Sage Journals: Discover world-class research

Abstract

Although match fixing threatens the integrity of competitions in sport, studies on the prevalence of match fixing are scarce. We measured the prevalence of competition manipulation by German elite athletes and the total percentage of these athletes who had been asked to participate in match fixing by using the randomized response technique. Approximately 8% of the athletes were asked to participate in match fixing, and approximately 7.5% were actually involved in competition manipulation during their careers. More than 30% of athletes reported an attempt to illegally influence referees’ decisions, but only 4.9% had ever directly participated in such attempts. Only the parameter of the financial status provides a different perspective. In general, this study shows that the dissemination of deviant behavior is not extremely high.

Keywords

match fixing competition fraud randomized response technique elite athletes cheater detection prevalence German squad athletes Olympic sport

Introduction

Cases of corruption, especially match fixing and competition fraud,¹ are not limited to the current discussion revolving around FIFA (Fédération Internationale de Football Association) and DFB (German Football Association). These topics are also frequently addressed in the popular media. However, many of the news reports lack a rigorous scientific analysis and are often deliberately sensationalized and fishing for attention. Public discussion and reports in the media are often limited only to the deviant behavior of athletes, namely,

Gambling competition fraud linked with specific interests of bettors,

Payment of bribes with the objective of competition fraud, and/or

Competition fraud or match fixing in specific sports (e.g., badminton during the 2012 Olympic Games in London or suspected cases in football²; see Table 1a in Maennig, 2008).

For analytical clarity, different forms of deviance in this field can be distinguished. In accordance with Hill (2013; see also Boeri & Severgnini, 2013; for a theoretical framework, see Emrich & Pierdzioch, 2015), we will distinguish between match fixing as follows:

For betting purposes (sports exogenous, financial motive) or “gambling match fixing” and

Match fixing to gain a competitive advantage (sports endogenous, ideal motive).

The first type of fixing is driven by the material interests of often non-sport-involved parties (i.e., profit maximization), while the second type is driven mainly by immaterial interests, which sometimes can later be transferred into economic benefits. The fixing act is the same; only the rewards and motivation differ. Although both forms threaten the integrity of sporting competitions, the second one is arguably more acceptable from the point of view of players and spectators.

Match fixing and competition fraud harm the integrity of competition (Emrich & Pierdzioch, 2015; Emrich, Pierdzioch, & Pitsch, 2014), which is a complex idea. It means that athletes deliver high performances while competing fairly under the rules of respective sporting codes but especially of Olympic sports, because morality plays a larger role there (Emrich et al., 2014).³ Athletes are expected to be “true athletes”⁴ and to guarantee the uncertainty of outcome of competitions by respecting the rules and playing fair. If this expectation is belied by detected manipulations, fans and consumers can reasonably be expected to loose interest in the competition and to quit watching (for a game theoretical modeling of this problem, see Büchel, Emrich, & Pohlkamp, 2014). Depending on the objective of the manipulation, whether winning a competition, monetary profit by sports betting fraud, or a combination of both, manipulation brings both the corruptor(s) and the corruptee(s) benefits as long as it remains undetected.

The integrity of sports competitions is an asset deserving of special protection in our society, from the point of view of the organizers of sporting events as well as of the athletes themselves and of consumers. National organizations, such as the German Football League (DFL), consider integrity “the greatest asset” (Deutschlandfunk, 2014). Resolutions passed by the German Olympic Sports Association (2013), the Union of European Football Associations (2014), the Council for Europe’s Convention (Council of Europe, 2014), and draft laws (Bayerisches Staatsministerium der Justiz, 2014) also attest to the significance attributed to the integrity of sports competitions.

Because we try to address this threat systematically, we need to measure the prevalence of this type of deviance and analyze the causes of match fixing and competition fraud. In the following article, we use the randomized response technique (RRT) to determine the extent of match fixing and competition fraud in elite German sport to reduce false positives. If you have any questions about sensitive topics, you may expect distortions based on social desirability. By anonymizing the question asked (indirect question), RRT offers the possibility of nevertheless appreciating the true result. Our article raises the following research issues:

Research Question 1: What is the proportion of elite athletes who, over the course of their career, have ever been actively involved in competitive manipulations and/or have received an invitation to participate in this type of corrupt activity?

Research Question 2: What is the percentage of athletes who have actively exerted influence on a referee or official to manipulate a competition?

Research Question 3: How many athletes have observed an active effort to influence a referee or official?

Research Question 4: Which socioeconomic and biographical variables affect the prevalence of corruption in sports?⁵

Our approach to the subject begins with a review of the current literature on the significance of integrity in athletic competitions. Next, the empirical methods are explained, and the results are presented. The article ends with a brief discussion.

State of the Research⁶

Different forms of deviant behavior such as doping or match fixing by participants jeopardize the integrity of athletic competitions. Surveys of spectators at the Olympic Games indicate that the consumers, that is, the spectators at sports competitions, likewise consider this integrity threatened primarily through doping, corruption, and commercialization (see “Olympia task force” at Johannes Gutenberg University in Mainz cited in Emrich et al., 2014).

A long list of publications has recently studied types and patterns of match fixing and competition fraud in sports, often based on single case analysis (Feltes, 2013; Petropoulos & Maguire, 2013), on single countries or regions (Qureshi & Verma, 2013; Richard, 2013; Tak, 2018; Tak, Sam, & Jackson, 2016) or on single sport events (Boeri & Severgnini, 2013; Cheloukhine, 2013; Harris, 2013).

As most of these studies contain suggestions to curb this form of deviance, suggestions derived from the presented cases, they can serve as a starting point to develop a middle-range theory linking special knowledge about the issues studied with highly abstract theories on the impact of determinants on deviance (such as microeconomics, for example, Forrest, 2013, or game theory, for example, Hakeem, 2013).

In soccer, an investigation analyzed case studies such as the Hoyzer scandal⁷ in Germany or the Italian scandal (Feltes, 2013). An early warning system seems to be ineffective, so that other measures (training, education, communication, or cooperation between stakeholders) are proclaimed. These suggestions widely concur with the INTERPOL/FIFA initiative to combat match fixing in soccer (Abbott & Sheehan, 2013), as well as with the role assigned to academics in the fight for the integrity of sport (Segal, 2013). These papers primarily address individual morality as a key factor in causing match fixing.

However, there are other studies that identify the impact of the legal framework for sport and sports competitions on the probability of (mostly) gambling match fixing. For Uganda, Richard (2013) identified the role of rules as well as the impact of a nonfunctioning control system as a result of his analyses of interviews. Similarly, changes to the legal and to the law enforcement system to fight match fixing are suggested for cricket in India (Qureshi & Verma, 2013) and for sports in Greece and Ireland (Petropoulos & Maguire, 2013) or Australian cricket, rugby, and A-league soccer (Misra, Anderson, & Saunders, 2013). For South Korean “K-League,” Tak et al. (2016) showed that the countermeasures introduced after the football scandal (education program, change of regulations and sports betting monitoring) have had a positive effect on the loss of legitimacy.

From another perspective, Tak (2018) considers the attributed responsibilities between institutions and individuals. Using documents and interviews with stakeholders in South Korea (football and motorboat racing) as a multimethod analysis, he notes that the attribution of institutional misconduct to the individuals is functional to protect the institution’s strength. Individual misconduct is perceived as a major crime because the institution is “too big to fail.”

The idea of curbing competition rigging by changes to sports-related laws is strongly supported by the regression analyses of Duggan and Levitt (2002). This study compared the empirical distribution of a large number of competition results in sumo wrestling to a theoretically assumed binomial distribution. Although not building on detected cases of competition fraud, their statistical analysis provided strong evidence that the incentive structure in sumo wrestling promoted the probability of corruption and of illegal trades among athletes (Duggan & Levitt, 2002).

For international tennis, Gunn and Rees (2008) show that match fixing is not linked to mafia structures or a general level of corruption in tennis. According to this analysis, it is suggested that people are vulnerable within the tennis sphere and that there are people outside tennis who try to exploit the vulnerable. Although these determinants are settled at the level of individual agents, they suggest changes to the law and law enforcement system, leading to a suggested “IT-based Intelligence and Case Management system” (Gunn & Rees, 2008, p. 26). Similarly, the IRIS (Institut de Relations Internationales et Strategique) study, explicitly aimed by definition at corruption in connection with betting, identified individual as well as structural determinants of corruption in sport, but recommended measures solely at the organizational/institutional level (IRIS, 2012; for EU members Asser Institute, Centre for International & European Law, 2014).

Determinants at the individual levels of athletes and referees are also identified by Boeri and Severgnini (2011, 2013). This work sought out the economic influences on match fixing or other specific determinants such as the date in the season when match fixing typically occurs. The most important determinants of match fixing in their models were individual economic drivers in players as well as the career concerns of referees.

In one of the rare cross-national studies, Tag, Sam, and Jackson (2018) considered match fixing as a social problem from the perspective of utility-maximizing actors based on different qualitative sources (policies, media releases, and scholarly works). The three core areas that are attributed to match fixing are criminal organizations and illegal betting markets, vulnerable individuals, and failure of governance.

Hill’s (2013) study differs in two ways from the research presented thus far. Although focusing on football and on match fixing betting, this study attempts to move beyond single-case analyses by developing different databases, compiling typical patterns of single cases, and thus enabling the first quantitative analyses. In addition, this publication offers recommendations for measuring individual morals, individual economic drivers, and the impact of law and law enforcement, thus accounting for the complex and multilevel phenomenon of match fixing.

Thus, the analysis by Preston and Szymanski (2003), for instance, went beyond single-case examples. Although their analysis was built on unique cases of match fixing in cricket, the authors were able to mathematically model the abstract patterns behind this kind of sporting deviance (Preston & Szymanski, 2003).

Although, as shown above, most of studies published thus far deal with determinants of match fixing derived from single-case analyses, a systematic investigation into its prevalence is still extremely scarce. One study hinting at prevalence estimations was published by the International Centre for Sport Security (2014). This report, which deals with known cases during the previous 3 years, attempts to describe the extent of the problem in different countries. The authors suggest that they were only able to see “the tip of the iceberg,” but they supply no figures describing the total size of the sports events analyzed, and prevalence estimations are impossible to estimate from their data. Gunn and Rees (2008) identified 73 tennis matches with a suspicious betting pattern, but as they did not provide a figure for the total number of matches, which were analyzed, the information is not sufficient.

The long list of single-case analyses (or, rather, few cases) presented above may have led to an overestimation of the prevalence of match fixing and competition fraud in sport. This logic can be seen in a report on match fixing in international tennis published in a London newspaper. A study describes a large number of corruption cases and cites statements and confessions of individual athletes who claimed to have been involved in match fixing (Harris, 2013). The study claims that unnamed “tennis officials” complained that the “tennis world” was afraid of revealing the true extent of the corruption in the sport. An accurate measurement of the total amount of match fixing cases in tennis cannot be estimated on the basis of limited, qualitative interviews.

Accurate figures on the extent of match fixing and competition manipulation are required if we wish to discuss these issues from a social-scientific position. Moreover, sports organizations need a reliable estimation of the total extent of the problem to adequately curb this behavior.

The only study thus far that includes a quantitative estimation of prevalence issues for match fixing using direct methods was published by the FIFPro Task Force Group for professional football in Eastern Europe in 2012. Using direct questioning, this study revealed that 11.9% of players were approached to consider match fixing and that 23.6% were aware that match fixing took place in their league. Nevertheless, the proportion of athletes who were ever involved in fixing matches remains unknown (FIFPro, 2012). An initial study on match fixing using RRT was carried out in amateur football in Germany, but as it was based on a relatively small sample, its results lack validity (Pitsch, Emrich, & Pierdzioch, 2015).

Method

Data Collection

An online-based RRT survey was conducted via postcards with a link to an online questionnaire from March to August 2013 with all-German A, B. and C elite athletes in Olympic sports.⁸ For the elite athletes of the Federal Republic of Germany, a file of the total population of all 5,548 elite athletes for the year 2005 was available. There is an estimated 10% fluctuation every year, that is, elite athletes who lose their status or others who join the national teams. This fluctuation is taken into account, because it is assumed that the willingness to provide answers on such a topic increases the more time has passed since the end of the athlete’s own career and the individual having left the national team.

These elite athletes were invited by letter to take part in the survey. Approximately 1,800 of the athletes contacted could not be reached by normal mail. A second letter in the form of a postcard was sent to the other 3,753 athletes. Furthermore, the athletes who were currently sponsored by “German Sports Aid”⁹ were also notified about the survey by the foundation. We have no additional information about the individuals who were contacted. A total of 425 athletes (79 from German Sports Aid information) participated in the survey. The return rate can, therefore, be estimated at approximately 11.3% of the contacted athletes.

Questionnaire

In accordance with these research parameters, the RRT survey included four questions:

Whether elite athletes have been approached to participate in manipulations,

Whether they have been personally involved in manipulations,

Whether they have knowledge that influence was exerted on a referee with the objective of competition rigging, and

Whether they have personally exerted an influence on a referee with the objective to manipulate a competition.

The online-based, standardized questionnaire contained additional questions about the sociodemographic data of the athlete’s career as well as direct questions about opinions and attitudes with respect to fairness, betting/gambling, and other forms of fraud in games. With all direct questions, care was taken (e.g., by determining the answer categories) to safeguard the respondents’ anonymity.

RRT—Data Analysis Processing

Numerous studies deal with the estimates of prevalence in socially sensitive behavior.¹⁰ It has been shown that direct questioning frequently leads to distorted results, because interviewees will often tailor their answers to provide more socially desirable responses. Thus, the prevalence of deviant behavior is usually underestimated in direct surveys, if the interviewees are actually willing to answer at all. The RRT (first developed by Warner, 1965, 1971; for an overview, see, for example, Lensvelt-Mulders, Hox, van der Heijden, & Maas, 2005; Wolter, 2012) is a method that offers protection in the form of anonymity to the interviewees, so that they can answer potentially embarrassing questions honestly.

The initial interest in this type of questioning is less focused on the behavior of the individual but, instead, on the percentage of “statistical units” in the collective, so that the loss of information at the personal level is accepted within the survey (for application areas of RRT in general, Lensvelt-Mulders et al., 2005, and in economic research, see, for example, Kerkvliet, 1994a, 1994b; in the area of deviant behavior in athletic competitions, see Dietz et al., 2013; Pitsch, Emrich, & Klein, 2005, 2007; Pitsch, Maats, & Emrich, 2009; Simon, Striegel, Aust, Dietz, & Ulrich, 2006; Striegel, 2008, in the area of competition-related drug use among recreational athletes or performance-enhancing drug consumption among students at different universities, see Frenger, Emrich, & Pitsch, 2016; Pitsch, Emrich, & Frenger, 2013).

Any inference about individual survey respondents is precluded by the method itself because of the indirect questioning. Anonymity is, therefore, protected. In this way, it is possible to achieve more reliable answers to questions about illegitimate and/or illegal practices than with direct questioning and the inevitable distortion effects it entails (Lensvelt-Mulders et al., 2005).

In addition to the questions about the criterion of competition fraud and match fixing, a supplementary instruction is given. The present study was conducted using the so-called “No” cheater detection (Feth, Frenger, Pitsch, & Schmelzeisen, 2017). In this study, we provide the option to answer one of the two questions (left or right; see the example in Figure 1). As a basis for the additional instruction, the respondents in the empirical study at hand were free to use the serial number of a euro note or one of the 10 randomly generated 10-digit numbers. Depending on that choice, the respondent must now answer the question on the left or on the right side. The question on the left (“Does every week have 7 days?”) should result in a “Yes” answer. The process is explained using an example in Figure 1.¹¹

Figure 1.

Example of a randomized response technique question similar to those asked from respondents.

The answers given and the actual existence of the property cannot be linked to one another by the researcher, as it is not known whether a response has been given due to the “say yes” instruction (here, the question “Does every week have 7 days?”) or due to the request to give an honest answer (here, answer the right question that asks for the embarrassing property). By knowing the distribution of the random numbers and their relative frequencies, probabilities result, according to which a respondent had received the instruction “answer yes” or “answer honestly.” The proportion of statistical units within the population can be indirectly and globally calculated only from the distribution of the answers and the known probabilities via maximum likelihood estimates in the idea of a conditional probability (see Feth et al., 2017).

Despite this protected procedure, cases occur in which the respondents do not follow the instructions (Coutts & Yann, 2008; Locander, Sudman, & Badburn, 1976; Shimizu & Bonham, 1978). The reasons for this cannot be researched by means of the method itself. Although deliberate manipulation is just as feasible as a misunderstanding of the instructions, in the literature on RRT methods, special techniques to monitor deviance from the instructions and to identify its proportion are labeled “cheater detection” methods. The first cheater detection variant of RRT was developed by Clark and Desharnais (1998). Further variants have been developed by Feth et al. (2017). As a result of the research on RRT, the method can be used for both exploratory prevalence estimation and for hypothesis testing. For those RRT variants with cheater detection, the distributions of error components of estimates is typically heavily skewed (Pitsch et al., 2013) due to an error component, which is introduced by the method when marginal cases (estimates close to zero or close to one) must be considered. This precludes in principle statistical tests that build on the precondition of normally distributed error components (e.g., t tests). Therefore, the bootstrap method to estimate (bias-corrected and accelerated) confidence intervals (Efron & Tibshirani, 1993) of the differences between estimates is recommended for statistical inference.¹² Using this method, the results of statistical tests are reported in a distinctive way (there is no t value or F value from an analysis of variance [ANOVA] or the degrees of freedom). In addition, the often-reported p value, indicating the estimated achieved significance level, becomes meaningless. Although this achieved significance level may as well be reported for bootstrap methods (e.g., Efron & Tibshirani, 1993) its value will vary between different bootstrap-simulations even if the null hypothesis were consistently rejected at the selected level for significance (here: p < .05).

Results

Descriptive Results

A total of 425 athletes participated in the survey. Because some participants did not answer particular questions (missing data), the figures of participants differ according to demographic characteristics. The gender ratio in the sample was divided between 52.3% male and 47.7% female (n = 396). The largest percentage (69.3%) was born in West Germany, while 25.3% were born in the GDR (German Democratic Republic), that is, the former East Germany, now the new federal states. A further 3.8% of the respondents were born outside Germany (n = 398). By means of the sample extraction, a proportion of athletes was expected to have already finished their athletic career. Their percentage stands at 59.1%; thus 40.9% of the athletes are currently still active (n = 397). In the context of being suspicious about manipulating competitions, the various financial situations of the athletes were analyzed (see the qualitative interviews; Kalb, Herrmann, & Emrich, 2015). In this context, the subjective evaluation of their own financial situation by the respondents was of particular interest. A total of 64.1% stated they were currently financially independent, while the other athletes depended on sponsorship through third parties (e.g., parents, German Sports Aid, n = 368; see Table 1).

Table 1.

Participant Demographic Details.

	Frequency (%)
Gender (n = 396)
Male	52.3
female	47.7
Born in . . . (n = 398)
(Western-)Germany	69.3
GDR/new federal states	26.9
Outside of Germany	3.8
Sporting career (n = 397)
Just started	3.3
Currently at its peak	27.7
Scheduled to finish this year	9.8
Finished before 2001	0.5
Finished between 2001 and 2003	3.0
Finished between 2004 and 2006	14.6
Finished between 2007 and 2009	24.9
Finished after 2009	16.1
Age (n = 398)
< 18	1.3
18-33	7.8
29-33	23.6
34-28	50.0
34-38	10.8
39-43	3.5
> 43	3.0
Financial status (n = 368)
I am currently financially independent.	64.1
I am currently dependent on third-party sponsorship (e.g., parents, i.e., sports help).	35.9
Perceived financial situation (n = 383)
I live in constant financial distress (e.g., permanent arrears on checking accounts and credit cards).	2.3
I feel financially overwhelmed (e.g., having problems coping with unexpected expenses).	15.9
I feel well looked after.	60.6
The financing of my current livelihood and pension benefits for the future is ensured.	21.1

Note. GDR = German Democratic Republic.

RRT Results

The empirical results of the four RRT questions are shown in Figures 2 and 3, with the bars indicating the respective percentages of honest “No” answers, cheaters, and honest “Yes” answers.

Figure 2.

Prevalence estimation using the randomized response technique with regard any experience and personal active involvement in a distortion of competition.

Figure 3.

Prevalence estimation using the randomized response technique with regard to experiencing and actively influencing a referee themselves with the objective of distorting competition (n = 351).

The prevalence of athletes who have been approached during their athletic career and asked to become involved in a competition manipulation was estimated at 8.42%. The estimated prevalence for participation in this form of manipulation was 7.47%. The estimation for cheating was 2.01%, or 9.96%, with the latter figure referring to an athlete’s own active involvement. Accordingly, the estimated proportion of the honest “No” answers in the case of the question about whether they had been approached to take part in a manipulation of competition was 89.57% and, in the case of the question about whether they had been actively involved in a manipulation of competition, 82.57% (cf. Figure 2).

Answering the question about whether they had ever experienced any influence on referees for the objective of manipulating competition (Figure 3), the prevalence was estimated at 32.99%. The estimated proportion of honest “Yes” respondents was 4.90% for the question about their own active exertion of influence on a referee. The estimated percentage of cheaters was at 5.44% with respect to an exertion of influence actually experienced and 4.62% with respect to their own active involvement therein. Correspondingly, the proportion of “No” answers was estimated at 61.56% and 90.48%, respectively.

When checking the differences in prevalence in terms of gender, current career status (active or already ended) and current financial status (i.e., currently financially independent or dependent on the sponsorship of third parties such as German Sports Aid, parents, etc.), it became apparent that only the financial status resulted in a significant difference in the prevalence with respect to the question about active exertion of influence on referees (cf. Table 2).

Table 2.

Different Prevalence Estimates by Financial Status (Bootstrap Significance Tests).

	Independent (n = 216)	Dependent (n = 118)	Difference	low/bcα	high/bcα	Significant at 5%
Honest “Yes”	0.0003	0.2265	0.2262	0.0209		*
Cheaters	0.0000	0.2848	0.2848	0.0542		*
Honest “No”	0.9997	0.4887	−0.5110		−0.1354	*

Further tests of the remaining RRT questions and factors of influence can be found in the tables of the supporting information (Supplemental Appendix tables).

Discussion

The study presented here was primarily conducted to generate an empirical description of the involvement of elite athletes in match fixing and competition fraud. The results add quantitative empirical evidence to the mostly single-case-based studies (few in number) presented thus far. In terms of the RRT questions, it should be noted that a share of 8.42% of the respondents had been approached over the course of their careers to take part in a distortion of competition. This result confirms the prevalence of approaching athletes for fixing purposes in sports, as suggested in the FIFPro study, which showed that 11.4% of professional soccer players had been approached to engage in fixing matches (FIFPro, 2012). The rather similar results from RRT and a direct question emphasize that being approached is not extremely embarrassing. Concerning the embarrassing RRT question for actual involvement in match fixing, the rate of admitters was 7.47%. In comparison to a similar study in amateur football (Pitsch et al., 2015), this is only about one half of the prevalence reported in that study. In addition, the share of cheaters was significantly higher for amateur football than in the present study. It is important that the results of the RRT questions be considered independently. In each case, we have seen corresponding proportions for the entire population, as we cannot connect the questions with each other.

Regarding the position of the referee, the estimated percentage of athletes who had exerted influence themselves was lower (prevalence interval = 4.90%-9.52%) than the percentage of those who had perceived that influence on a referee was exerted (prevalence interval = 32.99%-38.43%). As attempts to exert influence on a referee can be perceived by more than one athlete, the chance that an athlete has actually experienced an exertion of influence widely exceeds the prevalence of an athlete’s own exertion of influence on the referee. That is why the percentage of athletes who have perceived that influence on a referee was exerted for the objective of distorting competition cannot be equated with the portion of competitions or referees on which influence has been exerted.

The figures concerning the role of referees in match fixing taken from different studies are far from being comparable, but shed light onto these issues from different perspectives. The role of referees is noted as pivotal in match fixing due to their large influence on the course and results of matches (Hill, 2009). Figures rank referees third in the rate of successful fixes behind team administration (ranked first) and players (ranked second, Hill, 2013). A different study with direct questioning estimated the rate of soccer referees who were never approached to fix a match above 89% with significant influences from age and the duration of the referee’s career (Rullang, Gassmann, Emrich, & Pierdzioch, 2016). Our result adds to these results the rate of athletes who have actively addressed referees to manipulate a competition (4.90%-9.52%). Nevertheless, it should be noted that these figures may only hold for some populations (players and referees in soccer vs. athletes and referees in a large list of Olympic sports). Evidently, there is further need for empirical clarification in different sports from these various perspectives to arrive at coherent interpretations.

As our results show, financial status plays a pivotal role in the exertion of influence on referees: for financially independent athletes, the estimated rate of honest “yes” responders was as low as 0.03%, while for financially dependent athletes, this rate exceeded 22%. A similar effect of career insecurity was presented for the involvement of referees in match fixing (Boeri & Severgnini, 2011) as well as for cricket players (Hill, 2013; Preston & Szymanski, 2003; Qayyum, 1998). Even though we have sampled former and active athletes, the argument persists, as we were able to show that career status has no impact on the results (see Supplemental Appendix).

These results clearly point toward individual economic drivers as effective determinants of match fixing and competition fraud. Typically, from most Olympic sports, even top athletes cannot necessarily entail an enormous financial benefit (Frenger, Pitsch, & Emrich, 2012). Therefore, additional sources of economic security and independence may contribute to reducing this reason for vulnerability. At this point, attention should be paid to the fact that former and/or active national-team athletes, that is, athletes from different Olympic sports, were involved in the survey. Olympic medals and high rankings in international tournaments can constitute a status value, which does entail significant financial advantages.

Therefore, the group of athletes most vulnerable to potential corruption can be reasonably expected to be athletes who are facing financial difficulties. At the same time, U.K. media reports of possible match fixing in tennis claim some of the “big names” in tennis are involved in corruption (“Match-Fixing Experts Fear Corrupt Stars Will Be Playing at Wimbledon,” 2013). From our results, we would rather assume that corruption would mostly be among the players in the lower ranks of tennis who are struggling to find sponsorship or revenue directly from the sport.

In addition to financial status, further variables were examined, namely, gender, current career status, and variables in terms of the specific sport type, such as summer/winter sports, single versus team sports, and the nature of the interaction with opponents (direct vs. indirect; see Supplemental Appendix tables). There were no differences in prevalence of the groups when taking these factors into consideration.

One possibility is that there are other variables to explain the behaviors that were not considered or surveyed. Another possible explanation is that this type of deviant behavior does not follow any specific pattern and thus contains a kind of stochastic component. Our results do not support concerns toward measures against match fixing and competition fraud that build on revised laws and law enforcement systems (Gunn & Rees, 2008; Hill, 2013; IRIS, 2012; Misra et al., 2013; Petropoulos & Maguire, 2013; Qureshi & Verma, 2013; Richard, 2013). These measures typically threaten individuals independently of their affiliation to certain groups. Significant differences in the match fixing and competition fraud prevalence have raised concerns regarding the probability of success of these measures. These concerns are not substantiated by our research.

As mentioned earlier, the prevalence estimations in the present study are not linked to the motives of influencing (winning a competition, bets, qualifications, etc.; Emrich & Pierdzioch, 2015; Hill, 2013). However, we can assume that, from a financial perspective, the exertion of influence in this field is not comparable with cases that were discussed in the financially very lucrative types of sports such as basketball in the United States (SID, 2008) or football (Maennig, 2008). Derived benefits, which also have monetary effects, must nevertheless be assumed, especially because winning medals and championships is sometimes “rewarded” with high amounts. That, of course, differs among countries.

The most important limitation of this study is that it is derived from a relatively small sample size. The global prevalence estimation with “No” cheater detection is definitely feasible with a sample of 300 persons (Feth et al., 2017). In the simulations carried out to identify suitable sample size, it becomes apparent that the standard error of an estimate diminishes with an increasing sample size. Hence, larger samples result in considerably more stable estimates. For the overall estimations, this means that the results can well be considered as reliable. The significance tests between different groups, however, are based on the prevalence estimation from the subgroups, each of which contains 170 to 180 respondents. Therefore, the estimate for the subgroups is worse and discriminatory tests are more difficult. For this reason, the results of the study should be seen as provisional and ought to be replicated using larger numbers of respondents.

If national and international sports organizations are genuinely interested in finding evidence-based anticorruption policies, they should support this field of research by providing access to elite athletes.

The fact that we examined a range of different sports might also have obscured the analysis. The type of sport was not surveyed, but only the class of sport (team or individual, etc.), as it would have been too easy to identify an athlete if the individual sport had been identified. This limitation must nevertheless be accepted when trying to generate reliable prevalence estimations for deviant behavior. Highly exact differentiations among respondents, not only by sport but also by age, gender, and level of success, immediately reduce the belief in the anonymity of the study and thus directly reduce the capacity of the research to provide reliable estimations.

Supplemental Material

Appendix_(1) – Supplemental material for Corruption in Olympic Sports: Prevalence Estimations of Match Fixing Among German Squad Athletes

Supplemental material, Appendix_(1) for Corruption in Olympic Sports: Prevalence Estimations of Match Fixing Among German Squad Athletes by Monika Frenger, Eike Emrich and Werner Pitsch in SAGE Open

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is part of a project funded by the Federal Institute for Sports Science in Germany (BISP; AZ 071801/12).

ORCID iD

Monika Frenger

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Monika Frenger, lecturer and researcher at Saarland University. Research in the field of sports sociology and sports economics. Main topics: deviant behaviour, voluntary work, methodology and sports organisations.

Eike Emrich, professor at Saarland University. Research in Economics of Education, Sports Economics, Institutional Economics, Organisational Theory and Sports Sociology.

Werner Pitsch, lecturer and researcher at Saarland University, Institute for Sport Sciences. His work focuses on prevalence estimation of deviance in sport, especially doping as well as social and economic preconditions and effects of deviance.

References

Abbott

Sheehan

(2013). The INTERPOL approach to tackling match fixing in football. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 263-287). New York, NY: Springer.

Asser Institute, Centre for International & European Law. (2014). Study on risk assessment and management and prevention of conflict of interest in the prevention and fight against betting-related match-fixing in the EU 28 (Final report). Brussels, Belgium: European Commission: DG Education and Culture. Retrieved from http://ec.europa.eu/sport/news/2014/docs/asser-excecutive-summary_en.pdf

Bayerisches Staatsministerium der Justiz. (2014). Entwurf eines Gesetzes zum Schutze der Integrität des Sports [Draft law to protect the integrity of sport]. Retrieved from https://www.justiz.bayern.de/media/pdf/gesetze/sport.pdf

Boeri

Severgnini

(2011). Match rigging and the career concerns of referees. Labour Economics, 18, 349-359.

Boeri

Severgnini

(2013). Match rigging in Italian professional soccer: The economic determinants of corruption. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 15-30). New York, NY: Springer.

Boruch

R. F.

(1971). Maintaining confidentiality of data in educational research: A systemic analysis. The American Psychologist, 26, 413-430.

Büchel

Emrich

Pohlkamp

(2014). Nobody’s innocent: The role of customers in the doping dilemma. Journal of Sports Economics, 17, 767-789.

Cheloukhine

(2013). Match fixing in soccer: Organization, structure and policing. A Russian perspective. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 113-132). New York, NY: Springer.

Clark

S. J.

Desharnais

R. A.

(1998). Honest answers to embarrassing questions: Detecting cheating in the randomized response model. Psychological Methods, 3, 160-168.

10.

Council of Europe. (2014). Council of Europe convention on the manipulation of sports competitions (CETS No. 125). Retrieved from http://www.coe.int/en/web/conventions/full-list/-/conventions/treaty/215

11.

Coutts

Yann

(2008). Sensitive questions in online surveys: Experimental results for the randomized response technique (RRT) and the unmatched count technique (UCT) (Electronic version, ETH Zurich Sociology Working Paper, No. 3). Retrieved from http://ideas.repec.org/p/ets/wpaper/3.html

12.

Davison

A. C.

Hinkley

D. V.

(1997). Bootstrap methods and their applications. Cambridge, UK: Cambridge University Press.

13.

Deutschlandfunk. (2014). Deutsche Fußball-Liga: Das höchste Gut ist die Integrität des Wettbewerbs [German Football League: The highest good is the integrity of the competition]. Retrieved from http://www.deutschlandfunk.de/deutsche-fussball-liga-das-hoechste-gut-ist-die-integritaet.1346.de.html?dram:article_id=304795

14.

Dietz

Ulrich

Dalaker

Striegel

Franke

A. G.

Lieb

Simon

(2013). Associations between physical and cognitive doping: A cross-sectional study in 2.997 triathletes. PLoS ONE, 8(11), e78702. doi:10.1371/journal.pone.0078702

15.

Duggan

Levitt

S. D.

(2002). Winning isn’t everything: Corruption in sumo wrestling. American Economic Review, 92, 1594-1605.

16.

Efron

Tibshirani

(1993). An introduction to the bootstrap. New York, NY: Chapman and Hall.

17.

Emrich

Pierdzioch

(2015). Theoretische Rahmung [Theoretical framing]. In Emrich

Pierdzioch

Pitsch

(Eds.), Falsches Spiel im Sport. Analysen zu Wettbewerbsverzerrungen. Schriften des Europäischen Instituts für Sozioökonomie e.V [Wrong game in Sports. Analyzes of competitive distortions. Wrtitings of the European Institute for Socio-Economics e.V] (Vol. 10, pp. 15-44). Saarbrücken, Germany: Universaar.

18.

Emrich

Pierdzioch

Pitsch

(2014). Die “Marke” Olympia und die besondere Bedeutung von Vertrauenskriterien—Eine Geschichte von Markt, Macht und Moral [The “Brand” Olympia and the special meaning of Trust Criteria-A story of market, power and moral]. In Haut

(Ed.), Leistungssport als Konkurrenz der Nationen. Sozioökonomische Bedingungen und Effekte [Competitive sport as a competition of Nations. Socioeconomic conditions and effects] (pp. 89-116). Saarbrücken, Germany: Universaar.

19.

Emrich

Pitsch

(2011). The doping control market: Are investments into a faked honesty more rewarding than honesty itself? In Emrich

Pitsch

(Eds.), Sport and doping. The analysis of an antagonistic symbiosis (pp. 115-134). Frankfurt am Main, Germany: Peter Lang.

20.

Feltes

(2013). Match fixing in Western Europe. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 15-30). New York, NY: Springer.

21.

Feth

Frenger

Pitsch

Schmelzeisen

(2017). Schriften des Europäischen Instituts für Sozioökonomie e.V. Cheater-detection for randomized response-techniques (Vol. 12). Saarbrücken, Germany: Universaar.

22.

FIFPro. (2012). FIFPro black book Eastern Europe. The problems professional footballers encounter research. Retrieved from https://www.fifpro.org/images/documents-pdf/BLACK-BOOK.pdf

23.

Forrest

(2013). Match fixing: An economics perspective. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 177-197). New York, NY: Springer.

24.

Frenger

Emrich

Pitsch

(2013). How to produce the belief in clean sports which sells. Performance Enhancement & Health, 2, 210-215.

25.

Frenger

Emrich

Pitsch

(2016). Sport-induced substance use: An empirical study to the extent of sport-related substance use within a German sports association. PLoS ONE, 11(10), e0165103. doi:10.1371/journal.pone.0165103

26.

Frenger

Pitsch

Emrich

(2012). Erfolg(+)reich und verdorben. Eine empirische Überprüfung verbreiteter Vorurteile im Sport [Success (+) rich and spoiled. An empirical review of common Prejudices in sports]. Sportwissenschaft, 42, 188-201.

27.

German Olympic Sports Association. (2013). Die Integrität des sportlichen Wettbewerbs sichern—Doping und Wettbetrug konsequent bekämpfen. Antrag des DOSB-Präsidiums und des Deutschen Tischtennis-Bundes. Beschluss der 9. Mitgliederversammlung am 7. Dezember 2013 [The integrity of the ensure athletic competition-doping and betting fraud fight consistently. Application of the DOSB-Presidium and the German Table Tennis Federation. Resolution of the 9th General Meeting on December 7, 2013]. Retrieved from https://cdn.dosb.de/alter_Datenbestand/Bilder_allgemein/Veranstaltungen/MV_Wiesbaden/Beschluss_TOP_15_Tischvorlage_Antrag_Praesidium_Dopingbekaempfung_MV_2013_Version_07.12.13__13.31_Uhr.pdf

28.

Gunn

Rees

(2008). Environmental review of integrity in professional tennis. Retrieved from http://www.sportingintelligence.com/wp-content/uploads/2011/01/Integrity-in-tennis.pdf

29.

Hakeem

F. B.

(2013). Sports-related crime: A game theory approach. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 247-260). New York, NY: Springer.

30.

Harris

(2013). Fixing, doping, whistle-blowing: Secrets that tennis prefers not to discuss. Retrieved from http://www.sportingintelligence.com/2013/06/24/fixing-doping-whistle-blowing-secrets-that-tennis-prefers-not-to-discuss-240601

31.

Hill

(2009). How gambling corruptors fix football matches. European Sport Management Quarterly, 9, 411-432.

32.

Hill

(2013). The insider’s guide to match fixing in football. Toronto, Ontario, Canada: McDermid.

33.

International Centre for Sport Security. (2014). Protecting the integrity of sport competition: The last bet for modern sport. An executive summary of the Sorbonne-ICSS integrity report sport integrity research programme, 2012-14. Retrieved from http://theicss.org/2019/03/12/icss-protecting-the-integrity-of-sport-competition-the-last-bet-for-modern-sport/

34.

Institut de Relations Internationales et Strategique. (2012). Sports betting and corruption: How to preserve the integrity of sport. Retrieved from http://www.spordiinfo.ee/est/g22s355

35.

Kalb

Herrmann

Emrich

(2015). Wettbewerbsverzerrungen im Sport aus Sicht von Athleten, Trainern und Funktionären [Competition distortions in sport from the perspective of athletes, coaches and officials]. In Emrich

Pierdzioch

Pitsch

(Eds.), Schriften des Europäischen Instituts für Sozioökonomie e.V. Falsches Spiel im Sport. Analysen zu Wettbewerbsverzerrungen [Wrong game in Sports. Analyzes of competitive distortions. Wrtitings of the European Institute for Socio-Economics e.V] (Vol. 10, pp. 55-108). Saarbrücken, Germany: Universaar.

36.

Kerkvliet

(1994a). Cheating by economics students: A comparison of survey results. Journal of Economic Education, 25, 121-133.

37.

Kerkvliet

(1994b). Estimating a logit model with randomized data: The case of cocaine use. Australian Journal of Statistics, 36, 9-20.

38.

Lensvelt-Mulders

G. K. L.

Hox

J. J.

van der Heijden

P. G. M.

Maas

C. J. M.

(2005). Meta-analysis of randomized response research: Thirty-five years of validation. Sociological Methods & Research, 33, 315-348.

39.

Locander

Sudman

Badburn

(1976). An investigation of interview method, threat and response distortion. Journal of the American Statistical Association, 71, 269-275.

40.

Maennig

(2008). Corruption in international sports and how it may be combated (IASE/NAASE Working Paper Series, No. 08-13). Retrieved from http://college.holycross.edu/RePEc/spe/Maennig_Corruption.pdf

41.

Misra

Anderson

Saunders

(2013). Safeguarding sports integrity against crime and corruption: An Australian perspective. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 135-155). New York, NY: Springer.

42.

Petropoulos

Maguire

(2013). Match fixing: Case studies from Greece and Ireland. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 89-99). New York, NY: Springer.

43.

Pitsch

Emrich

Frenger

(2013). Doping im Breiten- und Freizeitsport. Zur Überprüfung von Hypothesen mittels RRT-gewonnener Daten [Doping in mass and recreational sports. For checking hypotheses using RRT-derived data]. In Kempf

Nagel

Dietl

(Eds.), Im Schatten der Sportwirtschaft [In the shadow of the sports industry] (pp. 253-264). Schorndorf, Germany: Hofmann.

44.

Pitsch

Emrich

Klein

(2005). Zur Häufigkeit des Dopings im Leistungssport. Ergebnisse eines www-survey [To the frequency of Doping in competitive sports. Results of a www survey]. Leipziger Sportwissenschaftliche Beiträge, 46, 63-77.

45.

Pitsch

Emrich

Klein

(2007). Doping in elite sports in Germany: Results of a www-survey. European Journal of Sport and Society, 4(2), 89-102.

46.

Pitsch

Emrich

Pierdzioch

(2015). Match Fixing im Deutschen Fußball: Eine empirische Analyse mittels der Randomized-Response-Technik [Match Fixing in German football: An empirical analysis using the Randomized response]. In Emrich

Pierdzioch

Pitsch

47.

Pitsch

Maats

Emrich

(2009). Zur Häufigkeit des Dopings im deutschen Spitzensport—eine Replikationsstudie [The frequency of doping in German elite sports - a replication study]. In Emrich

Pitsch

(Eds.), Sport und Doping: Zur Analyse einer antagonistischen Symbiose [Sport and Doping: To analyze an antagonistic symbiosis] (pp. 19-36). Frankfurt am Main, Germany: Peter Lang.

48.

Preston

Szymanski

(2003). Cheating in contests. Oxford Review of Economic Policy, 19, 612-624.

49.

Qayyum

M. M.

(1998). Report of judicial commission. Lahore: Pakistan Cricket Board.

50.

Qureshi

Verma

(2013). It is just not cricket. In Emrich

Pierdzioch

Pitsch

(Eds.), Schriften des Europäischen Instituts für Sozioökonomie e.V. Falsches Spiel im Sport. Analysen zu Wettbewerbsverzerrungen (Vol. 10, pp. 69-88). New York, NY: Springer.

51.

Richard

M. M.

(2013). Impact of sports betting and corruption: Reflections from Uganda. In Emrich

Pierdzioch

Pitsch

(Eds.), Schriften des Europäischen Instituts für Sozioökonomie e.V. Falsches Spiel im Sport. Analysen zu Wettbewerbsverzerrungen (Vol. 10, pp. 31-67). New York, NY: Springer.

52.

Rullang

Gassmann

Emrich

Pierdzioch

(2016). Empirische Befunde zu Beeinflussungsversuchen bei Schiedsrichtern im Amateurfußball [Empirical findings on attempts at influencing referees in amateur football]. Sportwissenschaft, 46, 232-240. doi:10.1007/s12662-016-0409-0

53.

Scheers

N. J.

(1992). Methods, plainly speaking: A review of randomized response techniques. Measurement and Evaluation in Counseling and Development, 24, 27-41.

54.

Segal

(2013). The role of the academe in sports integrity: The objectives and shape of a sports integrity training course. In Haberfeld

M. R.

Sheehan

(Eds.), Match-fixing in international sports: Existing processes, law enforcement and prevention strategies (pp. 289-301). New York, NY: Springer.

55.

Shimizu

I. M.

Bonham

G. S.

(1978). Randomized response techniques in a national survey. Journal of the American Statistical Association, 73, 35-39.

56.

Sport-Informations-Dienst (SID). (2008). NBA-Schiedsrichter wegen Wettbetrugs verurteilt [NBA referee convicted of betting fraud]. Retrieved from http://www.focus.de/sport/mehrsport/basketball-nba-nba-schiedsrichter-wegen-wettbetrugs-verurteilt_aid_321070.html

57.

Simon

Striegel

Aust

Dietz

Ulrich

(2006). Doping in fitness sport: Estimated number of unreported cases and individual probability of doping. Addiction, 101, 1640-1644.

58.

Striegel

(2008). Doping im Fitness-Sport. Eine Analyse zwischen Dunkelfeld und sozialer Kontrolle [Doping in the fitness sport. An analysis between darkfield and social control]. Baden-Baden, Germany: Nomos.

59.

Tak

(2018). Too big to jail: Match-fixing, institutional failure and the shifting of responsibility. International Review for the Sociology of Sport, 53, 788-806.

60.

Tak

Sam

M. P.

Jackson

S. J.

(2016). The politics of countermeasures against match-fixing in sport: A political sociology approach to policy instruments. International Review for the Sociology of Sport, 53, 30-48.

61.

Tak

Sam

M. P.

Jackson

S. J.

(2018). The problems and causes of match-fixing: Are legal sports betting regimes to blame? Journal of Criminological Research, Policy and Practice, 4, 73-87.

62.

Union of European Football Associations. (2014). UEFA-Erklärung zur Integrität der Wettbewerbe [UEFA statement on the integrity of the competitions]. Retrieved from http://de.uefa.com/uefachampionsleague/news/newsid=2088790.html

63.

Warner

S. L.

(1965). Randomized-response: A survey technique for eliminating evasive answer bias. Journal of the American Statistical Association, 60, 63-69.

64.

Warner

S. L.

(1971). The linear randomized response model. Journal of the American Statistical Association, 66, 884-888. Retrieved from http://www.jstor.org/stable/2284247

65.

Wolter

(2012). Heikle Fragen in interviews [Sensitive questions in interviews]. Wiesbaden, Germany: VS Verlag für Sozialwissenschaften.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.48 MB

Corruption in Olympic Sports: Prevalence Estimations of Match Fixing Among German Squad Athletes

Abstract

Keywords

Introduction

State of the Research 6

Method

Data Collection

Questionnaire

RRT—Data Analysis Processing

Results

Descriptive Results

RRT Results

Discussion

Supplemental Material

Appendix_(1) – Supplemental material for Corruption in Olympic Sports: Prevalence Estimations of Match Fixing Among German Squad Athletes

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

Supplemental Material

Notes

Author Biographies

References

Supplementary Material

State of the Research⁶