Dropout and social inequality: young people’s reasons for leaving organized sports

ABSTRACT

Few quantitative studies have looked at how reasons for quitting organized sports vary according to social backgrounds. The present paper addresses this gap by investigating how youths’ perceptions of six reasons for dropping out of organized youth sports vary according to three types of social inequality: socioeconomic status, ethnicity, and gender. We utilize data from the 2015 Young in Oslo survey, where organized youth sport dropouts in grades 8–13 rated the importance of six reasons for dropping out (N = 2355, response rate = 72%). Two findings stand out. First, gender differences were subtle and often related to ethnicity; they were more pronounced among majority youths than minority youths. Second, well-resourced majority youths were more likely to consider a lack of sports skills and friends who quit as important to dropout while minority youths with less resources more often highlighted discouragement from parents and sport expenses.

KEYWORDS:

• Social inequality
• organized sports
• youth sport
• withdrawal
• dropout

Introduction

Organized sports are valuable to many young people because they facilitate health, fitness, joy, and social benefits. Although most young Norwegians participate in organized sports for long or short periods, access to sports goods is distributed unevenly. Mitigating social inequalities in sports thereby requires a good understanding of how inequalities in participation develop.

The forces behind these social inequalities are intricate. As a start, socioeconomic inequalities are rising in many Western societies (Aaberge 2016; Piketty and Goldhammer 2014). Next, the last decades have seen high levels of immigration which implies an added layer of challenges to social inequalities (Anthias 2001; Nobis and El-Kayed 2019). Third, gender inequalities are still contested and debated. (Anthias 2001; Strandbu, Bakken, and Sletten 2019; Toffoletti and Palmer 2019). In organized sport, as in society at large, these three forces intersect to influence sport participation (Breuer and Wicker 2008; Nobis and El-Kayed 2019), and the need for intersectional studies of social inequality across social categories are in great demand (Choo and Ferree 2010; McCall 2005).

Simultaneously, to grasp how the uneven distribution of sports goods emerges, we must study sports participation as a process because the outside forces – SES, gender, and majority/minority status – might affect sports participation differently when it comes to sport recruitment, continuation, and dropout. As an example, gender differences in sport participation could come from girls not being recruited, more girls dropping out, or both.

In addition to social inequalities affecting sports participation in diverse phases of sports careers, understanding sports participation shifts should include the participants’ rationales. Athletes might have one reason for starting sports, another for retaining participation, and still more reasons for dropping out.

We know that social inequality in sports participation exists; girls and youths with low SES and minority backgrounds are more likely to drop out (Bakken 2019a). We have a general idea of the common explanations for dropout (Crane and Temple 2015), and participation levels according to social backgrounds (Nobis and El-Kayed 2019; Strandbu, Bakken, and Sletten 2019). Nonetheless, how patterns of social inequality connect to sports dropouts is essentially a black box: How do youths from different social backgrounds differ in their perception of their dropout?

Understanding social inequality in dropout is not the same as understanding participation barriers because those who drop out have already negotiated these. Still, social inequality is a challenge in sports dropouts (Bakken 2019a). Are minority girls overrepresented in sport dropouts because they prioritize schoolwork over sports? Do girls drop out more often because they lack skills, or is this a stereotype of the past? Are financial costs the only reason youths with lower SES drop out more, and is it unrelated to ethnicity? Many of these questions have not been adequately answered.

We will focus on six reasons for ending sports and discuss how SES, ethnicity, and gender might matter for youths’ reasons for dropping out. We use one theory to understand each social force: Bourdieu’s theory of social class, status and power (1978, 1984) as a backdrop to understand socioeconomic differences, hegemonic masculinities (Connell and Messerschmidt 2005) to understand gender differences, and Strandbu, Bakken, and Sletten’s (2019) studies of immigrant sports to understand ethnic differences. The data is from a survey of Norwegian youths (aged 12–19, N = 2355). For the results, we present simple frequencies of reasons for organized sports dropout before regressing reasons for dropout on social background. We end by discussing how the results matter for current research on social inequality in sports and the practical sides of handling social inequality in young athletes’ sports participation.

Contextual background and reasons for dropout

There is a challenging ambiguity in Norwegian youth sports: 93% of youths have participated in organized sports at some point in their childhood or youth (Bakken 2019a), and most youths drop out. Dropout rates are somewhat greater for girls than boys and substantially larger for minority girls and youths with lower SES (Bakken 2019a). While the minority-majority gap appears because many minority girls never start with organized sports and some drop out more frequently, youths with lower SES disappear more evenly during recruitment and dropout (Bakken 2019a, 2019b).

The ‘sport for all’ policy is present in government white papers (i.e. Meld. St. nr. 26 2011–2012) and strategic documents within the Norwegian Olympic and Paralympic Committee and Confederation of Sports (NIF). Two goals of inclusion underlie these documents: (a) recruiting and retaining as many as possible and (b) closing social differences in participation rates. Children’s Rights in Sports contribute to these goals by restricting competitive arenas until the age of 13: (1) No rankings or results until the age of 11, (2) rankings and results should only be used if it serves a purpose when children are 11 and 12, (3) children between 6 and 8 can only participate in local competitions, preferably in the club, and (4) children can compete regionally when they are between 9 and 11 (NIF 2015). NIF receives a substantial financial government support, and even more money is spent by municipalities to build and maintain sports facilities (Meld. St. nr. 26 2011–2012). NIF is the largest voluntary organization in Norway, most of the voluntary work is done by parents, and sports are mostly organized independent of the school-system. Therefore, there is a strong interdependence between NIF and the government: NIF has leverage towards the government in their sheer number of members and their role as a social arena promoting young people’s lifelong health, while they also depend on substantial financial support from the government (Bergsgard and Norberg 2010).

A few trends challenge the social equality ideal in sports. Sports clubs have become more competitive and professionalized (Seippel 2010, 2019). These developments can increase demands for sports skills, participation fees, and time, especially after the age of 13, when the Children’s Rights in Sports no longer apply (NIF 2015). Parallel to developments inside sports, today’s youth generation is also more serious or concerned about schoolwork, maintaining a good appearance, and being physically active (Eriksen et al. 2017; Hegna, Ødegård, and Strandbu 2013; Seippel 2016). Youths’ reasons for dropout are important because they illustrate experiences in sports and problematic patterns of dropout, may highlight barriers that lead to structural inequality in sport dropout and highlight possible improvements in the organization of sports.

Based on previous research, six reasons were included in the survey: lacking skills, friends quit, lacking suitable alternatives in the club, too expensive, parents discourage participation, and prioritizing schoolwork. First, the need to feel competent and master new skills become more challenging as children grow. Youths often report that they quit sports because it is too serious in terms of time demands and athletic skills (Persson et al. 2020), indicating that it is common to quit because they (1) lack skills. Second, many youths participate in sports to be with friends, even in their late teens (Jakobsson, Lundvall, and Redelius 2014). Increasing dropout rates with age can create a snowball effect (i.e. Gatouillat, Griffet, and Travert 2020), where youths who initially drop out for other reasons take friends with them. The snowball effect could thereby explain why many youths drop out because their (2) friends quit (Persson et al. 2020). Third, at a level more relevant for club structure, youths drop out because they (3) lack suitable alternatives in the club: teams dissolve, training hours are too late or inflexible, or training is too skill-demanding (Persson et al. 2020).

Fourth, professionalization can increase the cost of participation. When sports become more competitive from the age of 12 (Espedalen, Grønkjær, and Strandbu 2021; Oslo Economics 2020), youths may drop out because it is (4) too expensive. Fifth, sports also involve and demand parents as active agents in their children’s sports engagement (Stefansen, Smette, and Strandbu 2018; Strandbu, Bakken, and Stefansen 2020; Strandbu et al. 2019). Because parents vary in their ability and willingness to support their children’s sport participation (Bakke, Solheim, and Hovden 2016), some youths may therefore drop out because (5) parents discourage participation. Last, a generation of youths who increasingly emphasizes school performance (Bakken 2016; Hegna, Ødegård, and Strandbu 2013; Vestel and Øia 2014), parallel to increasing sport demands, could mean youths drop out because they (6) prioritize schoolwork. In the next section, we demonstrate how these six reasons for dropout connect to SES, ethnic background, and gender.

Theoretical perspectives and hypotheses

We take a categorical intersectional approach to social inequality: to understand structural inequality, we must compare both disadvantaged and resourceful social groups (i.e. McCall 2005). Therefore, categorical intersectionality concerns both the ‘main effects’ of SES, gender, and ethnicity on dropout-reasons, and the intersections between these categories. In the following, we apply theories supplied with relevant empirical contributions to outline a set of hypotheses for how SES, gender, and ethnicity impact reasons for dropout. We limit our hypotheses to those most relevant for social inequality in youths’ reasons for dropout.

Socioeconomic status

Economic and cultural capital are keys to understanding socioeconomic gaps in sports participation (Andersen and Bakken 2019; Bourdieu 1978, 1984; Laberge and Kay 2002; Wilson 2002). Economic capital comprises financial means and material possessions and is a critical cause of sports participation’s socioeconomic gap (Andersen and Bakken 2019; Bourdieu 1984). Cultural capital involves knowledge (e.g. education), skills (e.g. athletics), and possessions (e.g. books) of status in the broader cultural context (Bourdieu 1984). People with similar economic, cultural, and social resources tend to have shared tastes in activities (e.g. preference for organized sports and specific types of sports), which predisposes them to think and act in similar ways (i.e. habitus; Bourdieu 1984). Although one important point of Bourdieu is that these preferences create class-based distinctions between sport-fields, recent research also suggest the family’s cultural and economic capital represent a systematic underrepresentation of families with lower SES in organized sports overall (Andersen and Bakken 2019; Breuer and Wicker 2008). Even though Bourdieu to a large extent is concerned with how action results from the logic of the particular field and previously acquired dispositions, he also emphasizes the situated reflexivity of actors (Bourdieu 1984). To grasp these actions, we therefore contend that youths’ reasons for dropping out from organized sports partially reflect a classed habitus forming their experiences of organized sports (Bourdieu 1978; Laberge and Kay 2002) as well as a situated reflexivity regarding the value of sports.

While some research show that organized activities tend to be more valued in upper- and middle-class families and working-class families more often appreciate unorganized play (Lareau 2002), other studies show a more complex picture where cultural and economic capital interact and overlap (Andersen and Bakken 2019; Wilson 2002). Some studies in Northern Europe show that youths with different socioeconomic resources value organized sports in similar ways (Jakobsson et al. 2012), but that youths’ with less socioeconomic resources engage in fewer organized activities – often because of parents’ difficulty of combining inflexible workhours with supporting their children in multiple activities (Sjödin and Roman 2018). This points to a socioeconomic gradient in organized sports based on access to economic and cultural resources. As children grow, sports demand more time and money (Espedalen, Grønkjær, and Strandbu 2021; Oslo Economics 2020). These developments most likely put more pressure on parents with less economic and cultural resources to support their children’s sport participation. Therefore, we expect that the reasons most clearly affected by SES are that sports are (H1) too expensive and somewhat more exploratory that parents discourage participation.

Gender

Hegemonic masculinity theory implies that patriarchal notions of masculinity and femininity maintain positions of power within a social field such as organized sports; boys tend to comply by incorporating the dominant masculinity, while girls more often incorporate traditional feminine values (Connell and Messerschmidt 2005). In sports, the dominant masculinity is often associated with confidence, competition, and assertiveness, while femininity is more often associated with relational attributes, such as empathy and friendship (Eagly, Wood, and Diekman 2000; Guillet et al. 2006). Girls risk being caught in a limbo where adopting dominant values of assertiveness and competition conflict with femininity norms (Guillet et al. 2006; Slater and Tiggemann 2010). When sports become more competitive, gendered sports withdrawal may occur as sports become less inclusive and more serious. Simultaneously, girls leave sports more often than boys because of inflexible training hours and dissolving teams (Persson et al. 2020). Girls also tend to spend more time on schoolwork (Zuzanek 2018), putting more pressure on their spare-time. Though there are small gender differences in Norwegian sports participation, theory and research suggest that the experiences of belonging and availability in sports are gendered (Dalen and Seippel 2019; Hovden and Tjønndal 2019; Persson et al. 2020). Therefore, we expect that more often than boys, girls consider their dropping out important because they (H2) lack suitable alternatives in the club, lack skills, prioritize schoolwork, and friends quit.

Ethnic background

Strandbu, Bakken, and Sletten (2019) suggested four mechanisms to explain the underrepresentation of minority youth in sports: (1) a culture from the parents’ country of origin that deemphasizes unfamiliar organized sports compared to schoolwork and activities in one’s ethnic and cultural community; (2) religion – primarily Islam – that puts restrictions on girls’ participation; (3) discrimination and racism; and (4) an overrepresentation of young people from families with lower SES.

The minority-majority gap in sports is primarily due to the underrepresentation of minority girls where parents’ culture and religion seem to impact participation (Kay 2006; Prieur 2002; Strandbu, Bakken, and Sletten 2019; Walseth 2006). Furthermore, Norwegian youths with immigrant backgrounds are more likely to have parents from more patriarchal societies (Prieur 2002), which could align more closely to norms of masculinity in the context of organized sports. In this sense, many minority boys may feel especially at home with masculine values in sports. In contrast, some minority girls may feel particularly alienated through cultural ideals that highlight their role as carriers of family culture and tradition. Based on the logic above, we first hypothesize that minority youths, more often than majority youths, consider their dropout important because they (H3) lack suitable alternatives in the club and their parents discourage participation.

Our last hypothesis concerns the intersections between SES, gender and ethnicity. Because we know that ethnic minority girls are particularly marginalized and prone to dropping out from sports, ethnic minority boys are particularly unlikely to dropout (when controlling for SES) (Bakken 2019a), and somewhat more patriarchichal and gendered norms are more salient in many counries of origin in the ethnic minority population (Prieur 2002; Strandbu, Bakken, and Sletten 2019), we suggest that (H4) the gender differences addressed in (H2) will be particularly pronounced among youths with an ethnic minority background.

Method

We analysed data from the 2015 Young in Oslo survey (Andersen and Bakken 2015). The survey covered diverse aspects of teenagers’ lives, including their leisure activities, sports participation, family background, relationship to parents, and mental health. It was administered during class in 88 lower- and upper-secondary schools, including every Oslo public school.

The total sample consisted of 23,381 young people from grade 8 (normally 12–13 years old) to grade 13 (normally 18–19 years old), representing approximately 70% of Oslo’s total population between 12 and 19 years of age. A random third of the respondents received questions about organized sports participation (response rate = 83%). In our analyses, we included those who reported that they currently did not participate in any organized sports but had done so previously. Because of concerns with anonymizing younger participants and questionnaire fatigue in a large population survey, we do not have data on the type of organized sports the youths participated in before they quit. However, all organized sports are part of the same sport organization (NIF) which means that they all adhere to some similar values such as parent volunteering and economic support systems from municipalities and the government. On a more general basis, the largest organized youth sports in Norway are soccer, handball, and cross-country skiing. The final sample consisted of 2355 youths (grade 8 = 11%, grade 9 = 13%, grade 10 = 16%, grade 11 = 22%, grade 12 = 19%, and grade 13 = 19%). Parents and students were informed in advance about the study and that participation was voluntary. The Norwegian Social Science Data Services approved all ethical aspects of the study.

Dependent variables

Respondents were asked, ‘What was the reason you stopped participating in organized sports?’ followed by six reasons for quitting organized sports: Skill requirements were too strict, Friends quit, The club did no longer provide suitable options, It became too expensive, My parents did not like that I participated, and I had to spend more time on schoolwork. To ease the interpretation of results, we recoded an initial three-point scale into a dichotomous variable, where 0 = not important at all and 1 = somewhat or very important.

Independent variables

Gender and ethnicity were dichotomous variables. Youths with both parents born abroad were considered minorities. More than half of the minority sample in grades 11–13 had parents born in Pakistan, Somalia, and Sri Lanka in the initial sample (younger teenagers were not asked because of the stricter demands for anonymity below grade 11). Most of the remaining minority youths had parents from other countries in Africa, Asia, and Eastern Europe. The final sample in the analyses consisted of 72.5% majority youths and 27.5% minority youths.

We use a composite measure of SES consisting of the Family Affluence Scale (FAS II; Currie et al. 2008) and cultural capital. FAS II was measured by an average score from 0 to 3 based on three questions: ‘Does your family have a car?’, ‘Do you have your own bedroom?’, and ‘How many times have you travelled somewhere on holiday with your family over the past year?’ FAS II shows high inter-rater agreement between children and parents (Andersen et al. 2008), satisfactory validity, and correlates as expected with a wide range of health behaviours, including physical activity (Currie et al. 2008). A potential objection to the FAS II-measure could be that owning a car may be more necessary for lower SES-families in the suburbs. Oslo have an underground subway system that connect the poorer suburbs to the city centre, meaning many poorer families can manoeuvre the city landscape without a car. Recent research has indicated that the ability to distinguish between the broad middle- and upper-classes with rising affluence levels is more challenging, though still applicable within a national context when not comparing results across time (Schnohr et al. 2013). Cultural capital was measured by asking the teenagers, ‘How many of your parents have higher education?’ (0 = none, 1 = one parent, and 2 = both parents) and ‘How many books would you say there are in your home (1 metre of bookshelf approximately equals 50 books)?’ Response options ranged from 0 to 5, where 0 = no books and 5 = more than 1000 books. Books as measured by home library size in adolescence is an important predictor over and above parent’s education of a range of cultural capital competencies such as literacy, numeracy and problem-solving (Sikora, Evans, and Kelley 2019). A national Norwegian youth report also validated the cultural capital measures satisfactorily against school grades (Bakken, Frøyland, and Sletten 2016). We created a composite measure of SES combining FAS II and the cultural capital measures (i.e. Bakken, Frøyland, and Sletten 2016), where SES was measured as quintiles of SES calculated from the initial sample of 23,381 youths.

We analysed the data using multiple logistic regression and calculated predicted probabilities dependent on the combination of values in independent variables using R v4.0.3 (R Core Team 2020). To guide the reader, we begin with interpreting the regression analyses and the probability plots. Following, we present differences in the social backgrounds for each reason to drop out of sports before moving to the discussion.

Results

Ranked by the proportion of our respondents’ relevance rankings (see also Table 1), the reasons for dropout are prioritize schoolwork (61%), lack skills (51%), friends quit (50%), lack suitable alternatives in the club (36%), too expensive (17%), and parents discourage participation (8%).

Table 1. Overview of variables in the study – frequencies and percentage of total sample.

	Frequencies	Percentage
Social background (IVs)
Socioeconomic status (low – high)
SES 1	261	11.1
SES 2	497	21.1
SES 3	562	23.9
SES 4	523	22.2
SES 5	512	21.7
Gender
Boys (=0)	968	41.1
Girls (=1)	1387	58.9
Ethnicity
Majority youths (=0)	1698	72.1
Minority youths (=1)	657	27.9
Reasons to quit sports (DVs = 1)
Prioritize schoolwork	1415	60.1
Lack skills	1182	50.2
Friends quit	1168	49.6
Lack suitable alternatives in the club	836	35.5
Too expensive	398	16.9
Parents discourage participation	190	8.1
Total	2355	100

Note: Frequencies and percentage of dependent variables show those who considered each reason to drop out important.

Table 2 contains the results from six logistic regression models demonstrating how each reason to quit depends on social background. Figure 1(a–f) supplement Table 2 and show predicted probabilities for respondents reporting the respective reasons for dropping out as important (i.e. detailed effect size of dependencies in our model). Each figure reports the predicted probabilities (y-axis) for four groups represented by one line for each group: minority girls (whole black lines), minority boys (dotted black lines), majority girls (whole grey lines), and majority boys (dotted grey lines), and how these predicted probabilities vary by SES (x-axis).

Figure 1. (a–f) Predicted probabilities for each reason to quit according to social background.

Table 2. Logistic regression analyses of reasons to quit by social background.

	Prioritize schoolwork	Lack skills	Friends quit	Lack suitable alternatives in the club	Too expensive	Parents discourage participation
SES (1 = low, 5 = high)	.01	.07	.12**	.01	−.34***	−.10
	(.04)	(.04)	(.04)	(.04)	(.06)	(.09)
Ethnicity (minority = 1)	.93***	−.63*	−.31	.70*	.06	1.02*
	(.17)	(.27)	(.27)	(.27)	(.32)	(.44)
Gender (girl = 1)	.84***	.31**	.14	.19	.22	−.44*
	(.10)	(.10)	(.10)	(.11)	(.14)	(.22)
SES*Ethnicity	−.09	.07	−.21*	−.21*	.10	−.03
	(.09)	(.09)	(.09)	(.09)	(.11)	(.14)
Ethnicity*Gender	−.60**	−.35	−.29	−.20	−.25	.11
	(.20)	(.19)	(.19)	(.20)	(.24)	(.31)
Constant	−.17	−.18	−.36*	−.74***	−.72***	−2.25***
	(.17)	(.08)	(.17)	(.17)	(.22)	(.35)
Nagelkerke’s R^2	.05	.05	.02	.01	.05	.06

Note: Beta-coefficients outside parentheses and standard deviations inside parentheses. *p ≤ .05, **p ≤ .01, ***p ≤ .001.

For example, all lines in Figure 1(e) trend downward, illustrating that the higher their SES, every group has a lower probability for reporting it is too expensive as an important reason for dropping out. In general, to study the effect of SES, look at the direction of the lines – upward, flat, or downward – in the figure and check for statistical significance in Table 2. To interpret the effect of gender, compare whole lines to dotted lines. To distinguish minority from majority youths, compare black to grey lines. For example, as both grey lines are above the black lines in Figure 1(b), majority youths are more likely than minority youths to report they quit sports because they lack skills. Table 2 (column 2, row 2) shows that this effect is significant (β = −.63, p ≤ .05), so we can be confident that the difference observed in Figure 1(b) is not by coincidence. Figure 1(b) also illustrates an apparent gender difference in the majority sample; the whole grey lines (majority girls) are higher in the plot than the dotted grey lines (majority boys), demonstrating that majority girls were more likely than majority boys to report they quit sports because they lack skills. This gender difference indicates an interaction effect between ethnicity and gender (i.e. there is a greater gender difference among the majority youths compared to minority youths). However, the effect is not significant (β = −.35, ns) as seen in Table 2 (column 2, row 5). In other words, the probability plot shows that there is an observable difference, but we cannot be confident that the difference is not due to chance. We summarize how social inequality relates to each reason for dropout below, concluding the results with a summary of how the data fit with the hypotheses.

Inspecting prioritize schoolwork as a reason to quit sports (Table 2, column 1; Figure 1(a)), there is only one finding that stands out: majority boys (grey dotted line) were less inclined to view school work as a reason for ending sports than all other groups in our model. Additionally, minority youths (black lines) were less likely to report they lack skills as a reason for ending sports than majority youth (Table 2, column 2; Figure 1(b)). Among youths with a majority background, girls (grey whole line) had a higher propensity than boys (grey dotted line) to report they lack skills as a relevant reason to quit. The black lines show that the gender differences for minority youths were minimal (Table 2; column 2). In short, the effect of gender depended on ethnicity, and majority boys were less likely to consider schoolwork important for their decision to drop out than the other youths in the sample.

For reporting friends quit as a reason to drop out from sports (Table 2, column 3; Figure 1(c)), SES was the dominant variable, although it interacted visibly with minority status. In Figure 1(c), SES mattered for all groups in the model (no lines are flat), but the effects ran in opposite directions; for majority youths (grey lines), the SES effect was positive: the higher the SES, the higher the probability of reporting their friends quit as a reason to drop out. For minority youths, the effect ran in the opposite direction; the higher the SES, the lower the probability of reporting that their friends quit as a reason for quitting sports. Gender did not have a significant effect on dropping out because friends quit. In short, the results show that those who theoretically had the most resources to sustain sports participation (majority boys with high SES) were also most likely to consider their friends quit as an important reason to drop out.

For youths who reported they quit because they lack suitable alternatives in the club (Table 2, column 4; Figure 1(d)), there were differences between minority and majority youths that varied with SES; among youths with lower SES, minorities mentioned more than majorities that they lack suitable alternatives in the club as an important reason for dropping out. Among youths with higher SES, majorities tended to report they lack suitable alternatives in the club more often than minorities. The effect of SES was only valid among minorities (declining black lines). The results showed that minority youths were marginally more likely to quit because they lack suitable alternatives in the club, but it was only valid for those with low SES.

The simplest model to interpret concerns sporting expenses (Table 2, column 5; Figure 1(e)). Socioeconomic background (SES) had a clear, negative, and significant effect. As illustrated in Figure 1(e), all lines declined, and it was difficult to distinguish between gender and minority status. This finding indicates strong support for part of H1: youths with lower SES are more likely to consider too expensive as an important reason to quit irrespective of their gender or ethnicity.

Parents discourage participation was without interaction effects as a reason for ending sports and, hence, not too difficult to interpret directly from Table 2 (column 6). Minority status had a positive effect, illustrated by Figure 1(f), where minority boys (dotted black line) and girls (whole black line) were above the grey lines representing the majority. The effect of gender was negative for both majorities and minorities; girls (whole lines) reported that their parents discourage participation less often than boys (dotted lines). The findings show that, as hypothesized in H3, minority youths considered parents discourage participation as a more important reason for dropout than majority youths. A more surprising finding is that there was no gender difference among ethnic minorities, as discussed below.

Combined, the results partially support our hypotheses for SES and gender (H1 and H2) and support H3 for a minority-majority gap. H4 proposed that gender differences were more pronounced among minority youths, which our results contradicted (see Table 3). We return to H4 in the discussion, as it is one of our study’s surprising and critical results.

Table 3. Overview of support for each hypothesis.

Hypothesis	Lack skills	Friends quit	Lack suitable alternatives in the club	Too expensive	Parents discourage participation	Prioritize schoolwork	Support for hypothesis
H1: SES	X	X	X	Yes	No	X	Partial
H2: Gender	Yes	No	No	X	X	Yes	Partial
H3: Ethnicity	X	X	X	X	Yes	Yes	Yes
H4: Gender × ethnicity	No	No	No	X	X	No	No

Note: X = not relevant for hypothesis.

Summary and discussion

This article examined reasons Norwegian youths provide for dropping out of organized sports and how these reasons depend on social backgrounds (SES, gender, ethnicity). Overall, the most common reasons for dropout were that young people prioritize schoolwork, lack skills, and their friends quit. The least common reasons were that sports are too expensive, and their parents discourage participation. The most common reasons for dropout herein align with quantitative and qualitative research showing that increased time demands and the seriousness of schoolwork and sports contradict sport participation for many youths in Western and Northern Europe (Carlman, Wagnsson, and Patriksson 2013; Persson et al. 2020; Gatouillat, Griffet, and Travert 2020).

According to SES and ethnicity, there are two distinct differences in dropout reasons and a more complex picture for gender. First, young athletes from families with lower SES drop out more often than youths with higher SES because sports are too expensive. Second, ethnic minorities are twice as likely as their ethnic majority counterparts to quit because their parents discourage participation. For gender (see Elling and Knoppers 2005), the results showed the importance of considering ethnic background effects when discussing gendered mechanisms of inclusion and exclusion in sports. An unexpected key finding is that compared to ethnic minority youths with lower SES, ethnic majority youths with higher SES more often report dropping out because they lack skills and their friends quit.

Overall, our findings indicated that youths from resourceful ethnic majority families more often quit sports because they lack skills and friends quit compared to less resourceful ethnic minorities. Parents who discourage participation and that sport participation is too expensive are more common reasons for dropout among ethnic minority families with fewer resources than their resourceful majority counterparts.

Empirical contributions

It is well known that financial constraints matter for socioeconomic inequality in sports (Andersen and Bakken 2019). We found that youths with lower SES are substantially more likely to experience sport as too expensive as a reason for sport dropout. Our study shows that many youths from less resourceful families may start with sports but struggle to continue when sports become more expensive, as illustrated by two recent reports showing that even for cheap sports like youth soccer, costs for tournaments and travelling are challenging for many families in the lower socioeconomic strata (Espedalen, Grønkjær, and Strandbu 2021; Oslo Economics 2020). Future research should look at how youths and parents experience sports expenses as barriers to entering sports relative to dropping out so that sports organizations and clubs know where to combat the socioeconomic gradient in sports participation (i.e. financial barriers to take up sports or to continue with sports).

For gender and ethnicity, previous research has indicated that experiences of sport belonging and availability are more inclusive for boys than girls (Dalen and Seippel 2019; de Haan and Knoppers 2020; Persson et al. 2020) and that minority girls face more barriers to sports participation (Kay 2006; Strandbu, Bakken, and Sletten 2019; Walseth 2006). Our results reflect some of these experiences in the dropout process but only in the ethnic majority population. These findings are unexpected and contrast with the assumption that gender differences in dropout experiences are greater in the minority sample. Furthermore, they demonstrate why it is essential to move beyond the research on participation and non-participation and to specify inclusion and exclusion experiences in sports recruitment, participation, and dropout.

We propose – in part – a selection effect to explain the findings; first, minority girls who negotiate barriers in the recruitment phase may be particularly suited to organized sports’ physical and competitive nature. Second, family is a primary recruitment agent in Norwegian organized sports (Skille 2005). As such, religious restrictions and family culture deemphasizing organized sports for minority girls (i.e. Strandbu, Bakken, and Sletten 2019) could therefore be a barrier to entering sports and matter less in dropout. Elling and Knoppers (2005) have highlighted that although minority girls participate less, they tend to value traditional masculine contact sports such as football, basketball, and martial arts more than many majority girls. Our results favour that minority girls are more open to the stereotypically masculine logic of sports (i.e. Elling and Knoppers 2005), mainly because we demonstrated that stereotypically gendered reasons to quit, such as lack skills and friends quit, are more common among majority boys than minority girls, and the gender gap is consistently present among ethnic majority youths, not minority youths. An additional explanation for this finding could be a ‘minority drive’ in a sporting context: immigrant youth in Norway have on average higher educational aspirations, put more effort into homework, and take school more seriously than youths from the ethnic majority (Lauglo 1999; Skauge and Hjelseth 2021). Youths with lower SES in our study are similar to the girls from the ethnic minority in that they less often drop out because they lack skills or their friends quit. Future research should explore these assumptions further, especially regarding ethnic minority girls in sports. Moreover, we add that it is essential for researchers to specify if inclusion and exclusion experiences relate to sport recruitment, participation, or dropout. We also stress that future research should compare the experiences of minorities with those of majority youths to identify social inequalities in sporting experiences and potential areas of sports that are more suited to minorities.

Implications for sport policy and integration in sports clubs

Two approaches to combat social inequality in sports underlie white papers and strategic plans in governments and sports organizations: to increase the overall level of participation and to decrease social inequalities. The two do not necessarily complement each other and may require different inclusion measures.

It is helpful to study why most youths quit sports to increase overall participation. The most important reasons for dropout in our study are schoolwork, lacking skills, and friends who quit, indicating similar dropout patterns with other studies: youth sports become too serious, time demanding, and lack prominent social aspects (Gatouillat, Griffet, and Travert 2020; Persson et al. 2020). Perhaps the single most effective measure for governments, sports organizations, and clubs is to create more diverse participation alternatives where some are less time-consuming and cost-inducive. One example is the umbrella organization TVERGA in Norway, which, while still in its infancy, receives funding from the government as an alternative to NIF to counsel local initiatives on self-organization of physical activities that favour physical exploration, flexibility, and creativity over competition (e.g. parkour, skateboarding, street dance) as listed on their website as of 15 October 2021. It is, however, necessary to be aware of the ambiguity of such policy tools: lowering the threshold for some, could lower the commitment from others (not seeing sports as too serious) and lead to further differentiation.

Two findings must inform policy and social inclusion work at the grass-roots level of sports clubs if the goal is to combat social inequality. Our results show that – although overall the least important reason for dropouts – sporting expenses are the only dropout reason where youths with lower SES are overrepresented. Simultaneously, youths with lower SES are as likely as youths with higher SES to say they quit sports because their parents discouraged them. This finding shows an important point: Apart from potential class-based distinctions between sports (e.g. Stempel 2005), youths with diverse socioeconomic backgrounds experience their sport dropout in similar ways with one exception: sport becomes too expensive. At a sport policy-level, this finding highlight how critical it is to take measures to reduce the financial costs of sports if the goal is sports for all. At the same time, more information about the cost drivers of sports is necessary to draft policies that effectively address the expenses sports currently demand. Sports clubs aiming to reduce socioeconomic inequality should pay close attention to sports participation expenses, how parents experience costs, and signs indicating children lack equipment. The stigma attached to poverty (e.g. Bakke, Solheim, and Hovden 2016) may be challenging for sports clubs to handle. Providing parents with systematic information acknowledging sports participation as expensive and detailing financial support systems could help sports clubs battle socioeconomic inequality. Our study also shows that many youths are aware of the role finances play in sports participation opportunities. Hence, we encourage sports clubs to approach youths as resourceful informants concerning their sports participation’s financial boundaries.

Second, minority youths who drop out of sports consider discouragement from their parents more critical than majority youths. However, minority and majority youths are equally likely to say that sports expenses made them drop out. Youths who quit because their parents discouraged them may or may not be socially excluded, which may or may not be a problem. As the reason (discouraging parents) is somewhat outside the control of sports clubs, an important question is whether sports organizations should adjust their sport activities to the needs of minorities. Because minority girls drop out at greater rates than other youths, identifying the needs of minorities would likely include more dialogue with minority parents about potential religious barriers to sport participation (i.e. Strandbu, Bakken, and Sletten 2017), and potentially dialogue about how increasing demands of youth sports may conflict with an ‘ethnic minority drive’ to do well in school (Skauge and Hjelseth 2021). To gain a more effective channel into communities of minority parents who do not engage in the sports club’s daily activities, a partial solution would be to work hard to include more minority parents in voluntary work where much of the socialization in sports clubs happen for minority parents (i.e. Espedalen, Grønkjær, and Strandbu 2021) The study by Espedalen et al. indicate that minority parents who are less involved in voluntary work sometimes want to be more included, but lack information of how to contribute. This willingness to engage in voluntary work suggest that volunteerism may be an effective channel to maintain dialogue with parents.

Limitations

We used data from the Young in Oslo survey (Andersen and Bakken 2015), and while population-based surveys are essential because they enable empirical generalizations, this strength comes with limitations. First, the reasons for quitting sports are broad, and even though those herein cover important reasons for dropping out from sports, it is not an exhaustive list. For instance, previous research has indicated that black Norwegian athletes experience more covert racism (Massao and Fasting 2010, 2014), yet we could not test how important racism and social exclusion experiences were in youths’ decisions to quit.

Second, we encourage awareness of how to interpret the six reasons for dropouts in our study, especially ‘lacking any suitable option in the club’. Whether ‘suitable options’ could refer to characteristics such as fewer practices, more practices, bad coaches, or too competitive a culture is challenging to know. We know from categorizing open-ended answers in the 2018 Young in Oslo survey that teams dissolving, having to change teams, or lacking coach or facilities were common themes that deductively fit this category (Persson et al. 2020). Young Norwegians may, however, have a different understanding of this category than researchers. We suggest interpreting the category as a measurement of mismatch between youths’ demands for sports and the content sports clubs supply. We direct the reader to Persson et al. (2020) for a qualitative and nuanced understanding of how youths’ descriptions of dropout fit our study’s dropout reasons.

Third, to handle the complexity of sports participation and social inequality, it was necessary to simplify social backgrounds. We collapsed a diverse set of nationalities and cultural backgrounds into the term ‘minority youth’. Over half the sample had parents born in Pakistan, Somalia, and Sri Lanka. The rest had parents from other countries in Africa, Asia, and Eastern Europe. We thereby view our results as a much-needed baseline to compare with more specific and nuanced findings. A similar note should be considered regarding the variations in sports types within the umbrella term ‘organized sports’. However, there are ample empirical data to support that there is systematic underrepresentation in organized sports of girls, ethnic minority youths, and youths with lower SES. This suggest that specific variations between specific sports should be advanced in future research.

Concluding remarks

We studied how young people’s SES, gender, and ethnicity affect their reasons for dropout from organized sports. Our point of departure was that organized sports function as a good because they can provide belonging, mastery, fun, and health. Equal access to these goods is not the same as equal participation, and dropping out from sports is not the same as being excluded. Therefore, one essential measure governments and sports organizations can employ to combat social inequality is to understand the differences and similarities between youths with diverging social backgrounds when they drop out of sports. When some groups systematically drop out more from sports than others (e.g. minority girls), their reasons for dropout may in part reflect visible barriers in sport such as financial costs and in part reflect cultural differences related to social class, gender, and ethnicity. The cultural differences in reasons for dropout point to a system of dispositions, in the sense of habitus in Bourdieu’s theory and highlight the importance of intersectional approaches to understand how these systems of dispositions affect participation in organized sports. Qualitative approaches are necessary to nuance how these systems of dispositions relate to practice and dropout of specific sports.

Our results should be viewed as a comparative baseline for future qualitative research and depict well-resourced youths from majority backgrounds who – compared to their less resourceful minority counterparts – drop out of sports because they lack skills and their friends quit. The contrast is youths with minority backgrounds and fewer resources who seem more at ease with the skills sports require from modern youths. Nonetheless, less resourceful minority youths are pulled away from sports by parents more often than resourceful majority youths because parents discourage participation or due to financial constraints. Family dynamics are more difficult for sports clubs to notice because they play out in the sports arena’s periphery. Therefore, sports organizations and local governments should create collaboratives that include central actors such as schools, health, and sports organizations. When social inequality inside sports partially stems from processes outside sports, governments need to develop incentives for sports organizations and other arenas in young people’s lives to connect. Given that many youths find sports demanding irrespective of their social background, more playful and less time-consuming sports alternatives outside the traditional organized sports arena are also a demand waiting for supply.

Data availability statement

Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data are not available. For more information on the Ungdata-surveys (which the Young in Oslo survey is part of), see http://www.ungdata.no/English/.

Acknowledgments

The authors are grateful for valuable input from Professor Åse Strandbu in the earlier stages of the work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Norwegian Olympic and Paralympic Committee and Confederation of Sports (NIF).

Notes on contributors

Lars Erik Espedalen

Lars E. Espedalen is a PhD-student in Sociology of Sports at The Norwegian School of Sport Sciences. Espedalen's research specialise in parental involvement, social inclusion, and social inequality in organized youth sports.

Ørnulf Seippel

Ørnulf Seippel has a PhD in sociology (University of Oslo) and is Professor of Sociology of Sports at the Norwegian School of Sport Sciences. His research interests are sport and organizations, sport politics, social inequality, social capital/integration and social networks.