Introduction

The question of whether it is a good thing for children to be coached by their own parents is met with a divided opinion in the world of sport and among the general public. On the one hand, one finds examples in literature, film, and television in which a certain parent-coach–child-athlete relationship (Weiss & Fretwell, 2005)Footnote 1 appears in a positive light (e.g., Schneider, 2021). On the other hand, one comes across findings that suggest that this form of “atypical coach–athlete relationships” (Jowett, 2008, p. 22) or “‘familial’ coach–athlete relationship[s]” (Jowett, Timson-Katchis, & Adams, 2007, p. 61) has a particular potential for conflict (e.g., Schmid, Bernstein, Shannon, Rishell, & Griffith, 2015).

To date, the benefits and risks of the corresponding dual role constellation (parent/coach, child/athlete) have received little attention in the education and research landscape (Graham, Dixon, & Hazen-Swann, 2016). This may be due, in part, to the fact that the explanatory object of the parent-coach–child-athlete relationship fits within a relatively young field of research: The systematic study of coach–athlete interaction did not take off until the beginning of the third millennium (Poczwardowski, Barott, & Jowett, 2006; Wylleman, 2000).

Nonetheless, research on parent-coach–child-athlete relationships should be pursued as it can be considered as part of the effort to achieve optimal talent development and individual development of athletes from the perspective of sport practice (Federal Institute of Sport Science [Bundesinstitut für Sportwissenschaft], 2023). The need to advance the study of parent-coach–child-athlete relationships may also be fueled by the fact that an increasing level of parental involvement is emerging in the field of youth sport in the present (Eliasson, 2019; Qunito Romani, 2020). A changing understanding of gender and parenting roles is hypothesized to trigger this Zeitgeist phenomenon (Trussell & Shaw, 2012).

As part of their research, some authors have addressed the elementary descriptive question of the number of parent-coaches found in the sport landscape. To date, this question has preferably been answered from the perspective of the parents (e.g., Gould & Martens, 1979). In general, Eliasson (2019, p. 1007) points out “that the exact number of parent-coaches and child-athletes of coaches in many countries remains unknown.” Given this background, it is innovative to determine the frequency of parent-coaches from the perspective of children, and, in addition, to ask the question about the relevance of parent-coaches to athletic performance. The question of how important coaching-by-parents is for the success of athletes has also not yet been the subject of quantitative research. In the light of these two existing research gaps, two studies were conducted. In contrast, the duration of such parent-coach–child-athlete collaborations has been measured on various occasions (McCann, 2005; Schmid et al., 2015; Weiss & Sisley, 1984). Furthermore, it was empirically found that in youth sports, it is primarily fathers who assume the coaching role for their children (Snyder & Purdy, 1982). It also became apparent that different types of parent-coaches can be identified (e.g., voluntary and full-time), although no one has yet taken a decidedly comparative look at the different forms. Thus, questions about the duration and ontogenetic localization of parent-coach–child-athlete collaborations, as well as about the gender distribution and different types of parent-coaches also became part of the research project.

Study 1

Study 1 was not only designed to answer the question of the frequency of occurrence; in parallel, two other parameters were at issue. The list of three research questions reads as follows:

  • What proportion of athletes were coached by their own parents during childhood and/or adolescence?

  • How long were child-athletes coached by their parents?

  • At what age were child-athletes coached by their parents?

Procedure

Study 1 is based on two main surveys, which resulted in a cross-sport substudy (“Cross substudy”) and a soccer-specific substudy (“Soccer substudy”). In both cases, data collection was conducted in the second half of 2019. In the Rhineland–Palatinate city of Speyer, 151 performance-oriented athletes were surveyed across all sports. The male and female athletes came from 14 sports; in terms of performance level, the athletes were each active at least at the state association level. In addition, 140 male and female soccer players in “Verbandsliga” to “Regionalliga” divisions were interviewed in the region around the city of Speyer (“Vorderpfalz” and “Südpfalz”). The soccer players belonged to eleven different teams. All male and female athletes were asked to answer the question in writing whether they had ever been coached by their own parents. In case of a positive answer, they were to indicate whether they had been coached by their father, mother or both parents. They were also asked to indicate when and for how long the respective collaboration had taken place. Finally, respondents were asked to indicate their year of birth.

As part of the various rounds of surveys, which usually took place at the beginning of a training session at the sports facility, the responses of male and female coaches were also recorded. The coaches were asked to answer the questions already mentioned both in terms of their own careers as active athletes (“Have you ever been coached by your own parents?”) and from a parent’s perspective (“Have you ever coached your own children?”). The data were collected and analyzed anonymously. For reasons of comprehensiveness, the results from the perspective of the athletes (“athletes’ survey”) will be cited decisively in this article.

Participants

A total of 291 athletes and 46 coaches were officially surveyed as part of the cross-sport and soccer-specific data collection efforts. However, only 330 people were involved, as seven respondents were both athletes and coaches (“playing coach”, “playing captain”). The seven “playing-coaches” were included in the respective data analysis regarding the “athletes’ survey” as well as the “coaches’ survey”. The analysis of the cross-sport survey officially included the responses of 109 male and 42 female athletes—the analysis of the soccer-specific survey included the responses of 87 male and 53 female players. The average age of athletes who participated in the cross-sport and soccer-specific surveys, respectively, was nearly 24 years on the cutoff date of December 31, 2019 (M = 23.844 years, SD = 6.808, N = 286). The average age of the coaches who participated in the two main surveys—plus one youth soccer coach who had been surveyed on an accompanying basis (“U17 Bundesliga”)—was just under 46 years on the same date (M = 45.638 years, SD = 10.741, N = 47; 42 men, 5 women).

Results

Frequency of the parent-coach–child-athlete alliancesFootnote 2

In the cross-sport survey, 29 of the 151 male and female athletes, or 19%, indicated that they had ever been coached by their own parents. In the soccer-specific survey, 47 of the 140 male or female players, or 34%, indicated that they had been coached by their own parents at some point. This difference is statistically significant (χ2 [1, N = 291] = 7.770, p = 0.005, V = 0.163). This significance is also present when the 16 soccer players interviewed in Speyer (“Cross substudy”) are added to the sample drawn at the regional level (“Soccer substudy”) (χ2 [1, N = 291] = 9.075, p = 0.003, V = 0.177). The descriptive characteristics of this comparison of all soccer players and those athletes playing an individual sport or a hand-oriented team sport can be found in Table 1.

Table 1 Distribution of interviewed athletes according to their sport category and their role as a parent-coached or nonparent-coached child (Study 1)
Full size table

Duration of parent-coach–child-athlete alliances

The duration of the alliance was practically always quantified by the interviewees in years and without monthly data. The measurement of the alliance duration within the framework of the data analysis was based in principle on the year-to-year overarching division of sports competitions into seasons. This means, for example, that the specification “2000 to 2005” was not calculated as six (calendar) years from a numerical point of view, but as five seasons.

First, we checked whether, from the athletes’ point of view, the duration of the parent-coach–child-athlete alliances differed between the genders (19 women, 56 men). Purely descriptively, it was found that when comparing female and male respondents, the medians of the two groups differed only slightly (MDwomen = 3 seasons, MDmen = 4 seasons). The same is true with respect to the comparison of the means (Mwomen = 4.474 seasons, Mmen = 4.661 seasons). A Mann–Whitney U test for large samples (Bös, Hänsel, & Schott, 2000) revealed that the distributions of the two gender groups did not differ significantly with respect to the duration of their collaboration with a father-coach or mother-coach (U = 477.5, n1 = 19, n2 = 56, z = 0.661, p = 0.509, two-sided, d = 0.153).Footnote 3

Next, we tested whether the duration of the parent-coach–child-athlete alliances differed between the sport categories (individual sports, hand-oriented team sports, soccer) from the perspective of the sport participants. A Kruskal–Wallis test showed that the assumption of no difference between the three groups must be rejected (H = 9.446, df = 2, p = 0.009, d = 0.679).Footnote 4 Dunn’s (1964) test was used for post hoc comparisons between the three groups. According to the results, the difference between the individual sport athletes (MD = 6 seasons) and the soccer players (MD = 2.5 seasons) can be considered statistically significant (z = 2.882, n1 = 11, n2 = 51, p = 0.004, two-sided, d = 0.787). This provided inferential statistical support for the assumption that alliances between parent-coaches and child-athletes last longer on average in the world of individual sports than in the world of soccer.

Ontogenetic localization of the parent-coach–child-athlete alliances

Using the written information about an athlete’s year of birth and when a parent-coach–child-athlete alliance began, it was possible to locate these special coach-athlete relationships in terms of life history or calendar (N = 75).

Figure 1 provides an overview of the temporal location of the parent-coach–child-athlete alliances within the life histories of the athletes. The phases of the collaboration are shown aggregated with the help of the educational stages that a person passes through during life. With this in mind, the majority of the identified parent-coach–child-athlete collaborations occurred during a time when the athletes were attending secondary school. Only 20% of the parent-coach–child-athlete alliances took place exclusively in a phase of life in which the children were still attending preschool or elementary school.Footnote 5

Fig. 1
figure 1

Ontogenetic location of the parent-coach–child-athlete alliances on the basis of the educational levels involved (athletes’ survey; N = 75)

Full size image

Gender of parent-coaches and child-athletes

At the level of all athletes interviewed, test statistics show a tendency to support the assumption that boys would be coached more often by their parents than girls (χ2 [1, N = 291] = 2.735, p = 0.098, V = 0.097). In all, there were 57 men and 19 women in the present study who had been coached by their parents at some point (Table 2). The group of parent-coaches concerned in the athletes´ survey (N = 76) comprises 67 fathers (88%) and 9 mothers.Footnote 6

Table 2 Distribution of the interviewed athletes according to their gender and their status as a parent-coached or nonparent-coached child (Study 1)
Full size table

Study 2

Study 2 was designed to focus on answering the question of the relevance of parent-coaches to athletic performance. It took place in the field of men’s soccer. The hypothesis was formulated directionally because in past qualitative studies, successful athletes were often coached by their parents (Prewitt-White, Fisher, Odenheimer, & Buchanan, 2016; Willms, 2009). The hypothesis was as follows:

The proportion of soccer players who have been coached by their parents turns out to be larger in higher-ranked leagues than in lower-ranked leagues.

Procedure

Study 2 was accompanied by another wave of surveys, in which lower-ranked players were visited in addition to soccer players in the “Oberliga” and “Verbandsliga”. The second wave of surveys was conducted in the second half of 2020. On the sidelines of training sessions, athletes were asked to answer in writing the question of whether they had ever been coached by their own parents. In case of a positive answer, they were to indicate whether they had been trained by their father, mother or both parents. In addition, the active players were asked in which calendar year or age they had joined a soccer club. The players’ year of birth was also recorded, as well as calendar information regarding the beginning and end of the parent-coach–child-athlete collaboration.

Participants

For the data analysis, data from a total of 331 male soccer players were finally available from both survey waves. At the time of the data collection, the athletes played for teams in the highest amateur leagues in states of Baden–Württemberg and Rhineland–Palatinate/Saarland (“Oberliga”), in high amateur leagues (“Verbandsliga”) in North Baden, as well as in the southwest region of Germany, and in lower divisions. The spectrum of lower-class leagues covered various regional divisions of the County League (“Bezirksliga”) as well as of the District League (“Kreisliga”) which is located below the County League and includes various playing levels (A, B, C and D). The lower-class teams belonged to the Westphalia Football and Athletics Association (“Fußball- und Leichtathletik-Verband Westfalen”) and the Southwest German Football Association (“Südwestdeutscher Fußballverband”).

Results

As part of the data analysis, a compressed cross tabulation was created (Table 3). A chi-square (χ2) test was then performed. Defining three categories in terms of playing strength, the test statistically calculated a relationship that was significant at the 1% level (χ2 [2, n = 331] = 11.950, p = 0.003, V = 0.190). This is a small effect size in the sense of Cohen (1988, as cited in Ellis, 2010, p. 41).

Table 3 Distribution of the interviewed male soccer players according to different levels of play and their status as a parent-coached or nonparent-coached child (Study 2)
Full size table

Discussion

The first study on the frequency of parent-coaches in sports reveals four main ideas in quantitative terms. First, the study suggests that a noteworthy number of male and female athletes who pursue their sport in a performance-oriented manner are coached by their own parents for a shorter or longer period of time. Across a variety of sports, the percentage—even in the case of changing perspectives—is consistently in the double digits. In the cross-sport survey in the city of Speyer, the percentage of parent-coached children was 19%. This percentage is only slightly reduced if the soccer players are left out of this sample; the percentage is then still 18% (Table 1). Even among the coaches interviewed (“coaches’ survey”), 5 out of 47 cases (11%) showed that the person in question had once been coached by a parent.Footnote 7

In addition, there are various sport-specific differences with regard to the frequency of parent-coaches. The proportion of parent-coaches was significantly higher in the group of soccer players (33%) than in individual sports and hand-oriented team sports (Table 1). Kemming (2008) and Reinders (2013) also point out in the context of their empirical studies that the parent-coach phenomenon can be observed especially in kids’ and youth soccer.

Wylleman and De Knop (1998) already observed that parents often start coaching their children after the kids have finished elementary school. The two authors found in the field of athletics and swimming that 20% of parent-coaches did not become coaches of their own children until the kids were in their teens and adolescents, respectively (cited in Wylleman & Lavallee, 2004, p. 514). In our study 1, we detected that 32% of the child-athletes had been coached by their parents starting when they were of secondary school age (Fig. 1). In addition, Wylleman and De Knop (1998) also seem to have encountered parent-coach–child-athlete alliances that did not begin until the transition from junior to senior age category or even later (13%) (cited in Wylleman, De Knop, Verdet, & Cecič-Erpič, 2007, p. 244; Wylleman & Lavallee, 2004, p. 514).Footnote 8

A fourth result which stands out is the statistically significant finding that alliances between parent-coaches and child-athletes—measured in seasons as previously described—are of longer duration in individual sports (MD = 6 seasons, M = 7.682, SD = 4.787) than among soccer players (MD = 2.5 seasons, M = 3.824, SD = 3.292). The duration of parent-coach–child-athlete alliances collected in study 1 is slightly lower than average values obtained in previous research (McCann, 2005; Weiss & Sisley, 1984). When all collaborations between parents and their own children in study 1 are considered (athletes’ survey and coaches’ survey), the mean alliance duration is 5.055 seasons (SD = 4.219, N = 95). If only the already completed parent-coach–child-athlete alliances are taken into account, then the average duration is 4.604 seasons (SD = 3.918, n = 84). In comparison, the average value in terms of alliance length in the retrospective study by McCann (2005)—following our calculation method—was 4.818 seasons (SD = 3.281, N = 11). The former parent-coaches (“dropout coaches”) in the study by Weiss and Sisley (1984, p. 336) had coached their own children for an average of 5.2 years (n = 97).Footnote 9

The evaluation of the developmental location of parent-coach–child-athlete alliances fully paves the way for a discussion. From the perspective of the athletes interviewed, a high number of collaborations were found to have begun after the child had completed—or with the completion of—elementary school (Fig. 1). Interestingly, among the alliances that had been formed at a later stage, there were often links that lasted only a short time: For 13 of the 24 identified child-athletes who had not been coached by their parents until secondary or tertiary school, the sporting alliance in question spanned a period of 2 seasons or less.

This could indicate a type of coach who is willing and able to accompany his or her son or daughter in a flexible manner. Ideally, this is not a parent who wants to make it easier or possible for the child and its peers to get started in a sport by taking on a coaching role (on a voluntary basis).Footnote 10 And it is probably just as little about a pronounced connoisseur who wants to accompany his offspring from childhood and over a long period in a sport.Footnote 11 It is possible that a type of father-coach or mother-coach has been identified in this way whose coaching career is to be considered independent of the athletic career of their own child. Both careers—that of the athlete and that of the coach—take their own direction over long stretches and overlap only at times (in “bridging situations”).Footnote 12

Finally, the finding that boys are more often coached by their parents than girls is quite inspiring, even if it only tends to be inferentially significant. This result can be made plausible by referring to the existence of gender role stereotypes. Eccles, Jacobs, and Harold (1990) as well as Würth and Saborowski (1999) have already shown in their respective studies that parents ascribe significantly more natural talent or more often an aptitude for competitive sports to their sons than to their daughters. The dominance of male respondents, both among coaches and athletes, can be attributed not only to the fact that sport is a male domain (e.g., Willms, 2009), but also to the fact that only 3 of the 14 sports surveyed can be named a “girls’ sport” (equestrian, swimming, volleyball). Since most of the coaches are male, the finding that boys are more often coached by their parents than girls, can also be seen as a variant of the leisure phenomenon that fathers spend more time with sons than with daughters in play, companionship, and gender-stereotypical activities (e.g., Yeung, Sandberg, Davis-Kean, & Hofferth, 2001). The observation that 88% of all parent-coaches were fathers can be understood as a form of “fathering through sport” (Kay, 2007, 2009) and as an expression of a new form of masculinity (“inclusive masculinity”; Anderson, 2009). With the help of children’s and youth sports, fathers become more involved in the family environment in a different way than is traditionally the case (Coakley, 2006; Graham & Dixon, 2014; Graham et al., 2016).

The second study on the performance relevance of parent-coaches in sports was based on a central assumption. The main hypothesis, which states that the proportion of male soccer players who have been trained by their parents is greater in higher-ranking leagues than in lower-ranking leagues, was confirmed empirically. This suggests that having been coached by one’s own parents has a direct or indirect beneficial influence on the development of the playing performance of young people.

The formulation just chosen regarding the ways in which the existence of a parent-coach figure influences the athletic development of youngsters alludes to the recognition that there is widespread unclarity about the subtleties of providing family support (Rees, Hardy, Güllich, Abernethy, Côté, Woodmann, Montgomery, Laing, & Warr, 2016). On average, child-athletes in soccer are coached by their father (the “normal case”Footnote 13) for 4–5 seasons.Footnote 14 This period, for which an individualized and intensive sporting cooperation between a parent and the child may be assumed, can in many cases have a beneficial effect on performance. This mechanism can be characterized with the title “coached-by-parents as a catalyst”. The interpretation approach “coached-by-parents as a catalyst” is based on the assumption that it is primarily or solely the father (or mother) in the function as a coach who has a certain added value. Perhaps the unique selling point of a father-trainer is to enable the son or daughter to train as individually as possible, so that, for example, unofficial training in their free time can serve as an optimal feeder for the official club training. In terms of training practice, this also means that initiatives in coach education are to be welcomed, particularly those aimed at qualifying moms and dads in the field of youth sports (Bavarian Football Association [Bayerischer Fußball-Verband & BFV], 2021).

Alternatively, or complementary, the function of a parent-coach can also be seen as a kind of mosaic stone of family assistance—as one mosaic stone among several. The willingness to take on a coaching position within one’s own son’s or daughter’s team can thus be interpreted as a signal that is, in principle, good. This signal might indicate that a player receives intensive support in many ways, not only in the public or organized sphere, but also at home. On the one hand, this sheds light on a coaching parent who is willing and able to take an in-depth look at the child’s sport. This means that a father’s voluntary involvement in an organized training of a certain soccer club may often be limited to just a few years, but the father’s expertise will be available to his son or daughter for long stretches of their childhood and youth. This can take the form of conversations around the kitchen table as well as one-on-one “extra shifts” on the soccer field, in the front yard, or in the park.Footnote 15 On the other hand, this brings the noncoaching parent and siblings into focus. Mothers, as well as brothers and sisters, also play a role in supporting a child-athlete in the private sphere (Blazo & Smith, 2018; Chafetz & Kotarba, 1999; Thompsons, 1999). Thus, the finding that players in higher leagues were more often coached by their own fathers than this was the case in lower divisions can also be understood as an indicator of qualitatively higher support for the player by the entire parental home. The mechanism referred to here can be described as “coached-by-parents as an indicator”.Footnote 16

However, both interpretations should be used with caution. On the one hand, the effect size of the relationship between the status of the child-athlete (as a parent-coached child or nonparent-coached child) and the level of performance achieved is small. On the other hand, the characteristic “parent-coached” is a category that indicates a social function and role but does not say anything about the concrete form of the associated leadership or educational behavior.

However, based on the present empirical results, the presence of a parent-coach figure in the biography of a soccer player can—as a rule—be considered as a performance-enhancing framework condition. Overall, the confirmation of our hypothesis joins many findings according to which the support that a young athlete receives from his or her family has a positive effect on the performance development (see most recently Coutinho, Mesquita, & Fonseca, 2018).

Limitations

Regarding the research methodology used in the present studies, some critical questions can be formulated, for instance on the veridicality of the memory performances as well as on the precision of the operationalizations. At many interview sites, the impression was that a large proportion—if not the majority—of interviewees first had to recapitulate autobiographical contexts in order to then be able to name the relevant years. The observation that a recollection cannot be retrieved episodically and therefore must first be generated (cf. Norman & Bobrow, 1979; Conway & Pleydell-Pearce, 2000) is also well known in sports science (e.g., Bette, Schimank, Wahlig, & Weber, 2002, p. 203). In fact, this should increase the likelihood of recall bias. In addition, detailed time measurements would also be desirable, for example based on monthly data. In general, however, the question arises to what extent participants in a survey framed by a fleeting encounter can still produce better recall performances.

Conclusion

Based on two studies on the frequency and performance relevance of parent-coaches in sports, two things can be stated: As a rough rule of thumb, it can be assumed that about one fifth of all performance-oriented senior athletes had been coached by their own parents for a shorter or longer period of time. Furthermore, according to an initial empirical examination, the fact of having been coached by one’s own parents can in principle be regarded as a conducive condition for the athletic development of adolescents. However, the presence of a parent-coach figure in an athlete’s biography is not per se a guarantee of tailored training and success in competition.