal and categorical data in structural equation 
modeling. In G.R. Hancock and R.O. Mueller (Eds.), Structural equation modeling: A 
second course (2nd ed., pp. 439-492). Charlotte, NC: Information Age Publishing Inc. 

Gucciardi, D. F., Jalleh, G., & Donovan, R. J. (2010). Does social desirability influence the 

relationship between doping attitudes and doping susceptibility in athletes? Psychology of 
Sport and Exercise, 11(6), 479–486. doi:10.1016/j.psychsport.2010.06.002 

Gucciardi, D. F., Jalleh, G., & Donovan, R. J. (2011). An examination of the Sport Drug Control 

Model with elite Australian athletes. Journal of Science and Medicine in Sport, 14(6), 
469–476. doi:10.1016/j.jsams.2011.03.009 

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: 

Conventional criteria versus new alternatives. Structural Equation Modeling: A 
Multidisciplinary Journal, 6(1), 1–55. doi:10.1080/10705519909540118 

Hurst, P., Ring, C., & Kavussanu, M. (2020). An evaluation of UK athletics’ clean sport 

programme in preventing doping in junior elite athletes. Performance Enhancement & 
Health, 7(3-4), 100155. doi:10.1016/j.peh.2019.100155 

IOC, I. (2015). Olympic charter. Lausanne, Switzerland: International Olympic Committee. 

IPC (2018). IPC Anti-doping code. Bonn, Germany: International Paralympic Committee. 

Retrieved from https://www.paralympic.org/sites/default/files/2020-
06/Sec%20ii%20chapter%201_2_IPC%20Anti-Doping%20Code.pdf 

Kavussanu, M., Boardley, I. D., Sagar, S. S., & Ring, C. (2013). Bracketed morality revisited: 

How do athletes behave in two contexts? Journal of Sport and Exercise Psychology, 
35(5), 449–463. doi:10.1123/jsep.35.5.449 

96 
 

Kavussanu, M., Hatzigeorgiadis, A., Elbe, A.-M., & Ring, C. (2016). The moral disengagement 

in doping scale. Psychology of Sport and Exercise, 24, 188-198. 
doi:10.1016/j.psychsport.2016.02.003 

Kavussanu, M., & Ring, C. (2017). Moral identity predicts doping likelihood via moral 

disengagement and anticipated guilt. Journal of Sport and Exercise Psychology, 39(4), 
293–301. doi:10.1123/jsep.2016-0333 

Kavussanu, M., Ring, C., & Kavanagh, J. (2014). Antisocial behavior, moral disengagement, 

empathy and negative emotion: A comparison between disabled and able-bodied athletes. 
Ethics & Behavior, 25(4), 297–306. doi:10.1080/10508422.2014.930350  

Kayser, B., Mauron, A., & Miah, A. (2007). Current anti-doping policy: A critical appraisal. 

BMC Medical Ethics, 8(1). doi:10.1186/1472-6939-8-2 

Lauré, P. (1997). Epidemiologic approach of doping in sport. A review. The Journal of Sports 

Medicine and Physical Fitness, 37, 218-224. 

Lazuras, L., Barkoukis, V., Mallia, L., Lucidi, F., & Brand, R. (2017). More than a feeling: The 

role of anticipated regret in predicting doping intentions in adolescent athletes. 
Psychology of Sport and Exercise, 30, 196–204. doi:10.1016/j.psychsport.2017.03.003 

Lazuras, L., Barkoukis, V., Rodafinos, A., & Tzorbatzoudis, H. (2010). Predictors of doping 

intentions in elite-level athletes: A social cognition approach. Journal of Sport and 
Exercise Psychology, 32(5), 694–710. doi:10.1123/jsep.32.5.694 

Lazuras, L., Barkoukis, V., & Tsorbatzoudis, H. (2015). Toward an integrative model of doping 

use: An empirical study with adolescent athletes. Journal of Sport and Exercise 
Psychology, 37(1), 37–50. doi:10.1123/jsep.2013-0232 

Legg, D., & Mason, D. S. (1998). Autonomic dysreflexia in wheelchair sport: A new game in the 

legal arena. Marquette Sports Law Journal, 8, 225. 

Lucidi, F., Zelli, A., Mallia, L., Grano, C., Russo, P. M., Violani, C. (2008). The social-cognitive 
mechanisms regulating adolescents' use of doping substances. Journal of Sports Sciences, 
26(5), 447-456. doi: 10.1080/02640410701579370 

Lustria, M. L. A., Noar, S. M., Cortese, J., Van Stee, S. K., Glueckauf, R. L., & Lee, J. (2013). A 
meta-analysis of web-delivered tailored health behavior change interventions. Journal of 
Health Communication, 18(9), 1039–1069. doi:10.1080/10810730.2013.768727 

Maennig, W. (2014). Inefficiency of the anti-doping system: Cost reduction proposals. Substance 

Use & Misuse, 49(9), 1201–1205. doi:10.3109/10826084.2014.912065 

Mazzeo, F., Santamaria, S., & Iavarone, A. (2015). “Boosting” in Paralympic athletes with spinal 

cord injury: Doping without drugs. Functional neurology, 30(2), 91. 
doi:10.11138/fneur/2015.30.2.091 

97 
 

Morente-Sánchez, J., & Zabala, M. (2013). Doping in sport: A review of elite athletes’ attitudes, 
beliefs, and knowledge. Sports Medicine, 43(6), 395–411. doi:10.1007/s40279-013-0037-
x 

Muthén, L. K., & Muthén, B. O. (1998-2015). Mplus User’s Guide (7th ed.). Los Angeles, CA: 

Muthén & Muthén.  

Nicholls, A. R., Lucas, R. W., Plata-Andrés, M., Bailey, R., Cope, E., Madigan, D. J., Koenen, 

K., Glibo, I., Theodorou, N. C., Laurent, J., Garcia, G., & Chanal, B. (in press). A 
feasibility randomised controlled trial examining the effects of the Anti-Doping Values in 
Coach Education (ADVICE) mobile application on doping knowledge and attitudes 
towards doping among grassroots coaches. BMJ Open Sport & Exercise Medicine.  

Ntoumanis, N., Gucciardi, D. F., Backhouse, S. H., Barkoukis, V., Quested, E., Patterson, L., … 

Kaffe, S. (2018). An intervention to optimize coach motivational climates and reduce 
athlete willingness to dope (CoachMADE): Protocol for a cross-cultural cluster 
randomized control trial. Frontiers in Psychology, 8. doi:10.3389/fpsyg.2017.02301 

Ntoumanis, N., Ng, J. Y. Y., Barkoukis, V., & Backhouse, S. (2014). Personal and psychosocial 
predictors of doping use in physical activity settings: A meta-analysis. Sports Medicine, 
44(11), 1603–1624. doi:10.1007/s40279-014-0240-4 

Ntoumanis, N., Quested, E., Patterson, L., Kaffe, S., Backhouse, S. H., Pavlidis, G., … 

Gucciardi, D. F. (2020). An intervention to optimise coach-created motivational climates 
and reduce athlete willingness to dope (CoachMADE): A three-country cluster 
randomised controlled trial. British Journal of Sports Medicine. doi:10.1136/bjsports-
2019-101963 

Perkins, H. W., & Berkowitz, A. D. (1986). Perceiving the community norms of alcohol use 

among students: Some research implications for campus alcohol education programming. 
International Journal of the Addictions, 21(9-10), 961–976. 
doi:10.3109/10826088609077249 

Petróczi, A., & Aidman, E. (2008). Psychological drivers in doping: The life-cycle model of 
performance enhancement. Substance Abuse Treatment, Prevention, and Policy, 3(1). 
doi:10.1186/1747-597x-3-7 

Petróczi, A., Mazanov, J., Nepusz, T., Backhouse, S. H., & Naughton, D. P. (2008). Comfort in 

big numbers: Does over-estimation of doping prevalence in others indicate self-
involvement? Journal of Occupational Medicine and Toxicology, 3(1), 19. 
doi:10.1186/1745-6673-3-19 

Pitsch, W., & Emrich, E. (2011). The frequency of doping in elite sport: Results of a replication 

study. International Review for the Sociology of Sport, 47(5), 559–580. 
doi:10.1177/1012690211413969 

98 
 

Pope, H. G., Wood, R. I., Rogol, A., Nyberg, F., Bowers, L., & Bhasin, S. (2014). Adverse 

health consequences of performance-enhancing drugs: An endocrine society scientific 
statement. Endocrine Reviews, 35(3), 341–375. doi:10.1210/er.2013-1058 

Rimer, B. K., & Kreuter, M. W. (2006). Advancing tailored health communication: A persuasion 

and message effects perspective. Journal of Communication, 56, S184–S201. 
doi:10.1111/j.1460-2466.2006.00289.x 

Ring, C., & Hurst, P. (2019). The effects of moral disengagement mechanisms on doping 

likelihood are mediated by guilt and moderated by moral traits. Psychology of Sport and 
Exercise, 40, 33–41. doi:10.1016/j.psychsport.2018.09.001  

Ring, C., & Kavussanu, M. (2017). The role of self-regulatory efficacy, moral disengagement 
and guilt on doping likelihood: A social cognitive theory perspective. Journal of Sports 
Sciences, 36(5), 578–584. doi:10.1080/02640414.2017.1324206 

Ring, C., Kavussanu, M., & Gürpınar, B. (2020). Basic values predict doping likelihood. Journal 

of Sports Sciences, 38(4), 357–365. doi:10.1080/02640414.2019.1700669 

Ring, C., Kavussanu, M., & Mazanov, J. (2019). Self-other judgments of doping likelihood and 
anticipated guilt in hypothetical situations. Psychology of Sport and Exercise, 41, 46–53. 
doi:10.1016/j.psychsport.2018.11.006 

Ring, C., Kavussanu, M., Simms, M., & Mazanov, J. (2018). Effects of situational costs and 
benefits on projected doping likelihood. Psychology of Sport and Exercise, 34, 88–94. 
doi:10.1016/j.psychsport.2017.09.012 

Rivis, A., & Sheeran, P. (2003). Descriptive norms as an additional predictor in the theory of 

planned behaviour: A meta-analysis. Current Psychology, 22(3), 218–233. 
doi:10.1007/s12144-003-1018-2 

Sage, L., Kavussanu, M., & Duda, J. (2006). Goal orientations and moral identity as predictors of 

prosocial and antisocial functioning in male association football players. Journal of 
Sports Sciences, 24(5), 455–466. doi:10.1080/02640410500244531 

Sparkes, A. C., & Brighton, J. (2019). Autonomic dysreflexia and boosting in disability sport: 

exploring the subjective meanings, management strategies, moral justifications, and 
perceptions of risk among male, spinal cord injured, wheelchair athletes. Qualitative 
Research in Sport, Exercise and Health, 12(3), 414–430. 
doi:10.1080/2159676x.2019.1623298 

Stapleton, J. N., Perrier, M.-J., Campbell, D. S., Tawse, H. L., & Martin Ginis, K. A. (2016). 

Social cognitive predictors of competitive level among athletes with physical disabilities. 
Psychology of Sport and Exercise, 22, 46–52. doi:10.1016/j.psychsport.2015.06.005 

Strelan, P., & Boeckmann, R. J. (2006). Why drug testing in elite sport does not work: 

Perceptual deterrence theory and the role of personal moral beliefs. Journal of Applied 
Social Psychology, 36(12), 2909–2934. doi:10.1111/j.0021-9029.2006.00135.x 

99 
 

Sullivan, P. J., Feltz, D. L., LaForge-MacKenzie, K., & Hwang, S. (2015). The preliminary 

development and validation of the Doping Confrontation Efficacy Scale. Psychology of 
Sport and Exercise, 16, 182–190. doi:10.1016/j.psychsport.2014.04.011 

Sullivan, P., & Razavi, P. (2017). Are athletes' doping-related attitudes predicted by their 

perceptions of coaches' confrontation efficacy? Substance Use & Misuse, 52(8), 1098-
1103. doi: 10.1080/10826084.2016.1272613 

Tabachnick, B. G., & Fidell, L. S. (2013). Using Multivariate Statistics (6th ed.). New York, 

NY: Pearson.  

Uvacsek, M., Nepusz, T., Naughton, D. P., Mazanov, J., Ránky, M. Z., & Petróczi, A. (2011). 
Self-admitted behavior and perceived use of performance-enhancing vs psychoactive 
drugs among competitive athletes. Scandinavian Journal of Medicine & Science in 
Sports, 21(2), 224–234. doi:10.1111/j.1600-0838.2009.01041.x 

Van de Vliet, P. (2012). Antidoping in Paralympic sport. Clinical Journal of Sport Medicine, 1. 

doi:10.1097/jsm.0b013e31824206af 

WADA. (2019). Who we are. Retrieved from https://www.wada-ama.org/en/who-we-are 

WADA. (2021). World anti-doping code. Retrieved from https://www.wada-

ama.org/sites/default/files/resources/files/2021_code.pdf 

Westmattelmann, D., Dreiskämper, D., Strauß, B., Schewe, G., & Plass, J. (2018). Perception of 

the current anti-doping regime – A quantitative study among German top-level cyclists 
and track and field athletes. Frontiers in Psychology, 9. doi:10.3389/fpsyg.2018.01890 

Whitaker, L., Backhouse, S. H., & Long, J. (2017). Doping vulnerabilities, rationalisations and 
contestations: The lived experience of national level athletes. Performance Enhancement 
& Health, 5(4), 134-141. doi:10.1016/j.peh.2017.06.001 

Woolway, T., Lazuras, L., Barkoukis, V., & Petróczi, A. (2020). “Doing what is right and doing 

it right”: A mapping review of athletes' perception of anti-doping legitimacy. 
International Journal of Drug Policy, 84, 102865. doi:10.1016/j.drugpo.2020.102865 

Yuan, K. H., & Bentler, P. M. (2000). Robust mean and covariance structure analysis through 

iteratively reweighted least squares. Psychometrika, 65(1), 43-58. 
doi:10.1007/bf02294185 

 

 

100