Psychology of Sport & Exercise 68 (2023) 102448

Contents lists available at ScienceDirect

Psychology of Sport & Exercise

journal homepage: www.elsevier.com/locate/psychsport

The interplay between parental behaviors and adolescents’ sports-related

values in understanding adolescents’ dropout of organized sports activities
Darun Jaf a, *, Stefan Wagnsson b, Therése Skoog c, Terese Glatz a, Metin Özdemir a
a
Center for Lifespan Developmental Research, Örebro University, Sweden
b
Sport Science at Karlstad University, Karlstad, Sweden
c
Department of Psychology, University of Gothenburg, Sweden

A R T I C L E I N F O A B S T R A C T

Keywords: Using short-term longitudinal data, the primary goal of the present study was to examine the interplay between
Parents’ behaviors adolescents’ sports-related intrapersonal (e.g., sports values) and interpersonal factors (e.g., perceived parental
Adolescents’ sports values involvement) in relation to sports dropout. A secondary goal was to explore the direction of effects in the as­
Sports dropout
sociation between intra- and interpersonal factors. A total of 420 adolescents (39% girls, Mage = 14.06; SDage =
Reciprocal associations
0.33) responded to a set of survey questions over two consecutive years. Results from structural equation
modeling suggested that parental involvement predicted adolescents’ dropout one year later, via sports values.
Further, the results suggested that the direction of influence is mainly from parents to adolescents. Overall, the
findings indicate that adolescents whose parents attend their practices and games perceive sports activities as
fun, important and useful; as a result, adolescents are less likely to dropout. The findings offer an improved
understanding of how parents’ behaviors may influence adolescents’ dropout of organized sports.

Organized sports activities have been identified as developmental perceived parental pressure and involvement in their children’s games
contexts that can promote positive youth development (Lerner, Lerner, and practices) as important correlates of youth sports dropout (Back,
& Benson, 2011; Mahoney, Larson, & Eccles, 2005). Generally, re­ Johnson, Svedberg, McCall, & Ivarsson, 2022; Bentzen, Hordvik, Sten­
searchers have reported positive associations between involvement in ersen, & Solstad, 2021; Crane & Temple, 2015). However, there are
organized sports and positive youth development across the academic, limitations in the literature and questions that need further attention.
behavioral, and psychological domains (Holt, 2007). Nevertheless, re­ First and foremost, the key correlates of sports dropout have mostly been
ports across several countries have consistently documented a substan­ examined as separate predictors; and, despite their close-inter-relations
tial decrease in youth’s sports participation rates, especially during the (Bremer, 2012; Holt & Knight, 2014a), separate lines of research have
transition between childhood and adolescence (Crane & Temple, 2015; focused on youth’s intrapersonal and interpersonal constraints on sports
Møllerløkken, Lorås, & Pedersen, 2015; Rullestad, Meland, & Mildest­ dropout. Further, the current understanding of sports dropout is mostly
vedt, 2021). The decline in sports involvement raises challenges for based on cross-sectional and qualitative studies (Back, Johnson, et al.,
parents, practitioners, and researchers who are interested in giving 2022; Crane & Temple, 2015; Møllerløkken et al., 2015), and longitu­
youth opportunities for successful adjustment. Hence, an understanding dinal studies are needed to identify and distinguish short-term and
of factors that could reduce sports dropout is important for promoting long-term influences on youth’s sports dropout. In an attempt to reduce
youth’s continued participation in a developmental setting that is linked this gap of knowledge, in the present study, we draw on the
to youth’s positive development (Lerner et al., 2011). Expectancy-Value Model (Fredricks & Eccles, 2004), to examine the
Researchers have consistently identified youth’s sports-related interactive roles of intrapersonal and interpersonal constraints on ado­
intrapersonal constraints (e.g., lack of enjoyment and not perceiving lescents’ sports dropout using short-term longitudinal data extending
sports as important/useful) and interpersonal constraints (e.g., over two years.

Abbreviations: EVM, The Expectancy-Value Model; YeS, Youth and Sports project; T1& T2, Time 1 and Time 2; SES, Socioeconomic status; SEM, Structural
equation modeling; CFA, Confirmatory factor analysis.
* Corresponding author. Center for Lifespan Developmental Research (LEADER), School of Behavioural, Social and Legal Sciences, Örebro University, Fakul­
tetsgatan 1, Örebro, 701 82, Sweden.
E-mail address: [email protected] (D. Jaf).

https://doi.org/10.1016/j.psychsport.2023.102448
Received 1 April 2022; Received in revised form 25 April 2023; Accepted 27 April 2023
Available online 1 May 2023
1469-0292/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

1. The expectancy-value model consistently demonstrated that youth who have higher attainment value
(i.e., the importance of doing well in an activity that is considered to
The Expectancy-Value Model (EVM) is a comprehensive model that strengthen the individual’s self-perception), intrinsic value (i.e., the
offers a theoretical understanding of individual differences in youth’s subjective interest or enjoyment experienced when participating in an
choices of activity involvement across the academic and extracurricular activity) and utility value (i.e., the importance of the task for current and
domains (Fredricks, Simpkins, & Eccles, 2006; Wigfield & Eccles, 2000). future goals) related to sports will be more likely to participate in sports
The model is often used to examine questions such as what drives an activities (Fredricks & Eccles, 2004; Jaf et al., 2021; Simpkins et al.,
individual to engage in a given activity, or why an individual pursues a 2012; Simpkins et al., 2015b). It has also been reported that the values
particular activity (e.g., sports) rather than another (e.g., music). Ac­ that youth attach to sports activities are not limited to guiding their
cording to the EVM (Eccles, 1983; Wigfield & Eccles, 1992), parents’ choice of activity enrolment but are also important for their commit­
socialization behaviors and youth’s subjective task values represent two ment once they have enrolled. For example, perceiving sports as
of the most important and direct determinants of youth’s choice of ac­ important, fun, and useful has been found to be negatively associated
tivity involvement. Overall, it is demonstrated that, through a variety of with youth sports dropout (Back, Johnson, et al., 2022; Boiché & Sar­
behaviors, parents can influence their children’s subjective task values, razin, 2009; Guillet, Sarrazin, Fontayne, & Brustad, 2006; Salguero,
which in turn predict the child’s choice of activity involvement (Jaf, Gonzalez-Boto, Tuero, & Marquez, 2003). Taken together the values that
Özdemir, & Skoog, 2021; Simpkins, Fredricks, & Eccles, 2015b). How­ youth attach to sports are important in understanding sports dropout. As
ever, arguments from the EVM and related empirical research have highlighted above proponents of the EVM argue that as one of the pri­
mainly focused on parents’ sports-related behaviors in the family mary socializers’ parents play a key role in the development of adoles­
context, such as parents’ own engagement in sports or playing sports cents’ sports-related values (Fredricks & Eccles, 2004; Horn & Horn,
with the child, in relation to the youth’s sports values and sports 2012). From a developmental perspective, initially, the parents’ role
participation (Jaf et al., 2021; Simpkins, Fredricks, & Eccles, 2012). involves introducing and exposing their children to various activities
Nevertheless, to our knowledge and with the exception of some scholars which scholars have consistently shown to influence the child’s
relying on cross-sectional data (Boiché & Sarrazin, 2009), there is a activity-related values and enrollment (Simpkins et al., 2012; Simpkins
limited understanding of whether the existing findings supporting the et al., 2015b). On the other hand, according to the EVM and others
EVM also hold across other contexts, such as the athletic domain and interested in parents’ behaviors in the athletic context (Fraser-Thomas,
specifically related to sports dropout. The EVM offers a great opportu­ Côté, & Deakin, 2008), during adolescence, parents generally have a
nity to examine the combined role of intrapersonal and interpersonal supportive role in youth sports, for instance, attending to the child’s
factors in relation to adolescents’ sports dropout, which may offer a games and practices, or offering encouragement, which has been shown
more comprehensive understanding of the role of parents’ behaviors in to be linked with youth’s concurrent sports-related motivational beliefs
their children’s sports dropout. Hence, in the present study, we rely on (Dorsch et al., 2021). However, to our knowledge, during this devel­
and extend arguments from the EVM to investigate the role of parental opmental stage, less is known about the interplay between parents’
behaviors in adolescents’ athletic contexts, such as parental involvement behaviors in the athletic context and youth’s sports-related values in
(e.g., attending practices and games) and encouragement (e.g., relation to sports dropout. In this study, extending previous knowledge
persuading youth to engage in sports), in relation to adolescents’ (Holt & Knight, 2014a; O’Neil & Amorose, 2021), and relying on the
sports-related values and dropout. EVM (Fredricks & Eccles, 2004), sports-related values are identified as a
potential mechanism that may explain the association between parents’
1.1. Parents’ sports-related socialization behaviors behaviors and adolescents’ sports dropout.

By being involved in their children’s sports activities, either by 1.3. The present study
attending practices and games or through encouragement (Simpkins
et al., 2015b), parents are investing “psychological resources”, like time The sophistication of the EVM (Fredricks & Eccles, 2004) in its
and interest (Connell & Wellborn, 1991). Through these behaviors, it is delineation of the mechanisms via which parents can influence their
argued that parents can continue to influence their children’s children’s motivational behaviors (e.g., choice of activity involvement)
sports-related experiences (e.g., enjoyment) and their continued enables the model to offer valuable insight into adolescents’ choices to
participation or dropout when they enter adolescence (Côté & Vierimaa, pursue or dropout from sports. Hence, to better understand the interplay
2014; Espedalen & Seippel, 2022; Fraser-Thomas, Côté, & Deakin, between intrapersonal and interpersonal factors on adolescents’ sports
2008). For example, it has been shown that parents’ involvement and dropout, we considered key arguments of the EVM. More specifically,
encouragement are positively linked to youth’s sports-related values the primary goal of the present study is to examine the indirect role of
(Bremer, 2012; Camacho-Thompson & Simpkins, 2020; Welk, Wood, & parents’ behaviors (e.g., attending games and practices, and encour­
Morss, 2003). Similarly, parents’ autonomy-supportive behaviors within agement of sports participation) in adolescents’ sports dropout, through
the athletic context have been shown to be positively associated with adolescents’ sports-related values. We hypothesize that by investing
adolescents’ enjoyment of and perceived importance of sports activities time and showing interest in their sports activities (via involvement and
(O’Neil & Amorose, 2021). Taken together, parents play an important encouragement) parents’ behaviors will be positively linked to their
role in youth’s sports-related motivational beliefs, the latter being one of adolescents’ perceiving sports as important, fun, and useful, which in
the most important correlates of sports dropout. However, except for the turn, will have a negative influence on dropout (see Figure 1).
findings of some qualitative studies (Fraser-Thomas, Côté, & Deakin, In line with the EVM (Fredricks & Eccles, 2004; Jacobs & Eccles,
2008; Stuart, 2003), little is known about the link between parents’ 2000) and considering the lack of empirical research (Back, Johnson,
behaviors in the athletic context and youth’s sports dropout, that is et al., 2022; Holt, Tamminen, Black, Mandigo, & Fox, 2009; Knight,
youth’s choice to pursue or dropout of sports activities. In the present 2019), we also examined the reciprocal associations between adoles­
study, using data collected over two-time points we aim to understand cents’ perceptions of parents’ behaviors (i.e., involvement and encour­
whether and how parents’ involvement and encouragement are linked agement) and their perceived sports-related values. According to the
to adolescents’ sports dropout. EVM (Fredricks, Simpkins, & Eccles, 2005; Jacobs & Eccles, 2000), the
path between parents’ behaviors and youth’s motivational beliefs may
1.2. Youth’s subjective values not necessarily be unidirectional, where parents influence youth’s
values. Rather, parents may also adjust their behaviors according to
In line with the EVM (Jacobs & Eccles, 2000), empirical research has their children’s beliefs and behaviors (Fredricks & Eccles, 2004). In fact,

2
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

Figure 1. The Hypothesized Half-Mediation Model

Examining the Indirect Effect of Parents’ Behaviors
on Adolescents’ Sports Dropout through Adolescents’
Sports Values
Note. 1 = Time 1 and 2 = Time 2. Ovals represent
latent constructs. The latent factors and the outcome
variable (i.e., sports dropout) were all regressed on
the covariates (age, gender, immigrant background,
and family SES).

in qualitative studies, parents have reported that their beliefs and atti­ participation in the study, and 6% at T2. A small proportion of the youth
tudes towards sports, and their behaviors in youth sports, have changed did not take part in the survey due to sickness at T1 (n = 6) and T2 (n =
alongside their youth’s enrolment (for an overview, see Knight, 2019). 7), and due to truancy or for other (unknown) reasons at T1 (6%) and T2
For example, some parents reported that they developed an interest in (7%). Most of the participating youth were from intact families (63%),
the same sport as their children, which resulted in them being more with employed parents (77% for mothers, and 89% for fathers), and 83%
involved as spectators (e.g., attending practices and games), and even perceived their financial situation to be just as good or better than their
starting in the same sport themselves (Côté, 1999). Similar findings have classmates. About one-third (29%) of the adolescents and their parents
been reported by others (Weiss & Hayashi, 1995). To explore the hy­ were born outside a Nordic country (Sweden, Denmark, Finland, or
pothetical bidirectional effects, in the present study we considered the Norway); most of them spoke Swedish at home (70%), while about a
direction of influence in the association between parents’ behaviors and third (30%) spoke a foreign language or a combination of a foreign
adolescents’ sports-related values. language and Swedish. The analytic sample for the present study
The gender of the child could play an important role in the EVM comprised 420 adolescents involved in after-school sports activities
(Eccles & Wigfield, 2020). It is argued that parents who hold (39% girls, Mage = 14.06 and SDage = 0.33).
gender-stereotypic beliefs about their children will also act on such
beliefs in their socialization behaviors. For example, it has been found
2.2. Procedure
that parents are more likely to buy sports-related equipment for their
sons than for their daughters (Simpkins et al., 2012; Simpkins et al.,
Parents received an information letter by regular post with a
2015b). Relatedly, according to the model (Eccles & Wigfield, 2020;
description of the YeS project and were asked to return a pre-paid en­
Fredricks & Eccles, 2004), cultural beliefs could also influence parents’
velope if they did not want their child to participate in the study. The
general and for instance sports-related socialization behaviors. In line
data collection was administered by 2–3 trained research assistants.
with these arguments, research has shown mean-level differences be­
Before the survey questions, research assistants presented a thorough
tween immigrant and native parents’ behaviors on a daily basis, such as
description of the project, a statement about the voluntary nature of
parental monitoring (Delforterie et al., 2016; Fernandez, Loukas, &
participation and confidentiality of information, and an assurance that
Pasch, 2018; Jaf, Özdemir, & BayramÖzdemir, 2020), and levels of
participants could quit the study whenever they wanted. The study was
parents’ own engagement in physical and sports activities (Jaf et al.,
approved by the Regional Ethics Review Board in Uppsala (Dnr: 2015/
2020). Nevertheless, these differences have been mostly observed on a
330).
mean level and have not entailed differences in developmental processes
(Jaf et al., 2020; Jaf et al., 2021; Simpkins et al., 2015b). Taken together,
3. Measures
based on arguments from the EVM (Eccles & Wigfield, 2020) and related
empirical findings, gender and immigration background were treated as
3.1. Adolescents’ perceptions of parents’ behaviors
covariates.

The items for adolescents’ perceptions of their parents’ behaviors

2. Method
were developed as part of the YeS project. They are in line with the EVM
(Fredricks & Eccles, 2004), and the practice of researchers interested in
2.1. Participants
parents’ behaviors in the athletic context (Boiché & Sarrazin, 2009; Lee
& MacLean, 1997).
Data were obtained from the Youth and Sports (YeS) project, which
Parents’ involvement in adolescents’ sports activity. This construct
has several aims, including an understanding of how intrapersonal and
was measured using the same three items at T1 and T2: “How often do
interpersonal factors might explain youth’s dropout and sports
your parents come and watch you practicing?”, “How often do your
involvement over time. We used data across two-time points, time 1 (T1)
parents come and watch your matches/games?”, and “How often do
and time 2 (T2), which were collected annually from public-sector
your parents talk to your coach about your performance?”. The items
schools in a medium-sized city in central Sweden. To match the socio-
were rated on a 5-point scale ranging from 1 (Never) to 5 (Always).
demographic characteristics of the city the schools were selected from
Similar items have been used to assess youth’s perceptions of their
different neighborhoods. The city was very close to the national average
parents’ involvement and have been shown to have acceptable internal
in terms of annual income (303 300 Swedish kronor/person compared
consistency and predictive validity (Boiché & Sarrazin, 2009; Bremer,
to the national average of 300 000 Swedish kronor/person) and in terms
2012).
of the unemployment rate (7%) which was equal to the national average
Parents’ encouragement of adolescents’ participation in sports.
(7%) (Statistics Sweden, 2017).
This construct was measured using the same 3 items at T1 and T2: “My
Out of the target sample (812 students in grade 7), 678 (83%, 46%
parents want me to play sports.”, “My parents want me to become a
girls, Mage = 14.09 and SDage = 0.39) were present during T1 and 83%
professional athlete.”, and “My parents get happy when I play sports.”,
responded to the survey again at T2 (N = 563, 44% girls, Mage = 15.09
rated on a 4-point scale ranging from 1 (Not correct at all) to 4 (Very true).
and SDage = 0.38). At T1, 4% of parents declined their children’s
Similar items have been used to assess youth’s perceptions of their

3
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

parents’ encouragement of sports activities, and have been shown to longitudinal attrition (Nagelkerke R2 = 0.03). The p-values for the un­
have acceptable internal consistency and predictive validity (Brown, standardized regression coefficients ranged between − .67 and .09.
1985; Fredricks & Eccles, 2005).
3.6. Plan of analysis
3.2. Adolescents’ sports-related values
Structural equation modeling (SEM) in Mplus version 8.4 was used to
Adolescents’ sports values were measured using the same items at T1 test the primary and secondary aims of this investigation. To assess
and T2. Different items were used to tap into the multidimensional model fit, we followed the conventional cutoff points in the literature
construct of sports-related values (Fredricks & Eccles, 2004). For (Wang & Wang, 2020), which comprise: (1) model chi-square statistics,
example, one item was used to assess adolescents’ intrinsic values: “I do (2) the comparative fit index (CFI >0.95), (3) the Tucker-Lewis index
sport because it’s fun”, rated on a 4-point scale ranging from 1 (Not (TLI >0.95), (4) the root mean square error of approximation (RMSEA
correct at all) to 4 (Very true). Two items were used to assess adolescents’ <0.10), and standardized root means residuals (SRMR <0.10). The
utility values: “I do sport because it makes me physically stronger”, “I do constructs used in the present study are measured through observed
sport because I can make new friends”, rated on a 4-point scale ranging indicators that are ordinal or dichotomous. Hence, following recom­
from 1 (Not correct at all) to 4 (Very true). One item was used to assess mendations on this topic, weighted least squares with means and vari­
adolescents’ attainment values: “I spend more time thinking about the ances adjusted (WLSMV) robust estimators were used throughout the
sport than anything else”, rated on a 5-point scale ranging from 1 analyses (Brown, 2015; Flora & Curran, 2004; Rhemtulla,
(Disagree) to 5 (Agree). Similar items have been used across a range of Brosseau-Liard, & Savalei, 2012).
studies and have been shown to have high internal consistency and
predictive validity (Fredricks & Eccles, 2005; Simpkins et al., 2012). 3.6.1. Measurement model
Confirmatory factor analysis (CFA) was used to evaluate the factorial
3.3. Sports Dropout structure of the measurement model and to evaluate measurement
invariance across groups and time. The measurement model comprised
Consistent with the literature we measured sports withdrawal as 3 distinct latent factors across two-time points, resulting in a total of six
domain-general dropout, that is adolescents who dropped out and did latent factors, with one latent factor measuring adolescents’ sports
not reenter the same or other sports activities (Carlman, Wagnsson, & values, and two distinct latent factors measuring parents’ behaviors. The
Patriksson, 2013; Gould & Petlichkoff, 1988; Jaf et al., 2020). Initially (i. latent factors were identified by setting the first factor loadings at 1.
e., at T1), we measured adolescents’ involvement in organized sports Measuring the same items repeatedly over time tends to result in
activities using one dichotomous item: “Are you involved in after-school correlated measurement errors (Little, 2013a), more specifically,
sports activities, for example, soccer, ice-hockey, horse riding, and indicator-specific effects including indicator-specific residuals are ex­
athletics?”. About one year later (i.e., at T2), we asked the same ques­ pected to remain over time (Geiser, 2021; Hermida, 2015); hence, we
tion. Based on the responses, we identified adolescents who were correlated the error terms of the same items that were measured
involved in sports activities during T1 (n = 420) but no longer involved repeatedly across T1 and T2.
during T2 (n = 52). Adolescents who persisted in sports activities over The latent factors were created based on arguments from the EVM
time were coded as 0, and those who were involved in sports activities at (Fredricks & Eccles, 2004) and related empirical findings (Simpkins
T1 but were no longer involved at T2 (i.e., who had dropped out of et al., 2015b) emphasizing the importance of adolescents’
sports) were coded as 1. activity-related values (e.g., intrinsic, utility, and attainment values) and
parents’ behaviors (e.g., parental involvement and encouragement) on
3.4. Socio-demographic characteristics and covariates youth’s involvement in extracurricular activities. Even though
activity-related values consist of theoretically distinct components, they
Participants reported their age and gender, and parental employ­ are highly correlated (Eccles & Wigfield, 1995) and often used as an
ment status was used as a proxy of socioeconomic status (SES). overarching single scale of values (Fredricks & Eccles, 2005; Simpkins,
Following the practice of earlier researchers (Peguero, 2011; Singh, Fredricks, & Eccles, 2015a). Thus, relying on previous research (Simp­
Stella, Siahpush, & Kogan, 2008), adolescents with both parents born kins et al., 2012; Østerlie, Løhre, & Haugan, 2019), and to obtain a
outside a Nordic country (Sweden, Norway, Finland, or Denmark) were parsimonious model, we measured adolescents’ sports-related values as
regarded as immigrants. Participants with both of their parents born in a a unidimensional latent factor comprising one item regarding intrinsic
Nordic country were regarded as Nordic. A small group of adolescents value, two items regarding utility value, and one item regarding
(7%) had only one parent born outside the Nordic countries. This group attainment value. For adolescents’ perceptions of their parents’ behav­
was more like the Nordic adolescents than immigrants on all study iors, two distinct latent factors were created to measure parental
variables, thus, were regarded as Nordic. Immigrant adolescents were involvement (3 items) and parental encouragement (3 items). We did
coded as 1, and Nordic adolescents as 0. In the present study, we not merge parents’ behaviors into one single latent factor since, unlike
controlled for the effects of the following covariates: age, gender, parental involvement, indicators of parental encouragement represent
immigrant background, and SES. parents’ desire for their children to be involved in sports and to become
professional athletes. Hence, in parental encouragement, the focus is
3.5. Attrition and missing data analysis rather on what parents want from their children (Fredricks & Eccles,
2004). Such behaviors can be perceived by youth as pressuring and have
Of the analytic sample at T1 (n = 420), 87% were also present at T2 a negative influence on their sports-related experiences (Dorsch et al.,
(n = 365). Thus, 13% of the adolescents who were in the study during T1 2021).
did not participate in T2. We performed binomial logistic regression to
examine whether the study variables (i.e., age, gender, SES, immigration 3.6.2. Measurement invariance
background, parents’ involvement and encouragement, and adoles­ Once the factorial structure of the measurement model was assessed,
cents’ sports-related values) were systematically related to longitudinal we evaluated measurement invariance across groups and time. Research
attrition. We regressed all study variables on the attrition variable which has consistently identified youth’s sports-related values (e.g., engaging
identified those who remained in the study on both occasions (0 = in sports for fun) as one of the most important factors correlated with
stayed) and those who were missing at the second data collection (1 = youth’s participation and continuation in organized sports activities
attrition). None of the study variables significantly predicted (Crane & Temple, 2015; Fredricks & Eccles, 2005). Similarly, in the

4
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

present study, most adolescents across T1 and T2 scored high on the indicated an acceptable model fit: χ 2(145) = 236.187, p < .001; CFI =
indicators measuring their perceived sports-related values. Conse­ 0.98 and TLI = 0.97; RMSEA = 0.039 (90% CI: 0.029, 0.047, p = .986);
quently, assessment of measurement invariance through multi-group and SRMR = 0.06. All of the standardized factor loadings for the latent
factor analysis resulted in indicators with several categories that had variables were statistically significant (see Table 1) and above the rec­
low observations or empty response options (i.e., zero cells), which in ommended cut-off value of 0.40 (Wang & Wang, 2020). Following the
turn produced incorrect correlation estimates. Solutions such as recommendations of (Liu et al., 2017; Svetina, Rutkowski, & Rutkowski,
collapsing items (Rutkowski, Svetina, & Liaw, 2019) or adding a value 2020) and the findings of (Chen, 2007; Cheung & Rensvold, 2002), we
(e.g., .5) to the zero frequency cells (Savalei, 2011) have been shown to decided upon a change in Δχ 2 statistics and change in CFI and RMSEA as
result in biased standard errors, scale reliability estimates, and model fit. the criteria for determination of measurement invariance. A decrease in
Therefore, no adjustment is recommended for indicators that consist of CFI ≥0.01 and an increase in RMSEA ≥0.015 is considered an indication
more than two response options (Savalei, 2011). Hence, we conducted of non-invariance. The findings indicated scalar invariance across
measurement invariance only for the latent factors measuring parent’s immigration background Δχ 2(26) = 58.96, p > .001; ΔCFI = − 0.008,
involvement and encouragement across time and groups (i.e., gender ΔRMSEA = 0.006, gender Δχ 2(26) = 31.70, p = .203; ΔCFI = − 0.002,
and immigration background). ΔRMSEA = 0.00, and over time Δχ 2(13) = 21.22, p = .069; ΔCFI =
− 0.001, ΔRMSEA = 0.00.
3.6.3. Structural model
To address the main goal of this investigation, which draws on ar­
4.2. The indirect effects of parents’ Behaviors on Adolescents’ sports
guments from the EVM (Fredricks & Eccles, 2004), we examined
dropout via adolescents’ sports-related values
whether adolescents’ sports values at T1 mediated the association be­
tween parents’ behaviors at T1 and adolescents’ dropout from sports
We examined whether adolescents’ sports values at T1 mediated the
activities at T2. Figure 1 shows that two separate mediation effects were
association between parental involvement and encouragement at T1 and
estimated within the same model: (1) concerning the link of parental
adolescents’ sports dropout at T2. The indices indicated an acceptable
involvement via adolescents’ sports values to sports dropout, and (2)
model fit between the specified SEM model and the observed data:
concerning the link of parental encouragement via adolescents’ sports
χ 2(117) = 194.774, p < .001; CFI = 0.97 and TLI = 0.96; RMSEA =
values to sports dropout. Following recommendations for mediation
0.042 (90% CI: 0.032, 0.053, p = .884); and SRMR = 0.10. The direct
analysis with two measurement occasions, we applied a half-mediation
effects revealed that adolescents’ sports-related values were stable over
model (Little, 2013a). More specifically, we controlled for the previous
time and parents’ involvement in sports activities at T1 positively pre­
levels of the mediator variable, in this case, adolescents’ sports values
dicted adolescents’ sports-related values one year later (see Table 2). By
(see, Figure 1), and given that we focused on adolescents that were
contrast, parental encouragement at T1 had no significant direct effect
already engaged in sports activities at time 1, per default, we also
on adolescents’ sports-related values one year later. As for the main
controlled for the previous levels of the outcome variable (i.e., sports
findings, results from the half-mediation model suggested that parental
participation/dropout). Further, throughout the analysis, all latent fac­
involvement in their sports activities positively predicted adolescents’
tors at T1 and the outcome variable at T2 were regressed on the cova­
riates (age, gender, immigrant background, SES). However, to avoid
Table 1
statistical overcontrolling (Little, 2013a), the covariates that did not
Standardized coefficients from the confirmatory factor analysis.
have marginal associations (p < .10) with the model variables were
removed from the models. Specifically, in line with the recommendation Observed variable Latent Time 1 Time 2
factor β (SE) β (SE)
of the American Statistical Association (Nuzzo, 2014; Wasserstein &
Lazar, 2016), we omitted covariates that did not have marginal effects Parents’ Involvement
from the final model if their respective 90% confidence intervals, cor­ How often do your parents come and watch .76 (.03) .76 (.04)
responding to marginal effects, included the value of zero. The recom­ you practicing? *** ***
mendations from the American Statistical Association were also used as How often do your parents come and watch .86 (.03) .84 (.04)
your matches/games? *** ***
the primary source of guidance to interpret results from the present
How often do your parents talk to your coach .76 (.03) .78 (.04)
study. about your performance? *** ***
The mediation coefficients examined in the present study are each
Scale reliability .84 .83
the product of two regression coefficients (i.e., product 1 = a1*b1, and
product 2 = a2*b1). Hence, the assumption of normal distribution of a Parents’ Encouragement

computed product term could be violated (Geiser, 2012). Given rec­ My parents want me to play sports .77 (.04) .86 (.04)
ommendations for testing the significance of indirect effects, we *** ***
My parents want me to become a .76 (.04) .82 (.04)
included 95% bias-corrected bootstrap (BC) confidence intervals based on
professional athlete *** ***
1000 bootstrap draws (MacKinnon, 2008; MacKinnon, Lockwood, & My parents get happy when I play sports .78 (.04) .76 (.04)
Williams, 2004). The second part of the analysis involved the exami­ *** ***
nation of the reciprocal relationships between (1) parents’ involvement Scale reliability .81 .86
and adolescents’ sports values, and (2) parents’ encouragement and
Adolescents’ Sports values
adolescents’ sports values, using latent autoregressive cross-lagged
modeling (Geiser, 2012; Selig & Little, 2012). For instance, using data I do sport because it is fun .76 (.06) .67 (.07)
across two-time points, we examined reciprocal relationships and the *** ***
I do sport because it makes me physically .60 (.05) .66 (.07)
stability and change of parents’ behaviors and adolescents’ sports values stronger *** ***
(i.e., the auto-regressive and cross-lagged effects of the respective I do sports because I can make new friends .64 (.05) .64 (.06)
constructs). *** ***
I spend more time thinking about sport than .63 (.05) .61 (.07)
anything else *** ***
4. Results
Scale reliability .75 .74
4.1. CFA and measurement invariance Note. β = standardized beta coefficients, and SE = standard error. Measurement
scale of each latent variable was established by setting the first indicator at 1. *p
Results from the CFA concerning the overall measurement model < .05; **p < .01; ***p < .001.

5
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

Table 2 adolescents’ sports values over time (see Table 3). Second, and most
Path estimates predicting adolescents’ sports values and dropout. importantly, the cross-lagged estimation revealed that perceived
Path β (SE) 95% CI parental involvement at T1 predicted adolescents’ sports values at T2.
Hence, above and beyond the stability of adolescents’ sports values over
LL UL
time, perceived parental involvement predicted adolescents’ sports
Covariates values one year later. The findings also revealed that there was no
Immigrant background → Parents’ − .29 (.07) − .402 − .127
involvement1 ***
reciprocal association between perceived parental encouragement and
Immigrant background → Dropout2 .37 (.07)*** .244 .500 adolescents’ values.
Gender → Parents’ encouragement1 .28 (.06)*** .173 .385
Gender → Sports values1 .21 (.07)*** .068 .310 5. Discussion
SES → Parents’ encouragement1 − .16 (.07)* − .306 − .041
Direct effects
Sports values1 → Sports values2 .77 (.10)*** .60 1.03 The present study examines an important gap in the literature con­
Parents’ involvement1 → Sports values2 .18 (.08)* .02 .33 cerning the interplay between intrapersonal and interpersonal factors in
Parents’ encouragement1 → Sports values2 − .04 (.11) − .36 .13 adolescents’ dropout of organized sports. Hence, the findings contribute
Sports values1 → Dropout2 − .39 (.10) − .57 − .22 to the literature by offering additional theoretical and empirical insights
***
into processes involved in adolescent’s sports dropout. Specifically, in
B (SE) LL UL the present study, we focused on the indirect role of parents’ behaviors
Indirect effect in the athletic context on adolescents’ dropout from sports via adoles­
Parents’ involvement1 → Sports values1 → − .08 (.05)* − .20 − .01 cents’ sports-related values. Further, we investigated reciprocal associ­
Dropout2
ations between parents’ behaviors and adolescents’ values to explore
Parents’ encouragement1 → Sports values1 → .02 (.05) − .06 .17
Dropout2
questions related to directionality.

Note. 1 = Time 1 and 2 = Time 2. β = standardized beta coefficients, B = un­

5.1. The role of parents’ behaviors in their adolescents’ choices to pursue
standardized beta coefficients, and SE = standard error. CI = confidence inter­
or dropout from sports
val; LL = lower limit; UL upper limit. 95% BC bootstrap CI reported for all
estimates. Confidence intervals that do not contain zero are statistically signif­
icant. * = p < .05; ** = p < .01; *** = p < .001 Building on arguments from the EVM (Fredricks & Eccles, 2004), the
present study extends existing knowledge on parents’ involvement in
youth sports by demonstrating that parents’ behaviors, in particular
sports values, which, in turn, were negatively associated with sports
their perceived involvement, are important not only for their adoles­
dropout. Contrary to our expectations, there was no significant indirect
cents’ sports-related values (Babkes & Weiss, 1999; Camacho-Thompson
effect of parental encouragement.
& Simpkins, 2020; Hoyle & Leff, 1997) but also for counteracting
dropout from sports. These findings are in line with the idea that
4.3. Reciprocal relationships external support is necessary for adolescents to maintain their values
and involvement in activities over time (Connell & Wellborn, 1991;
The indices for the cross-lagged model of parents’ involvement and Côté, 1999; Jacobs & Eccles, 2000). There are two important aspects.
adolescents’ sports values indicated acceptable model fit between the First, as they grow older, the amount of time adolescents spend with
specified model and the observed data: χ 2(201) = 303.592, p < .001; their parents decreases dramatically, while the positive qualities of
CFI = 0.97 and TLI = 0.97; RMSEA = 0.037 (90% CI: 0.028, 0.045, p = parent-adolescent relationships, such as parental closeness and support,
.996); and SRMR = 0.08. First, we found strong stability estimates for decrease (Smetana, Robinson, & Rote, 2015). At the same time, most
perceived parental involvement and encouragement, and for adolescents spend more time with their peers, and the importance of
relationships with peers increases as they become more intimate with
Table 3
them. Despite these changes during adolescence, the present study in­
Reciprocal associations between parents’ behaviors and adolescents’ sports dicates that having parents who attend practices and games may play a
values. crucial role in the extent to which adolescents value sports activities,
and in their continued participation in sports. Second, although theo­
β (SE) 95% CI
retical models of sports dropout, such as the developmental model of
LL UL sports participation (Côté, Baker, & Abernethy, 2003), have highlighted
Covariates the importance of parental behaviors in relation to youth sports dropout,
Immigrant background → Parents’ involvement1 − 27 (.07) − .41 − .13 less is known about potential underlying mechanisms. The studies that
***
1 have addressed this issue are mostly qualitative (Fraser-Thomas, Cote, &
Gender → Parents’ encouragement 27 (.06)*** .17 .38
SES → Parents’ encouragement1 − .16 (.07)* − .27 − .05 Deakin, 2008; Stuart, 2003), but findings from the present study suggest
Gender → Parents’ encouragement2 .12 (.07)† .01 .24 that adolescents’ sports-related values mediate the link between par­
SES → Parents’ encouragement2 .15 (.09)† .01 .29 ents’ behaviors and dropout, and therefore parents’ behaviors seem to
Gender → Sports values1 .20 (.07)** .07 .33 have an indirect influence on adolescents’ dropout.
Stability paths
Sports values1 → Sports values2 .77 (.08)*** .61 .94
Contrary to our expectation, there was no direct or indirect effect of
Parents’ involvement1 → Parents’ involvement2 .76 (.04)*** .67 .84 parental encouragement on adolescents’ sports-related values and
Parents’ encouragement1 → Parents’ .59 (.08)*** .43 .75 dropout from sports. To some extent, these findings are consistent with
encouragement2 the literature, where no associations between parental encouragement
Cross-lagged paths
and children’s perceived sports-related competence and intrinsic moti­
Sports values1 → Parents’ involvement2 .02 (.07) − .12 .15
Parents’ involvement1 → Sports values2 .19 (.08)** .04 .34 vation have been observed (Babkes & Weiss, 1999). It should be noted
Sports values1 → Parents’ encouragement2 − .10 (.09) − .28 .08 that this null finding could be related to how parental encouragement
Parents’ encouragement1 → Sports values2 − .04 (.09) − .22 .13 was measured in our and some of the prior studies. Guided by the EVM
Note. 1 = Time 1 and 2 = Time 2. β = standardized beta coefficients and SE = (Eccles, 2007), parental encouragement was conceptualized in the pre­
standard error. CI = confidence interval; LL = lower limit; UL upper limit. sent study as adolescents’ perceptions of being encouraged by their
Confidence intervals that do not contain zero are statistically significant. † = p < parents to play sports, and of expectations to excel in sports. However,
.10; * = p < .05; ** = p < .01; *** = p < .001 adolescents could have perceived these parental behaviors and

6
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

expectations as a source of pressure rather than support. The behaviors, responsibilities in a sports association, e.g., as a coach, board member or
especially those involved in becoming a professional athlete, are to some sports official (Riksidrottsförbundet, 2019; Wagnsson & Patriksson,
extent overlapping with performance-focused parenting (Dorsch et al., 2005). The uniqueness of the Swedish sports culture is even more
2021) and can be perceived by adolescents as involving excessive evident when Sweden is compared with other countries across Europe
parental expectations. In fact, excessive parental behaviors (e.g., high (Eurobarometer, 2019). Sweden has the highest proportion of its pop­
expectations on performance) have been associated with a range of ulation (19%) involved in sports-related voluntary work. Taken
negative outcomes in youth sports including perceived pressure to play together, the evidence suggests that parents and other adult figures
sports (Holt & Knight, 2014a; Dorsch et al., 2021; O’Neil & Amorose, could be involved in youth sports as a result of the Swedish sports cul­
2021). Further, qualitative findings have revealed that youth whose ture rather than being influenced by their children.
parents set excessive expectations on sports-related achievements are It is also important to highlight that despite a well-established sports
more likely to focus on meeting their parents’ expectations in future culture, Sweden like other countries (Breuer, Feiler, & Wicker, 2015;
sports-related events (Sagar & Lavallee, 2010) rather than on their own Rullestad et al., 2021; Toselli & Belcastro, 2017) faces challenges related
preferences to remain active or dropout. Nevertheless, it is also impor­ to gender stereotypic behaviors and inequality in sports participation
tant to acknowledge that the association between parental encourage­ rates between Swedish and immigrant families. For instance, our find­
ment, whether performance-oriented or not, and youths’ motivational ings suggested that girls were less likely to be encouraged by their
beliefs (e.g., enjoyment or sports) depends on whether parents are parents to participate in sports and reported lower sports-related values
autonomy-supportive or controlling during their interactions with the than boys. Further, our findings also indicated that immigrant parents
child (O’Neil & Amorose, 2021), and if they provide positive or negative were less likely to be involved in their adolescents’ organized sports
evaluations related to youths’ sports-related performances (Dorsch activities, and immigrant adolescents were more likely to dropout of
et al., 2021). Taken together, future research may benefit from identi­ sports activities than their Swedish counterparts. Taken together, future
fying different dimensions of parental encouragement and accounting research could benefit from considering the implications of broader
for parenting style (e.g., autonomy-supportive vs. controlling), to better contextual factors (e.g., cultural norms and values) for a more compre­
understand when it is perceived as supportive and/or pressurizing, and hensive understanding of parental behaviors in youth sports contexts
how it influences adolescents’ sports values and continued participation and adolescents’ sports dropout. Especially, to promote equal opportu­
in sports. nities to participate in organized sports activities regardless of gender
and immigrant background of parents and their adolescents.
5.2. Reciprocal associations between parents’ behaviors and adolescents’
sports-related values 6. Strengths and limitations

In line with seminal papers on the dynamic nature of socialization The present study has some limitations that need to be addressed.
(Bell, 1968; Sameroff, 1975a, 1975b), proponents of the EVM have First, even though we reveal that parents’ behaviors, particularly
argued that the link between parents’ behaviors and youth’s motiva­ parental involvement, predict change in adolescents’ sports-related
tional beliefs is reciprocal (Fredricks & Eccles, 2004; Jacobs & Eccles, values over time, the items used to tap into parents’ involvement pro­
2000). However, to our knowledge, reciprocal associations have not vide limited information about the types of behaviors that parents
been examined in the literature on parental behaviors in youth’s engage in when they attend games and practices or talk with the coach.
sporting contexts (for a review see Crane & Temple, 2015; Møllerløkken Parents might engage in a range of behaviors (e.g., cheering on their
et al., 2015; Back, Stenling, et al., 2022). Consequently, we examined child vs. criticizing their child) that might determine whether their
the data for potential bidirectional associations between parents’ be­ involvement is perceived as supportive or pressurizing (Holt & Knight,
haviors in the athletic setting and adolescents’ sports-related values over 2014b). Including both supportive and pressurizing behaviors within
time. Our findings provide additional support for the results obtained the same model may offer an additional understanding of the roles of
from the proposed mediation model (see Figure 1), where parents’ be­ different types of parental involvement in youth’s sports-related values
haviors, especially parental involvement, play an important role in ad­ and their future commitment to sports. Several qualitative studies (for
olescents’ development of sports-related values. Contrary to our an overview, see Holt & Knight, 2014b) have provided a list of parental
expectations, there were no significant effects of adolescents’ behaviors in sports contexts that youth regard as desirable (e.g., main­
sports-related values on parental behaviors (e.g., parental involvement taining a positive tone and body language) or as undesirable (e.g.,
and encouragement). Thus, findings from the present study suggest that criticizing their child or the team). Future research could benefit from
the direction of influence is from parent to adolescent, not the other way using this type of knowledge for a more detailed understanding of
around. effective parental behaviors.
Overall, these findings can be explained by broader contextual fac­ Second, we focused mainly on parents’ behaviors as a potential
tors, namely the prevailing sports culture in Nordic countries such as source of influence on adolescents’ sports-related values and dropout.
Sweden. Consistent with the ecological view of human development We did not control for other important factors, such as peer and coach
(Bronfenbrenner, 2005), proponents of the EVM argue that contextual relations, that have been linked to youth’s overall experiences, sports
factors, such as cultural norms and values, play a key role in parents’ values, and sports dropout (Back, Stenling, et al., 2022; Crane & Temple,
general and specific behaviors (Eccles, 2007; Eccles & Wigfield, 2020). 2015; Møllerløkken et al., 2015). To develop a more comprehensive
In Sweden, sports and physical activities are an important part of understanding of adolescents’ sports dropout, research would benefit
contemporary culture, which is clearly shown in governmental policy from examining the joint effect of the activities of the people in a sports
statements, alongside the substantial financial resources (Bergsgard, context (e.g., parents, peers, and adult-leaders/coaches). Third, even
Borodulin, Fahlen, Høyer-Kruse, & Iversen, 2019; Fahlén & Stenling, though repeated data over two-time points was used to address the main
2016) provided to support the Swedish Sports Confederation. Further, it goal of the study, a minimum of three waves of data is preferred (Little,
is important to acknowledge that the sports activities investigated here 2013b) to properly examine mediation models and to distinguish
took place outside Swedish school hours. Hence, the family (e.g., parents within-person processes from between-person differences that may exist
or other adult care-takers) was found to play a key role in facilitating in the data (Hamaker, Kuiper, & Grasman, 2015). Hence, future research
youth’s participation in organized sports. In fact, national reports have with extensive longitudinal data is required to rigorously support our
revealed that a large majority of parents engage in their children’s findings concerning the mediation effects and to address questions
organized sports activities, not only by offering transportation and related to stable trait-like individual attributes. Finally, even though not
attending games and practices but also by taking, mostly pro bono, part of the aim of this study, we do not know if adolescents who did not

7
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

dropout of organized sports over time remained in the same type of Data availability
sports activity.
Despite its limitations, the present study has some considerable The data that has been used is confidential.
strengths. To our knowledge, existing research has mainly studied
youth’s dropout from sports by focusing separately on interpersonal and Acknowledgments
intrapersonal factors. To develop a more comprehensive understanding,
we applied theoretical arguments from the EVM (Fredricks & Eccles, The data used in this study were collected as part of the Youth &
2004), alongside existing knowledge, to examine the joint effect of Sports Project (YeS), at the Center for Lifespan Developmental Research,
interpersonal factors (e.g., parents’ involvement and encouragement) Örebro University. Responsible for planning, implementing, and
and intrapersonal factors (e.g., sport is fun, important, and useful) on financing the collection of data were Metin Özdemir, Ph.D. (Principal
adolescents’ dropout from sports activities. Hence, one of the main Investigator), and Therése Skoog, Ph.D. (Co-Investigator).
strengths of the study is that it offers additional insights into the pro­
cesses involved in adolescents’ dropout. A second strength concerns the References
use of adolescents’ perceptions of their parents’ behaviors. Researchers
have emphasized that effective socialization is influenced by how ado­ Babkes, M. L., & Weiss, M. R. (1999). Parental influence on children’s cognitive and
lescents perceive their parents’ intentions (Grusec & Goodnow, 1994; affective responses to competitive soccer participation. Pediatric Exercise Science, 11
(1), 44–62.
Smetana et al., 2015). It has been shown that children’s perceptions of Back, J., Johnson, U., Svedberg, P., McCall, A., & Ivarsson, A. (2022). Drop-out from team
their parents’ behaviors are a more important source of influence on sport among adolescents: A systematic review and meta-analysis of prospective studies.
children’s sports-related experiences than what their parents report Psychology of Sport and Exercise, Article 102205.
Back, J., Stenling, A., Solstad, B. E., Svedberg, P., Johnson, U., Ntoumanis, N.,
(Babkes & Weiss, 1999; Hoyle & Leff, 1997; Leff & Hoyle, 1995). The Gustafsson, H., & Ivarsson, A. (2022). Psychosocial predictors of drop-out from
short-term prospective design of the present study is a third strength that organised sport: A prospective study in adolescent soccer. International Journal of
needs to be highlighted, especially given that most studies on youth Environmental Research and Public Health, 19(24), Article 16585.
Bell, R. Q. (1968). A reinterpretation of the direction of effects in studies of socialization.
dropout have used cross-sectional designs (Back, Stenling, et al., 2022;
Psychological Review, 75(2), 81–95.
Crane & Temple, 2015; Møllerløkken et al., 2015). Thus, our findings Bentzen, M., Hordvik, M., Stenersen, M. H., & Solstad, B. E. (2021). A longitudinal
offer important insight into the long-term role of intrapersonal con­ transitional perspective on why adolescents choose to quit organized sport in
Norway. Psychology of Sport and Exercise, 56, Article 102015.
straints (e.g., perceiving sport as important/useful) and interpersonal
Bergsgard, N. A., Borodulin, K., Fahlen, J., Høyer-Kruse, J., & Iversen, E. B. (2019).
constraints (e.g., parental involvement in their children’s games and National structures for building and managing sport facilities: A comparative
practices) in youth sports dropout. A final strength of the present study analysis of the nordic countries. Sport in Society, 22(4), 525–539.
concerns the examination of the direction of influence between Boiché, J. C., & Sarrazin, P. G. (2009). Proximal and distal factors associated with
dropout versus maintained participation in organized sport. Journal of Sports Science
perceived parental behaviors and adolescents’ sports-related values. and Medicine, 8(1), 9.
Bremer, K. L. (2012). Parental involvement, pressure, and support in youth sport: A
narrative literature review. Journal of Family Theory & Review, 4(3), 235–248.
7. Conclusions Breuer, C., Feiler, S., & Wicker, P. (2015). Sport clubs in Germany. In Sport clubs in Europe
(pp. 187–208). Springer.
Findings from the present study have important implications for both Bronfenbrenner, U. (2005). Interacting systems in human development. Research
paradigms: Present and future. In U. Bronfenbrenner (Ed.), Making human beings
researchers and practitioners. Specifically, they contribute to the liter­
human: Bioecological perspectives on human development (pp. 67–93). Sage.
ature by offering an improved understanding of how parents’ behaviors Brown, B. A. (1985). Factors influencing the process of withdrawal by female adolescents
in the athletic context influence adolescents’ sports dropout. The results from the role of competitive age group swimmer. Sociology of Sport Journal, 2(2),
111–129.
indicate that adolescents whose parents attend their practices and games
Brown, A. T. (2015). Data issues in CFA: Missing, non-normal, and categorical data. In
or talk with the coach of the team perceive sports activities as fun, Confirmatory factor analysis for applied research (2 ed., pp. 333–379). The Guilford
important and useful. In turn, these adolescents are less likely to dropout Press.
of sports activities one year later. Further, the results of our reciprocal Camacho-Thompson, D. E., & Simpkins, S. D. (2020). Parental involvement in organized
after-school activities and adolescent motivational beliefs. Applied Developmental
analysis add to the literature by demonstrating that the direction of Science, 1–16.
influence may flow from parental involvement to changes in adoles­ Carlman, P., Wagnsson, S., & Patriksson, G. (2013). Causes and consequences of
cents’ perceived sports-related values over time, not the other way dropping out from organized youth sports. Swedish Journal of Sport Research, 2(1),
26–54.
around. Given that dropout from sports increases during adolescence Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement
(Norberg, 2019; Rullestad et al., 2021), researchers and practitioners invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464–504.
interested in prolonging adolescents’ continued involvement in sports Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing
measurement invariance. Structural Equation Modeling, 9(2), 233–255.
might adopt parent-focused strategies to influence parents’ involvement Connell, J. P., & Wellborn, J. G. (1991). Competence, autonomy, and relatedness: A
in their children’s sports activities. For example, professionals in the motivational analysis of self-system processes. In M. R. Gunnar, & A. L. Sroufe (Eds.),
sports domain might inform parents that by attending games or prac­ Self Processes and Development: The Minnesota symposia on child psychology, 23 pp.
43–77). Lawrence Erlbaum Associates.
tices, which is desired by young people (Holt & Knight, 2014b), they can
Côté, J. (1999). The influence of the family in the development of talent in sport. The
play an important role in their children’s sports-related values (e.g., Sport Psychologist, 13(4), 395–417.
sport is fun, important and useful), which have been identified as key Côté, J., Baker, J., & Abernethy, A. (2003). From play to practice: A developmental
framework for the acquisition of expertise in team sports.
correlates of sports dropout (Back, Johnson, et al., 2022; Crane &
Côté, J., & Vierimaa, M. (2014). The developmental model of sport participation: 15
Temple, 2015; Møllerløkken et al., 2015). In sum, the present study years after its first conceptualization. Science & Sports, 29, S63–S69.
demonstrates that the importance of parental behaviors is not limited to Crane, J., & Temple, V. (2015). A systematic review of dropout from organized sport
adolescents’ sports-related values, but also applies to their continued among children and youth. European Physical Education Review, 21(1), 114–131.
Delforterie, M. J., Verweij, K. J., Creemers, H. E., van Lier, P. A., Koot, H. M.,
participation in sports over time. Branje, S. J., Huizink, A. C. J. E., & health. (2016). Parental solicitation, parental
control, child disclosure, and substance use. native and immigrant Dutch adolescents,
21(6), 535–550.
Dorsch, T. E., Wright, E., Eckardt, V. C., Elliott, S., Thrower, S. N., & Knight, C. J. (2021).
Declaration of competing interest
A history of parent involvement in organized youth sport: A scoping review. Sport,
Exercise, and Performance Psychology, 10(4), 536.
This manuscript is not being considered for publication elsewhere. Eccles, J. S. (1983). Expectancies, values, and academic behaviors. In I. J. T. S (Ed.),
The study was conducted in accordance with the ethical standards of the Achievement and achievement motivations (pp. 75–146). San Francisco, CA: Freeman.
Eccles, J. S. (2007). Families, schools, and developing achievement-related motivations
American Psychological Association, and we do not have any interests and engagement. In Handbook of socialization:theory and research (pp. 665–691). The
that may influence the research. Guilford Press.

8
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

Eccles, J. S., & Wigfield, A. (1995). In the mind of the actor: The structure of adolescents’ Liu, Y., Millsap, R. E., West, S. G., Tein, J.-Y., Tanaka, R., & Grimm, K. J. (2017). Testing
achievement task values and expectancy-related beliefs. Personality and Social measurement invariance in longitudinal data with ordered-categorical measures.
Psychology Bulletin, 21(3), 215–225. Psychological Methods, 22(3), 486.
Eccles, J. S., & Wigfield, A. (2020). From expectancy-value theory to situated expectancy- MacKinnon, D. P. (2008). Single mediator model. In Introduction to statistical mediation
value theory: A developmental, social cognitive, and sociocultural perspective on analysis (pp. 47–78). Taylor & Francis.
motivation. Contemporary Educational Psychology, 61, Article 101859. MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence limits for the
Espedalen, L. E., & Seippel, Ø. (2022). Dropout and social inequality: Young people’s indirect effect: Distribution of the product and resampling methods. Multivariate
reasons for leaving organized sports. Annals of Leisure Research, 1–18. Behavioral Research, 39(1), 99–128.
Eurobarometer. (2019). Sport and physical activity. https://ec.europa.eu/sport/. htt Mahoney, J. L., Larson, R. W., & Eccles, J. S. (2005). Organized activities as
ps://op.europa.eu/en/publication-detail/-/publication/9a69f642-fcf6-11e8-a96d- developmental contexts for children and adolescents. In J. L. Mahoney, R. W. Larson,
01aa75ed71a1/language-en/format-PDF/source-82747397. & J. S. Eccles (Eds.), Organized activities as contexts of development: Extracurricular
Fahlén, J., & Stenling, C. (2016). Sport policy in Sweden. International journal of sport activities, after-school and community programs. Psychology Press.
policy and politics, 8(3), 515–531. Møllerløkken, N. E., Lorås, H., & Pedersen, A. V. (2015). A systematic review and meta-
Fernandez, A., Loukas, A., & Pasch, K. E. (2018). Examining the bidirectional analysis of dropout rates in youth soccer. Perceptual and Motor Skills, 121(3),
associations between adolescents’ disclosure, parents’ solicitation, and adjustment 913–922.
problems among non-hispanic white and hispanic early adolescents. Journal of Youth Norberg, J. R. (2019). Statens stöd till idrotten – uppföljning 2019 C. f. i.-. CIF.
and Adolescence, 47(12), 2569–2583. Nuzzo, R. (2014). Statistical errors. Nature, 506(7487), 150.
Flora, D. B., & Curran, P. J. (2004). An empirical evaluation of alternative methods of O’Neil, L., & Amorose, A. J. (2021). Autonomy support and control within mother-father
estimation for confirmatory factor analysis with ordinal data. Psychological Methods, parenting structures: A person-centered approach in youth sport. Psychology of Sport
9(4), 466. and Exercise, 53, Article 101854.
Fraser-Thomas, J., Cote, J., & Deakin, J. (2008). Examining adolescent sport dropout and Østerlie, O., Løhre, A., & Haugan, G. (2019). The expectancy-value questionnaire in
prolonged engagement from a developmental perspective. Journal of Applied Sport physical education: A validation study among Norwegian adolescents. Scandinavian
Psychology, 20(3), 318–333. Journal of Educational Research, 63(6), 869–883.
Fraser-Thomas, J., Côté, J., & Deakin, J. (2008). Understanding dropout and prolonged Peguero, A. A. (2011). Immigrant youth involvement in school-based extracurricular
engagement in adolescent competitive sport. Psychology of Sport and Exercise, 9(5), activities. The Journal of Educational Research, 104(1), 19–27.
645–662. Rhemtulla, M., Brosseau-Liard, P.É., & Savalei, V. (2012). When can categorical variables
Fredricks, J. A., & Eccles, J. S. (2004). Parental influences on youth involvement in be treated as continuous? A comparison of robust continuous and categorical SEM
sports. In M. R. Weiss (Ed.), Developmental sport and exercise psychology: A lifespan estimation methods under suboptimal conditions. Psychological Methods, 17(3), 354.
perspective (pp. 145–164). Fitness Information Technology. Riksidrottsförbundet. (2019). Idrottsrörelsen i siffror, 10.
Fredricks, J. A., & Eccles, J. S. (2005). Family socialization, gender, and sport motivation Rullestad, A., Meland, E., & Mildestvedt, T. (2021). Factors predicting physical activity
and involvement. Journal of Sport & Exercise Psychology, 27(1), 3–31. and sports participation in adolescence. Journal of Environmental and Public Health,
Fredricks, J. A., Simpkins, S. D., & Eccles, J. S. (2005). Family socialization, gender, and 2021.
participation in sports and instrumental music. In Developmental pathways through Rutkowski, L., Svetina, D., & Liaw, Y.-L. (2019). Collapsing categorical variables and
middle childhood: Rethinking contexts and diversity as resources (pp. 41–62). Erlbaum. measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 26
Geiser, C. (2012). Data analysis with Mplus. Guilford press. (5), 790–802.
Geiser, C. (2021). Longitudinal structural equation modeling with Mplus: A latent state–trait Sagar, S. S., & Lavallee, D. (2010). The developmental origins of fear of failure in
perspective. adolescent athletes: Examining parental practices. Psychology of Sport and Exercise,
Gould, D., & Petlichkoff, L. (1988). Participation motivation and attrition in young 11(3), 177–187.
athletes. Children in sport, 3, 161–178. Salguero, A., Gonzalez-Boto, R., Tuero, C., & Marquez, S. (2003). Identification of
Grusec, J. E., & Goodnow, J. J. (1994). Impact of parental discipline methods on the dropout reasons in young competitive swimmers. The Journal of Sports Medicine and
child’s internalization of values: A reconceptualization of current points of view. Physical Fitness, 43(4), 530.
Developmental Psychology, 30(1), 4–19. Sameroff, A. J. (1975a). Early influences on development: Fact or fancy? Merrill-Palmer
Guillet, E., Sarrazin, P. G., Fontayne, P., & Brustad, R. J. (2006). Understanding female Quarterly of Behavior and Development, 21(4), 267–294.
sport attrition in a stereotypical male sport within the framework of Eccles’s Sameroff, A. J. (1975b). Transactional models in early social relations. Human
Expectancy–Value model. Psychology of Women Quarterly, 30(4), 358–368. Development, 18(1–2), 65–79.
Hamaker, E. L., Kuiper, R. M., & Grasman, R. P. (2015). A critique of the cross-lagged Savalei, V. (2011). What to do about zero frequency cells when estimating polychoric
panel model. Psychological Methods, 20(1), 102. correlations. Structural Equation Modeling, 18(2), 253–273.
Hermida, R. (2015). The problem of allowing correlated errors in structural equation Selig, J. P., & Little, T. D. (2012). Autoregressive and cross-lagged panel analysis for
modeling: Concerns and considerations. Computational Methods in Social Sciences, 3 longitudinal data. In B. Laursen, T. D. Little, & N. A. Card (Eds.), Handbook of
(1), 5. developmental research methods (pp. 265–278).
Holt, N. L. (2007). Positive youth development through sport. Routledge. Simpkins, S. D., Fredricks, J. A., & Eccles, J. S. (2012). Charting the Eccles’ expectancy-
Holt, N. L., & Knight, C. J. (2014a). Overview of issues associated with parenting in youth value model from mothers’ beliefs in childhood to youths’ activities in adolescence.
sport. In Parenting in youth sport: From research to practice (pp. 5–15). Routledge. Developmental Psychology, 48(4), 1019–1032.
Holt, N. L., & Knight, C. J. (2014b). Parental behaviors at competitions in Parenting in youth Simpkins, S. D., Fredricks, J., & Eccles, J. S. (2015a). Families, schools, and developing
sport: From research to practice (pp. 55–68). Routledge. achievement-related motivations and engagement. In J. E. Grusec, & P. D. Hastings
Holt, N. L., Tamminen, K. A., Black, D. E., Mandigo, J. L., & Fox, K. R. (2009). Youth (Eds.), Handbook of socialization: Theory and research (2 ed., pp. 614–636). Guilford
sport parenting styles and practices. Journal of Sport & Exercise Psychology, 31(1), Publications.
37–59. Simpkins, S. D., Fredricks, J., & Eccles, J. S. (2015b). In P. J. Bauer (Ed.), The role of
Horn, T. S., & Horn, J. L. (2012). Family influences on children’s sport and physical parents in the ontogeny of achievement-related motivation and behavioral choices, 80. The
activity participation, behavior, and psychosocial responses. In G. Tenenbaum, & Society for Research in Child Development. https://doi.org/10.1111/mono.12165.
R. C. Eklund (Eds.), Handbook of sport psychology (3 ed., pp. 685–711). Singh, G. K., Stella, M. Y., Siahpush, M., & Kogan, M. D. (2008). High levels of physical
Hoyle, R. H., & Leff, S. S. (1997). The role of parental involvement in youth sport inactivity and sedentary behaviors among US immigrant children and adolescents.
participation and performance. Adolescence, 32(125), 233. Archives of Pediatrics and Adolescent Medicine, 162(8), 756–763.
Jacobs, J. E., & Eccles, J. S. (2000). Parents, task values, and real-life achievement- Smetana, J. G., Robinson, J., & Rote, W. M. (2015). Socialization in adolescence. J. H. o.
related choices. In Intrinsic and extrinsic motivation (pp. 405–439). Elsevier. s. T., & research, 60–84.
Jaf, D., Özdemir, M., & Bayram Özdemir, S. (2020). The role of parents’ disapproval of Stuart, M. (2003). Sources of subjective task value in sport: An examination of
peers and monitoring on immigrant and native youth’s participation in organized adolescents with high or low value for sport. Journal of Applied Sport Psychology, 15
sports in Sweden: The mediating role of engagement in delinquent activities. The (3), 239–255.
Journal of Early Adolescence, 41(3), 437–471. https://doi.org/10.1177/ Svetina, D., Rutkowski, L., & Rutkowski, D. (2020). Multiple-group invariance with
0272431620919159 categorical outcomes using updated guidelines: An illustration using M plus and the
Jaf, D., Özdemir, M., & Skoog, T. (2021). Immigrant and Swedish adolescents’ lavaan/semtools packages. Structural Equation Modeling: A Multidisciplinary Journal,
involvement in organized sports activities: An expectancy-value approach. BMC 27(1), 111–130.
Psychology, 9(12), 1–15. https://doi.org/10.1186/s40359-021-00522-9 Toselli, S., & Belcastro, M. G. (2017). Participation in sports, body composition, and
Knight, C. J. (2019). Revealing findings in youth sport parenting research. Kinesiology fitness characteristics in children according to ethnic background. Scandinavian
Review, 8(3), 252–259. Journal of Medicine & Science in Sports, 27(12), 1913–1926.
Lee, M., & MacLean, S. (1997). Sources of parental pressure among age group swimmers. Wagnsson, S., & Patriksson, G. (2005). Föräldrarnas engagemang i barnens
European Journal of Physical Education, 2(2), 167–177. idrottsföreningar–myt eller verklighet. Svensk idrottsforskning(2).
Leff, S. S., & Hoyle, R. H. (1995). Young athletes’ perceptions of parental support and Wang, J., & Wang, X. (2020). Structural equation modeling: Applications using Mplus (2
pressure. Journal of Youth and Adolescence, 24(2), 187–203. ed.). John Wiley & Sons.
Lerner, R. M., Lerner, J. V., & Benson, J. B. (2011). Positive youth development: Research Wasserstein, R. L., & Lazar, N. A. (2016). The ASA statement on p-values: Context, process,
and applications for promoting thriving in adolescence. Advances in Child and purpose, 70 pp. 129–133). Taylor & Francis.
Development and Behavior: Positive youth development, 41, 1–17. Academic Press.
Little, T. D. (2013a). Longitudinal structural equation modeling. Guilford press.
Little, T. D. (2013b). Mediation and moderation. In Longitudinal structural equation
modeling (pp. 286–324). Guilford Press.

9
D. Jaf et al. Psychology of Sport & Exercise 68 (2023) 102448

Weiss, M. R., & Hayashi, C. T. (1995). All in the family: Parent-child influences in Wigfield, A., & Eccles, J. S. (1992). The development of achievement task values: A
competitive youth gymnastics. Pediatric Exercise Science, 7(1), 36–48. theoretical analysis. Developmental Review, 12(3), 265–310.
Welk, G. J., Wood, K., & Morss, G. J. P.e. s. (2003). Parental influences on physical
activity in children. An exploration of potential mechanisms, 15(1), 19–33.

10