De Meester et al.

International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48

http://www.ijbnpa.org/content/11/1/48

RESEARCH Open Access

Extracurricular school-based sports as a motivating

vehicle for sports participation in youth:
a cross-sectional study
An De Meester1,2*, Nathalie Aelterman3,4, Greet Cardon1, Ilse De Bourdeaudhuij1 and Leen Haerens1

Abstract
Background: Extracurricular school-based sports are considered to be an ideal means of reaching children who are
not active in community sports. The purposes of this study were to examine the extent to which pupils not engaging
in community sports do participate in extracurricular school-based sports, and to assess whether extracurricular
school-based sports participants are more physically active and/or more autonomously motivated towards sports in
daily life than children who do not participate in extracurricular school-based sports.
Methods: One thousand forty-nine children (53.7% boys; M age = 11.02 years, SD = 0.02) out of 60 classes from 30
Flemish elementary schools, with an extracurricular school-based sports offer, completed validated questionnaires to
assess physical activity (Flemish Physical Activity Questionnaire) and motivation (Behavioral Regulations in Physical
Education Questionnaire). Multilevel regression analyses were conducted to examine the data generated from these
questionnaires.
Results: More than three quarters of the children (76%) reported participating in extracurricular school-based sports
during the current school year and 73% reported engaging in organized community sports. Almost two third of the
children (65%) not participating in community sports stated that they did participate in extracurricular school-based
sports. Extracurricular school-based sports participants were significantly more physically active than children not
participating in extracurricular school-based sports (β = 157.62, p < 0.001). Significant three-way interactions
(sex × extracurricular school-based sports participation × community sports participation) were found for
autonomous motivation, with boys engaging in extracurricular school-based sports but not in community sports
being significantly more autonomously motivated towards sports than boys not engaging in community or
extracurricular school-based sports (β = 0.58, p = 0.003). Such differences were not noted among girls.
Conclusions: If extracurricular school-based sports are offered at school, the vast majority of elementary school
children participate. Although extracurricular school-based sports attract many children already engaging in community
sports, they also reach almost two third of the children who do not participate in community sports but who might
also be optimally motivated towards sports. As children participating in extracurricular school-based sports are more
physically active than children who do not participate, extracurricular school-based sports participation can be
considered to contribute to an active lifestyle for these participating children.
Keywords: Extracurricular school-based sports, Community sports, Physical activity, Motivation, Self-determination theory

* Correspondence: [email protected]
1
Department of Movement and Sports Sciences, Ghent University,
Watersportlaan 2, 9000 Ghent, Belgium
2
The Policy Research Centre on Sports, funded by the Flemish Government,
Brussels, Belgium
Full list of author information is available at the end of the article

© 2014 De Meester et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly credited.
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 2 of 15
http://www.ijbnpa.org/content/11/1/48

Background to organized community sports which makes participation

Research suggests that a considerable amount of children not limited to sports-talented children [16]. Hence, in such
and adolescents drop out from sports when they grow a context, extracurricular school-based sports programs
older [1,2]. Consequently, by the end of elementary school are typically put forward as programs that reach those chil-
a considerable amount of children does not meet the dren who are not participating in community sports [17].
health-related recommendations of sixty minutes or more This is different to the situation in the US, where extracur-
moderate to vigorous physical activity (MVPA) per day ricular school-based sports can sometimes be selective and
[3]. Sports are considered one specific form of PA that highly competitive in nature [18]. Despite these positive
contributes to children’s and adolescents’ overall engage- inclusive features of extracurricular school-based sports, lit-
ment in MVPA [4]. When compared to other forms of tle research has been conducted to explore which children
physical activities, sports participation typically involves participate in extracurricular school-based sports programs,
physical exertion and skill development, and the competi- and whether such programs indeed reach children who are
tion of individuals or teams against one another [5]. In order not active in community-based sports. Hence, the first aim
for more children to meet the health-recommendations of the present study was to investigate how many children
regarding PA, the need arises to develop and implement not engaging in community sports do participate in extra-
promotion strategies for an active lifestyle, with the promo- curricular school-based sports programs.
tion of sports participation among children and adoles- There is already some evidence suggesting that the
cents as one possible strategy. provision of PA organized at school is effective to increase
Schools are considered to be ideal settings for implement- children’s activity levels, with children being more physic-
ing such promotion strategies since children of all socioeco- ally active if there is an extensive PA-offer at school
nomic backgrounds spend considerable amounts of time at [19,20]. According to Powers et al. [14], pupils’ PA-levels
school. Furthermore, while required educational training at school could be improved by increasing the participa-
degrees differ between countries, in most countries teachers tion, duration, and frequency of existing programs and by
are required to have a teacher education degree and most implementing programs before school and at lunchtime.
school are equipped with the necessary facilities to provide The potential of such programs for the promotion of PA
PA opportunities [6-8]. Although Physical Education (PE) is among adolescents is also acknowledged by other authors
acknowledged to be an important intra-curricular medium [15,21,22]. However, in most of these studies, it is not clear
within the school context for the promotion of sports whether certain groups of pupils, characterized by com-
participation among youth [9,10], PE hours in most school mon demographic (sex, age, SES), psychosocial (e.g., mo-
curricula are limited, and in many countries not mandatory tivation, social support) or behavioral (e.g., participation in
[11]. Furthermore, within the limited curricular time, phys- community sports) characteristics are more or less repre-
ical education’s diverse learning objectives are not exclu- sented in extracurricular school-based sports, and whether
sively related to promoting PA and sports participation. pupils participating in extracurricular school-based sports
Within the school-community context possible additional are overall more or less active than their non-participating
sources for PA opportunities, including sports participation, counterparts. Pupils who participate in extracurricular
thus need to be explored [12]. school-based sports are physically active during the time
Due to considerable differences between different educa- spent in extracurricular school-based sports but it is not
tional systems, various terms are used interchangeably within clear whether they are more or less physically active outside
the literature to describe possibilities for school-based sports these extracurricular school-based sports. Consequently, a
activities outside the PE curriculum (extracurricular school- second purpose of the present study was to investigate
based sports). Most authors apply the term extracurricular whether the group of pupils participating in extracurricular
school-based sports when referring to the provision of activ- school-based sports differs from other groups (not partici-
ities at school outside the formal PE curriculum, most often pating in any organized sports or participating in commu-
during after-school hours and lunch breaks [13-15]. In the nity sports) in terms of overall activity levels.
current paper, extracurricular school-based sports are de- Besides differing in PA-levels, children participating in
fined in a similar way: ‘all sports activities not included in extracurricular school-based sports might also differ from
the curriculum but organized by the school either during those not participating in extracurricular school-based sports
lunch break, during after-school hours or on Wednesday- in their motivation towards sports. Specifically, some pupils
afternoon (typically a free afternoon in Flemish elementary might be optimally motivated toward sports, but not get the
schools) in which pupils can voluntarily participate’. chances (e.g. no parental support) to participate in sports
Extracurricular school-based sports are considered to outside the school context [23]. For this group of pupils,
be easily accessible in terms of location and costs. They participation in extracurricular school-based sports might
usually involve participation together with class- or school- provide the only extracurricular opportunity to be active.
mates and are generally less competitive when compared Hence, a third aim of the present study was to investigate
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 3 of 15
http://www.ijbnpa.org/content/11/1/48

differences in motivation towards sports between children participate in the present cross-sectional study. The schools
participating in extracurricular school-based sports and were randomly selected from a list of all schools located in
children not engaging in any form of organized sports. two Flemish provinces (West Flanders and East Flanders)
In the present study, the concept of motivation was that participate in an extracurricular school-based sports
approached from the perspective of the Self-Determination project developed by the Flemish School-sports Foundation
Theory (SDT) [24]. This macro-theory of motivation pro- and the Flemish Commissariat for Promotion of Physical
vides an understanding of why people initiate and persist in Development, Sports and Outdoor Recreation. This project
behaviors [25] and distinguishes between autonomous mo- provides logistical support to the participating schools who, in
tivation, controlled motivation and amotivation for sports. return, organize at least one weekly hour of extracurricular
Autonomous motivation involves the regulation of behavior school-based sports during after-school hours on a weekday
with the experiences of volition, psychological freedom, and under the guidance of a PE teacher. At the moment of data
reflective self-endorsement and is considered the most opti- collection, one school no longer participated in the extracur-
mal form of motivation [26]. The second type of motivation ricular school-based sports project and it therefore did not
is controlled motivation, which refers to the pressured en- meet the inclusion criteria. In addition, four principals of the
gagement in an activity. Autonomous motivation and con- other contacted schools refused to participate. This resulted in
trolled motivation are contrasted with amotivation, which a final sample of 30 schools, of which eleven (36.67%) were
exists when people lack intentionality or engage in behav- located in an urban area, fourteen in a suburban area (46.67%)
iors for unknown reasons [24]. For example, an amotivated and five in a rural area (16.67%). Two classes in each school
pupil claims to have no idea why he should bother partici- were randomly selected to participate in the present study
pating in extracurricular school-based sports. Hagger et al. (one class in grade 5 and one class in grade 6).
[27] examined the relation between motivation for PE and The children in the selected classes and their parents
motivation for sports in leisure time, outside the school. received an information letter and with the exception of 33
However, the relationship between extracurricular school- (response rate = 96.95%), all parents gave informed consent
based sports participation and motivation towards sports for their child to participate. In total, 1049 children (563
has, to our knowledge, not yet been investigated. If we find boys; 53.67%) with a mean age of 11.02 years (SD = 0.02,
that extracurricular school-based sports programs mainly range 9-13 years) filled out a set of questionnaires during a
attract those children who are already autonomously moti- randomly chosen lesson. All the questions were read out
vated towards sports, and furthermore there remains a large loud by the first author and clarified when necessary.
group of children that is not optimally motivated and is also In the present study extracurricular school-based sports
not reached, the question arises how the promoting role of were defined as ‘all sports not included in the curriculum
extracurricular school-based sports can be optimized so that but organized by the school either during lunch break,
groups with less optimal motivational profiles are reached. during after-school hours or on Wednesday-afternoon, in
In summary, the primary aim of this study was to investi- which pupils can voluntarily participate’. In Flanders,
gate how many pupils not engaging in community sports do (Belgium) elementary school-hours are traditionally orga-
participate in extracurricular school-based sports. The sec- nized between 8.30am and 12pm and between 1pm and
ond aim was to assess if extracurricular school-based sports 4pm. On Wednesday, the lessons end at 12pm and the
participants were more physically active in daily life than pupils are free in the afternoon. Many schools therefore
children who do not participate in extracurricular school- organize activities (including extracurricular school-based
based sports, while controlling for their participation in sports) on Wednesday-afternoon in which pupils can
community sports. The last aim was to evaluate whether participate on a voluntary basis. In Flanders, extracurricular
children who participate in extracurricular school-based school-based activities can be organized at school or in the
sports, compared to those who do not, are better (i.e., more schools’ neighborhood. Extracurricular school-based sports
autonomously) or relatively worse (i.e., more controlled) delivered in the school’s neighborhood are not substantively
motivated or amotivated to participate in sports in general. different from those delivered at school, they merely take
As the transition from childhood to adolescence is known place in a different location (e.g., adjacent grass fields or
to be a risk-period for decreased PA and sports participation sports halls off school campus). Extracurricular school-
[28,29], the target group of the current study consisted of based sports activities are also sometimes organized for
children of the last two grades of elementary school. groups of schools, rather than for one single school, for
which then a central location (yet still close to each of
Methods the participating schools) is often chosen. Extracurricular
Participants and procedure school-based sports during the weekend were not in-
Principals of 35 elementary schools in Flanders (the cluded in our definition because Flemish extracurricular
Dutch-speaking northern part of Belgium) with an extra- school-based sports are rarely or never organized during
curricular school-based sports offer were contacted to the weekend.
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 4 of 15
http://www.ijbnpa.org/content/11/1/48

The study protocol was approved by the Ethical within the ongoing school year or if they had registered
Committee of Ghent University. for upcoming extracurricular school-based sports activities
within the ongoing school year. Pupils were then ques-
Measures tioned about the frequency of participation in extracurricu-
Demographic variables lar school-based sports within the current school year:
Demographic variables such as sex, date of birth and resi- “How many hours do you spend on extracurricular school-
dence were obtained through a number of questions at the based sports (a) during lunch break, (b) during after-school
beginning of the questionnaire. Pupils were also asked to fill hours, (c) on Wednesday-afternoon?”. Multiple-choice an-
out their parents’ occupations. The Four Factor Index of swers to each of the three sub-questions varied from less
Social Status [30] was then used to determine the pupils’ than 1 hour per month to more than 4 hours per week,
socioeconomic status (SES). This index determines a per- with seven sub-categories in between. Children’s weekly
son’s SES, based on three parental factors: marital status, time spent in extracurricular school-based sports was cal-
educational factor and occupational factor. A fourth factor culated based on the answers on the three above mentioned
applies only to retired persons. Given the young age of the sub-questions. An additional question was added about the
participants in the current study, we focused on the occu- specific sports they personally participated in (“Which of
pational factor of both parents. Parents’ occupations were the following extracurricular school-based sports did you
given a score from 1 to 9 and accordingly categorized in participate in during the current school year?”). Pupils
two socio-economic groups: low and medium SES (54.72%) could, similar to the question about the extracurricular
on the one hand and high SES (45.28%) on the other hand. school-based sports offer, indicate multiple answers in a list
of 50 sports or fill out a sport that was not included in the
Self-reported physical activity list. Finally, to broaden the window to multiple school
PA-levels were determined using the Flemish Physical years, pupils responded to the question “How long have
Activity Questionnaire (FPAQ) [31]. The pen and paper you been participating in extracurricular school-based
version of the FPAQ has been shown to be a reliable (test- sports?” via a multiple-choice answer, ranging from 1 (this
retest reliability coefficients ranging from 0.69 to 0.93) and is the first school year) to 3 (more than 2 school years).
valid (concurrent validity: r = 0.27-0.44 with respect to Next to the questioning on extracurricular school-based
accelerometer data) instrument to assess different dimen- sports, three categories of physical activities were mea-
sions of usual PA and sedentary behavior in children [32]. sured using validated questions [34]: a) Intra-curricular
To determine the validity of the questionnaire in the PA (e.g. “How many hours of PE do you participate in per
current study, we calculated correlations between self- week?”), b) walking and cycling (e.g. “How much time do
reported PA levels and accelerometer-measured activities you usually spend cycling on a weekday?”, not distinguish-
in a convenience sample of 61 children who filled out the ing between walking and cycling for active transportation
FPAQ. Similar to previous studies, significant positive and leisure time), and c) non-school based sports partici-
correlations (r = 0.311; p = 0.045) were found between PA pation (both organized as well as unorganized sports).
levels and accelerometer-measured activities. An add- Sports participation was assessed by asking the partici-
itional file contains a detailed methodology of the validity pants to indicate the main sport(s) they engaged in (with a
study in our subsample (see Additional file 1). For the maximum of 3 sports). For each sport, children had to
present study a distinction was made between the assess- report the frequency, the usual time spent on that activity
ment of extracurricular school-based sports and other and whether or not they practiced their sport(s) in a sports
types of physical activity. Questions with regard to extra- club (that is to make the distinction between organized
curricular school-based sports were similar as in the study community sports versus unorganized sports) [35]. Sports
of Haerens et al. [33]. In a first question, the extracurricu- participation did not include extracurricular school-based
lar school-based sports offer at school was addressed sports, which was clarified to the pupils by means of an
(“Which of the following extracurricular school-based oral explanation by the researcher and the written instruc-
sports does your school offer during the current school tions in the questionnaire. Children’s overall weekly PA was
year?”). Pupils could indicate multiple answers in a list of calculated by making the sum of the time (in minutes) spent
50 sports or fill out a sport that was not included on the list in extracurricular school-based sports, intra-curricular PA-
in the space provided. Pupils were then asked about their time, time spent on walking and cycling, and time spent in
personal extracurricular school-based sports participation, (non-school based) sports participation.
and were asked to indicate ‘yes’ or ‘no’ as an answer to the
question “Did you participate in extracurricular school- Motivation towards sports
based sports in the current school year?”. Pupils were An adapted version of the Dutch Behavioral Regulations
instructed to indicate ‘yes’ to this question if they had pre- in Physical Education Questionnaire (BRPEQ) [36] was
viously participated in extracurricular school-based sports used to assess motivation towards sports. The BRPEQ
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 5 of 15
http://www.ijbnpa.org/content/11/1/48

was validated in previous research in the context of PE (version 2.26) was used to conduct the multilevel regres-
[36] and is an adapted version of the BREQ-2 [37]. To ob- sion analyses. Multilevel modeling does not require stron-
tain an optimal balance in controlled motivation items, ger assumptions than ordinary modeling with regard to
the BRPEQ contains 6 additional items compared to the normality of the data [39]. Multilevel models therefore
BREQ-2. Two other adaptations were made: one item was assume that the distribution at each level comes from a
reformulated in order to avoid a double negative and the Gaussian distribution. A technique for graphically asses-
word ‘personally’ was added in another item because SDT sing this assumption for each residual is to plot a normal
emphasizes personal endorsement rather than the more score plot [40]. The normal score plots showed that, for
general importance of engagement in activities [36]. Simi- eleven out of twelve conditions (4 variables at 3 levels
lar to the original BREQ-2, the sports context (all forms of respectively), the data were normally distributed.
sports in leisure time, whether or not organized within the First, a three-level null model (school, class, pupil) or
community) was used to frame the items of the BRPEQ intercept-only model including only the dependent variable
used in the current study. was estimated for PA (null model 1), autonomous motiv-
The questionnaire consists of twenty items. Sixteen items ation (null model 2), controlled motivation (null model 3),
with the stem ‘I participate in sports because…’ were used and amotivation (null model 4), respectively. These null
to measure autonomous motivation (8 items, α = 0.87; e.g., models partitioned the total variance of the examined
“I participate in sports because I find it a pleasurable activ- dependent variable into the between-pupil (Level 1),
ity.”) and controlled motivation (8 items, α = 0.70; e.g., “I between-class (Level 2) and between-school (Level 3) vari-
participate in sports because I feel guilty if I don’t.”). Amoti- ance and further served as a baseline with which explana-
vation was measured by four items (α = 0.77; e.g., “I don’t tory models were compared.
see why I should bother participating in sports.”). Partici- In a second step, age and SES were inserted in the
pants responded to each of the twenty items via a 5-point models as covariates. Furthermore, all three types of mo-
Likert scale from 1 (not at all true for me) to 5 (very true tivation were entered as additional covariates in the model
for me). for PA (model 1a). Controlled motivation and amotivation
were entered as additional covariates in the model with
Analyses autonomous motivation as a dependent variable (model
SPSS (version 19.0) was used to describe the characteris- 2a). Similar strategies were followed for the two other
tics of the sample, to describe the percentage of pupils motivation-models [autonomous motivation and amotiva-
participating in extracurricular school-based sports that tion as covariates in the model for controlled motivation
either did or did not engage in community sports and to (model 3a); autonomous and controlled motivation as
analyze the sports offer in the extracurricular school- covariates in the model for amotivation (model 4a)]. To
based sports programs. Pearson’s chi-squared tests were compare these models with their respective intercept-only
then used to determine the relationships between sex, models, likelihood ratio tests were conducted. If a likeli-
extracurricular school-based sports participation and hood ratio test was statistically significant, the model with
community sports participation. The hierarchical struc- covariates was considered to be a better fit than the
ture of the data, with 1049 children being nested within intercept-only model [41].
60 classes nested within 30 different schools, and the ad- In a third step, sex, extracurricular school-based
equate sample size for conducting multilevel analyses sports participation and participation in community
[38], allowed us to employ multilevel regression analyses sports were included as predictors in each of the models
to examine whether children participating in extracurric- (model 1b, model 2b, model 3b and model 4b). These
ular school-based sports are (1) more physically active in predictors were entered in the models as dichotomous
daily life and (2) better motivated to participate in sports variables (sex: 0 = boy, 1 = girl; extracurricular school-
than children who do not participate in extracurricular based sports participation: 0 = no extracurricular school-
school-based sports. The total weekly time engaging in based sports participation in the current school year, 1 =
PA (in minutes) was calculated based on the reported extracurricular school-based sports participation in the
time spent in leisure time sports including both commu- current school year; participation in community sports:
nity sports and unorganized sports, intra-curricular PA 0 = no participation in community sports in the current
(PE and other weekly mandatory sports and PA), extra- school year, 1 = participation in community sports in the
curricular school-based sports, and walking and cycling current school year). In the same model, we also explored
(not distinguishing between walking and cycling for the two-way interaction effects between each of these
active transportation or leisure time). Since all necessary three predictors (sex × extracurricular school-based sports
data regarding each of the variables used in the analyses participation; sex × community sports participation; and
were obtained from all the participants, there were no extracurricular school-based sports participation × com-
missing data. The IGLS estimation method in MLwiN munity sports participation) and the three-way interaction
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 6 of 15
http://www.ijbnpa.org/content/11/1/48

effect (sex × extracurricular school-based sports participa- of the boys (n = 466) and 68.11% of the girls (n = 331).
tion × community sports participation). To interpret sig- More than one third of the boys (36.70%) and one quarter
nificant interactions, the regression equations were of the girls (25.08%) spent more than one hour per week
repeated several times by changing the reference category on extracurricular school-based sports, while 43.35% of
to obtain coefficients for all combinations of the used the boys and 43.81% of the girls engaged in extracurricular
variables (sex, extracurricular school-based sports partici- school-based sports for 10 to 60 minutes weekly, the
pation and community sports participation). Here again remaining 19.96% of the boys and 31.12% of the girls spent
likelihood ratio tests were conducted to determine if the less than 10 minutes per week in extracurricular school-
models with predictors were better fits than the models based sports. Participation in community sports did not
only including the covariates. significantly differ by sex [χ2(1) = 0.94, p = 0.33]. Almost
In a final step, only statistically significant predictors three-quarters of the boys (74.07%) and the girls (71.40%)
ameliorating each of the models were retained, resulting reported participating in community sports (Table 3) with
in the most parsimonious and interpretable models. The 29.45% of the participants spending less than two hours
most parsimonious models for PA, controlled motivation per week on community sports, 43.68% between two and
and amotivation therefore only included the covariates six hours per week and 26.75% more than six hours per
and the three main effects as predictor (model 1c, model week. The group of children neither participating in extra-
3c and model 4c). Since all the interaction-effects were curricular school-based sports nor community sports
significant in the model for autonomous motivation, all consisted of 40.00% boys and 60.00% girls.
predictors were retained in the final model (model 2b).
The level of significance was for all statistical analyses Sports offer in extracurricular school-based sports
defined as lower than 0.05. In Flanders, as noted earlier, extracurricular school-
based sports are organized directly after school hours,
Results during lunch breaks and on Wednesday-afternoon.
Descriptives for extracurricular school-based sports and Overall, soccer and swimming were the most frequently
community sports participation offered sports (in 90% of the schools), followed by athlet-
Overall, the majority of the children (75.98%) reported ics (86.7%), netball (83.3%) and hockey (76.7%). Other
participating in extracurricular school-based sports during popular sports were badminton (66.7%), dance, tennis,
the current school year. Almost one third of the children softball (46.7%) and basketball (43.3%). Some sports were
(31.87%) spent more than one hour per week on extracur- only offered by a minority of the schools: rugby, volleyball,
ricular school-based sports, while 43.54% engaged in squash (33.3%), table tennis, handball, rope skipping,
extracurricular school-based sports for 10 to 60 minutes skating (30%), ring stick (26.7%), cycling (23.3%), korfball
weekly, the remaining 25.00% of the children spent less (20%), self-defense and martial arts (16.7%), wall climbing
than 10 minutes per week in extracurricular school-based (6.7%), gymnastics and circus techniques (3.3%).
sports. A significantly positive relation was found between During after school hours, the content of the extracurric-
extracurricular school-based sports participation and com- ular school-based sports was completely determined by the
munity sports participation [χ2(1) = 26.25, p < 0.001]. More supervising teacher and typically consisted of popular
than three-quarters (76.79%) of the extracurricular school- sports such as soccer, hockey and badminton, combined
based sports participants and more than half (60.32%) of with an initial warm-up game (e.g. dodgeball, tag and ball
the non-participants were member of at least one sports tag). During lunch breaks, ball sports and dance were the
club. Only a small group of children, namely 9.53% (n = most frequently offered activities, while on Wednesday
100) of the total sample (N = 1049), participated neither in afternoon the extracurricular school-based sports offer
extracurricular school-based sports nor in community mostly consisted of once-a-year-offered sports activities like
sports, while 17.64% of the children reported participating swimming, athletics (most often running races) or skating.
in extracurricular school-based sports at least once a year
but not in community sports (Table 1). Self-reported physical activity according to participation
Similarly, the majority of the children (72.83%) reported in extracurricular school-based sports
participating in community sports, and of this group To examine if children participating in extracurricular
80.10% participated in extracurricular school-based sports school-based sports and/or community sports are more or
also. Of the children not engaging in any community less physically active than their non-participating peers,
sports, 64.91% (n = 185) reported participating in extracur- a three-level model was estimated for weekly-based PA.
ricular school-based sports in the current school year. Variance at school-level was not significantly different
Participation in extracurricular school-based sports from zero [3.1%, χ2(1) = 1.20, p = 0.273]. The random part
significantly differed by sex [χ2(1) = 33.73, p < 0.001]. As of the null model revealed that the variances at class-level
shown in Table 2, participation was reported by 82.77% [6.5%, χ2(1) = 4.38, p = 0.036] and at pupil-level [90.4%, χ2
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 7 of 15
http://www.ijbnpa.org/content/11/1/48

Table 1 Cross tabulation extracurricular school-based sports and community sports

No extracurricular school-based Extracurricular school-based Total
sports participation sports participation
No community sports participation N 100 185 285
% within community sports 35.09% 64.91% 100.00%
% within extracurricular 39.68 % 23.21% 27.17%
school-based sports
% of total 9.53% 17.64% 27.17%
Community sports participation N 152 612 764
% within community sports 19.90% 80.10% 100.00%
% within extracurricular 60.32% 76.79% 72.83%
school-based sports
% of total 14.49% 58.34% 72.83%
Total N 252 797 1049
% within community sports 24.02% 75.98% 100.00%
% within extracurricular 100.00% 100.00% 100.00%
school-based sports
% of total 24.02% 75.98% 100.00%

(1) = 495.58, p < 0.001] were significantly different from were found to be significantly more physically active than
zero, suggesting that both the class-environment and indi- children not participating in extracurricular school-based
vidual pupil level characteristics may have an influence on sports [β = 157.62, S.E. = 28.54, χ2(1) = 30.50, p < 0.001]
pupils’ PA. Age, SES, autonomous motivation, controlled while children participating in community sports were
motivation, and amotivation were entered as covariates significantly more physically active [β = 57.07, S.E. = 27.95,
in the predictor models (Model 1a). Both age [β = 50.92, S. χ2(1) = 4.17, p = 0.041] than children not participating in
E. = 19.31, χ2(1) = 6.96, p = 0.008] and autonomous motiv- community sports. As none of the interaction effects
ation [β = 118.24, S.E. = 18.23, χ2(1) = 42.09, p < 0.001] between sex, participation in extracurricular school-based
were found to be significantly related to pupils’ PA-levels. sports and participation in community sports were statisti-
Pupil’s sex, extracurricular school-based sports participation cally significant, these were not included in the final model.
and community sports participation were then entered
as predictors in the next step. As shown in Table 4, Motivation towards sports and participation in
girls were significantly less physically active than boys extracurricular school-based sports
[β = -135.45, S.E. = 22.89, χ2(1) = 35.03, p < 0.001]. Autonomous motivation
After controlling for different covariates (see Table 4), In the null model for autonomous motivation, only the
children participating in extracurricular school-based sports variance at pupil-level was significantly different from

Table 2 Cross tabulation extracurricular school-based sports and sex

No extracurricular school-based Extracurricular school-based Total
sports participation sports participation
Boys N 97 466 563
% within sex 17.23% 82.77% 100.00%
% within extracurricular school-based sports 38.49% 58.47% 53.67%
% of total 9.25% 44.42% 53.67%
Girls N 155 331 486
% within sex 31.89% 68.11% 100.00%
% within extracurricular school-based sports 61.51% 41.53% 46.33%
% of total 14.78% 31.55% 46.33%
Total N 252 797 1049
% within sex 24.02% 75.98% 100.00%
% within extracurricular school-based sports 100.00% 100.00% 100.00%
% of total 24.02% 75.98% 100,00%
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 8 of 15
http://www.ijbnpa.org/content/11/1/48

Table 3 Cross tabulation community sports and sex zero [95.0%, χ2(1) = 496.44, p < 0.001], which implies that
No community Community Total there was no significant between-school and between-class
sports sports variance in pupils’ autonomous motivation towards sports.
participation participation
With regard to the included covariates (Table 5) controlled
Boys N 146 417 563 motivation and amotivation were, unlike age and SES, sig-
% within boys 25.93% 74.07% 100.00% nificantly related to pupils’ autonomous motivation. Higher
% within 51.23% 54.58% 53.67% scores on amotivation were related to lower scores on au-
community sports tonomous motivation [β = -0.55, S.E. = 0.03, χ2(1) = 395.72,
% of total 13.91% 39.75% 53.67% p < 0.001], while higher scores on controlled motivation
Girls N 139 347 486 were related to higher scores on autonomous motivation
% within girls 28.60% 71.40% 100.00% [β = 0.26, S.E. = 0.03, χ2(1) = 73.76, p < 0.001].
As the three-way interaction effect between sex, partici-
% within 48.77% 45.42% 46.33%
community sports pation in extracurricular school-based sports and partici-
% of total 13.25% 33.08% 46.33%
pation in community sports was statistically significant,
averages of autonomous motivation according to sex, par-
Total N 285 764 1049
ticipation in extracurricular school-based sports and par-
% within sex 27.17% 72.83% 100.00% ticipation in community sports were calculated for each
% within 100.00% 100.00% 100.00% possible combination by changing the reference categories
community sports
of each of the involved variables. Table 6 shows that boys
% of total 27.17% 72.83% 100.00% were in general more autonomously motivated to engage

Table 4 Relationship between sex, extracurricular school-based sports participation, community sports participation
and weekly self-reported physical activity
Weekly self-reported physical activity (Model 1)
Parameter Null model 1 Model 1a Model 1b Model 1c
Fixed part β (S.E.) β (S.E.) β (S.E.) β (S.E.)
Intercept 581.04 (21.78) 588.48 (24.10) 446.73 (62.62) 499.99 (38.15)
Age 64.88 (19.48)*** 52.60 (19.13)** 50.92 (19.31)**
SES (high)a −15.65 (24.06) −35.33 (23.35) −36.61 (23.31)
Autonomous motivation 156.94 (17.57)*** 117.70 (18.28)*** 118.24 (18.23)***
Controlled motivation 15.05 (19.05) 4.59 (18.37) 5.95 (18.43)
Amotivation 34.64 (19.89) 27.01 (19.14) 24.67 (19.17)
Pupil sex (girl)b −102.38 (74.18) −135.45 (22.89)***
Extracurricular school-based sports participation (yes)c 187.22 (68.93)* 157.62 (28.54)***
d
Community sports participation (yes) 85.47 (77.59) 57.07 (27.95)*
Girl x extracurricular school-based sports 59.36 (92.40)
Girl x community sports 40.27 (95.77)
Extracurricular school-based sports x community sports 11.23 (85.54)
Girl x extracurricular school-based −184.65 (114.12)
sports x community sports
Random part σ2 (S.E.) σ2 (S.E.) σ2 (S.E.) σ2 (S.E.)
School-level variance 3 712.82 (4 501.84) 5 858.29 (4 095.83) 4 360.46 (3 835.24) 4 343.75 (3 958.93)
Class-level variance 11 240.89 (5 295.85)* 7 318.49 (4 011.82) 8 408.89 (4 135.84)* 9 040.08 (4 321.37)*
Pupil-level variance 151 314.31 (6 796.73)*** 136 386.27 (6 125.50)*** 124 771.37 (5 604.52)*** 125 725.64 (5 647.62)***
Deviance test model 15 545.97 15 434.61*** 15 344.24*** 15 353.79***
χ (df)
2
11.36 (5) 90.37 (12) 80.18 (8)
Note. Values in parentheses are standard errors.
*p < .05; **p < .01, ***p < .001.
Reference category = 0; a0 = low and medium SES, 1 = high SES; b0 = boy, 1 = girl; c0 = no participation in extracurricular school-based sports in the current
school year, 1 = participation in extracurricular school-based sports in the current school year; d0 = no participation in community sports in the current school
year, 1 = participation in community sports in the current school year.
Model 1b and 1c were both compared to model 1a to calculate χ2.
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 9 of 15
http://www.ijbnpa.org/content/11/1/48

Table 5 Relationship between sex, extracurricular school-based sports participation, community sports participation
and autonomous motivation towards sports
Autonomous motivation towards sports (Model 2)
Parameter Null model 2 Model 2a Model 2b
Fixed part β (S.E.) β (S.E.) β (S.E.)
Intercept 4.15 (0.04) 4.13 (0.03) 3.38 (0.10)
Age 0.03 (0.03) 0.03 (0.03)
SES (high) a
0.05 (0.04) −0.02 (0.04)
Controlled motivation 0.30 (0.03)*** 0.26 (0.03)***
Amotivation −0.64 (0.03)*** −0.55 (0.03)***
Pupil sex (girl)b 0.32 (0.13)**
c
Extracurricular school-based sports participation (yes) 0.58 (0.12)***
Community sports participation (yes)d 0.90 (0.13)***
Girl x extracurricular school-based sports −0.55 (0.16)***
Girl x community sports −0.45 (0.16)**
Extracurricular school-based sports x community sports −0.49 (0.14)***
Girl x extracurricular school-based sports x community sports 0.58 (0.19)**
Random part σ (S.E.)
2
σ (S.E.)
2
σ2 (S.E.)
School-level variance 0.02 (0.01)* 0.00 (0.01) 0.01 (0.01)
Class-level variance 0.01 (0.01) 0.01 (0.01) 0.01 (0.01)
Pupil-level variance 0.66 (0.03)*** 0.43 (0.02)*** 0.36 (0.02)***
Deviance test model 2 556.76 2 116.65*** 1947.14***
Χ2 (df) 440.11 (4) 169.51 (11)
Note. Values in parentheses are standard errors.
*p < .05; **p < .01, ***p < .001.
Reference category = 0; a0 = low and medium SES, 1 = high SES; b0 = boy, 1 = girl; c0 = no participation in extracurricular school-based sports in the current
school year, 1 = participation in extracurricular school-based sports in the current school year; d0 = no participation in community sports in the current school
year, 1 = participation in community sports in the current school year.

in sports than girls. Only within the group of children Controlled motivation
participating in community sports but not in extracur- The random parts of the null model for controlled motiv-
ricular school-based sports, no significant differences ation showed that the variances at both the class- and the
were noted between boys and girls. For both boys and pupil-level differ significantly from zero with the pupil-
girls, participation in community sports was related to level variance [94.7%, χ2(1) = 496.43, p < 0.001] largely
the highest levels of autonomous motivation. It can also exceeding the class-level variance [5.2%, χ2(1) = 5.40, p =
be seen that boys who participate in extracurricular 0.020]. Of all the covariates that were inserted in the
school-based sports but not in community sports were model, only age was not significantly related to controlled
significantly more autonomously motivated than boys motivation (Table 7). Children with a high SES were over-
not engaging in either one of those forms of organized all less controlled motivated towards sports [β = -0.11, S.
sports [β = 0.58, S.E. = 0.19, χ2(1) = 8.96, p = 0.003], E. = 0.04, χ2(1) = 8.41, p = 0.04]. Pupils’ autonomous motiv-
whereas such differences were not noted among girls. ation [β = 0.26, S.E. = 0.03, χ2(1) = 79.68, p < 0.001] and

Table 6 Comparison of autonomous motivation towards sports between different groups

No community sports participation Community sports participation
No extracurricular school-based Extracurricular school-based No extracurricular school-based Extracurricular school-based
sports participation sports participation sports participation sports participation
β (S.E.) β (S.E.) β (S.E.) β (S.E.)
a b c
Boys 3.38 (0.10) 3.97 (0.07) 4.28 (0.09) 4.38 (0.04)c
a# a# b
Girls 3.71 (0.08) 3.74 (0.07) 4.16 (0.07) 4.28 (0.05)b#
A β is significantly different from another β if they have other superscript letters.
#
indicates a significant sex difference.
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 10 of 15
http://www.ijbnpa.org/content/11/1/48

Table 7 Relationship between sex, extracurricular school-based sports participation, community sports participation
and controlled motivation towards sports
Controlled motivation towards sports (Model 3)
Parameter Null model 3 Model 3a Model 3b Model 3c
Fixed part β (S.E.) β (S.E.) β (S.E.) β (S.E.)
Intercept 1.94 (0.03) 2.00 (0.03) 2.01 (0.10) 2.08 (0.06)
Age 0.01 (0.03) 0.00 (0.03) 0.00 (0.03)
SES (high) a
−0.12 (0.04)** −0.11 (0.04)** −0.11 (0.04)**
Autonomous motivation 0.25 (0.03)*** 0.26 (0.03)*** 0.26 (0.03)***
Amotivation 0.23 (0.03)*** 0.23 (0.03)*** 0.22 (0.03)***
Pupil sex (girl)b −0.04 (0.13) −0.09 (0.04)*
c
Extracurricular school-based sports participation (yes) 0.09 (0.12) 0.04 (0.05)
Community sports participation (yes)d 0.05 (0.13) −0.10 (0.05)*
Girl x extracurricular school-based sports 0.04 (0.16)
Girl x community sports −0.12 (0.19)
Extracurricular school-based sports x community sports −0.12 (0.14)
Girl x extracurricular school-based sports x community sports 0.03 (0.19)
Random part σ (S.E.)
2
σ (S.E.)
2
σ2 (S.E.) σ2 (S.E.)
School-level variance 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Class-level variance 0.02 (0.01)* 0.02 (0.01)* 0.02 (0.01)* 0.02 (0.01)*
Pupil-level variance 0.40 (0.02)*** 0.36 (0.02)*** 0.36 (0.02)*** 0.36 (0.02)***
Deviance test model 2 045.87 1 946.34*** 1 933.49 1 936.01
Χ2 (df) 99.53 (5) 12.85 (11) 10.33 (7)
Note. Values in parentheses are standard errors.
*p < .05; **p < .01, ***p < .001.
Reference category = 0; a0 = low and medium SES, 1 = high SES; b0 = boy, 1 = girl; c0 = no participation in extracurricular school-based sports in the current
school year, 1 = participation in extracurricular school-based sports in the current school year; d0 = no participation in community sports in the current school
year, 1 = participation in community sports in the current school year.
Model 3b and 3c were both compared to model 3a to calculate χ2.

their amotivation [β = 0.22, S.E. = 0.03, χ2(1) = 50.26, p < positively related to amotivation [β = 0.20, S.E. = 0.03, χ2
0.001] were both positively associated with their controlled (1) = 49.49, p < 0.001] (Table 8). Since none of the inter-
motivation. The interaction effects were not retained in action effects were statistically significant in the model for
the final model for controlled motivation since none of amotivation, they were not retained in the final model.
these interaction effects were statistically significant. Children participating versus not participating in extra-
As can be seen in Table 7, boys were overall more con- curricular school-based sports did not significantly differ
trolled motivated than girls [β = 0.09, S.E. = 0.04, χ2(1) = with regard to amotivation [β = -0.02, S.E. = 0.04, χ2(1) =
5.1, p = 0.024], while children participating in community 0.20, p = 0.65]. Similarly children participating in commu-
sports were less controlled motivated than the reference nity sports did not, compared to children not participating
category [β = -0.10, S.E. = 0.05, χ2(1) = 4.21, p = 0.040]. in community sports, significantly differ in terms of amoti-
Children participating versus not participating in extracur- vation [β = -0.01, S.E. = 0.05, χ2(1) = 0.03, p = 0.86]. There
ricular school-based sports did not significantly differ with is, however, a significant difference between boys’ and
regard to controlled motivation. girls’ amotivation towards sports with girls being signifi-
cantly less amotivated than boys [β = -0.10, S.E. = 0.04, χ2
Amotivation (1) = 7.13, p = 0.007].
Similar to autonomous motivation, only the variance at
pupil-level was significantly different from zero [99.1%, χ2 Discussion
(1) = 510.56, p < 0.001], implying a lack of between-school In the present study extracurricular school-based sports
and between-class variance in pupils’ amotivation towards participation was investigated as one possible vehicle to
sports. Table 7 shows that autonomous motivation was offer opportunities for children to be physically active.
negatively related to amotivation [β = -0.50, S.E. = 0.03, χ2 Children in the last two grades of elementary school
(1) = 403.54, p < 0.001] while controlled motivation was were chosen as the target population of the present
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 11 of 15
http://www.ijbnpa.org/content/11/1/48

Table 8 Relationship between sex, extracurricular school-based sports participation, community sports participation
and amotivation towards sports
Amotivation towards sports (Model 4)
Parameter Null model 4 Model 4a Model 4b Model 4c
Fixed part β (S.E.) β (S.E.) β (S.E.) β (S.E.)
Intercept 1.38 (0.03) 1.40 (0.03) 1.49 (0.10) 1.47 (0.06)
Age 0.03 (0.03) 0.03 (0.03) 0.03 (0.03)
SES (high) a
−0.05 (0.04) −0.05 (0.04) −0.05 (0.04)
Autonomous motivation −0.50 (0.02)*** −0.50 (0.03)*** −0.50 (0.03)***
Controlled motivation 0.21 (0.03)*** 0.20 (0.03)*** 0.20 (0.03)***
Pupil sex (girl)b −0.08 (0.12) −0.10 (0.04)**
Extracurricular school-based sports participation (yes)c 0.00 (0.11) −0.02 (0.04)
Community sports participation (yes)d −0.10 (0.13) −0.01 (0.05)
Girl x extracurricular school-based sports −0.16 (0.15)
Girl x community sports 0.06 (0.15)
Extracurricular school-based sports x community sports 0.04 (0.14)
Girl x extracurricular school-based sports x community sports 0.11 (0.18)
Random part σ (S.E.)
2
σ (S.E.)
2
σ2 (S.E.) σ2 (S.E.)
School-level variance 0.01 (0.01) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Class-level variance 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Pupil-level variance 0.50 (0.02)*** 0.34 (0.02)*** 0.33 (0.02)*** 0.34 (0.02)***
Deviance test model 2 265.68 1 848.46*** 1 837.07 1 841.31
Χ2 (df) 417.22 (4) 11.39 (11) 7.15 (7)
Note. Values in parentheses are standard errors.
**p < .01, ***p < .001.
Reference category = 0; a0 = low and medium SES, 1 = high SES; b0 = boy, 1 = girl; c0 = no participation in extracurricular school-based sports in the current
school year, 1 = participation in extracurricular school-based sports in the current school year; d0 = no participation in community sports in the current school
year, 1 = participation in community sports in the current school year.
Model 4b and 4c were both compared to model 4a to calculate χ2.

study, since this age coinciding with the transition from sports has the advantage that it does not require additional
childhood to adolescence is typically characterized by a transportation as extracurricular school-based sports are
decrease in both PA and sports participation [28]. The organized at school or in its neighborhood at times when
results of the current study demonstrate that, if extra- the children are already at school. The school environment
curricular school-based sports are offered in elementary is also a familiar setting in which children may consider
school, an overall high participation rate in extracurricular the barrier to participate in sports as relatively low. Espe-
school-based sports is found. cially for the less sports talented children, the step towards
The results furthermore reveal that extracurricular more selective and competitive community sports may
school-based sports not only reach children who already seem challenging. For this group of children, extracurricu-
participate in community sports, but also almost two third lar school-based sports can be a medium to get them
(64.91%) of the children who do not yet participate in com- acquainted with different kinds of sports without being
munity sports. This means that some pupils who do not pressured to compete. In a long-term perspective, extra-
participate in sports outside the school context, nonethe- curricular school-based sports may then serve as a vehicle
less engage in extracurricular school-based sports activities. to community sports participation.
In contrast to community sports, the vast majority of Despite the promising findings of the present study, it
extracurricular school-based sports programs in Flanders must be noted that in our sample, only schools with an
are free of charge. It is possible that some children would extracurricular school-based sports offer were included.
like to participate in community sports, but do not get that Research from the Flemish Policy Centre Culture, Youth
opportunity due to financial constraints. Eime et al. [23,42] and Sports (2010) [43] shows that only 26% of the Flemish
indicated that transportation issues might be another rea- elementary schools actually offer extracurricular school-
son why children do not engage in community sports out- based sports. As extracurricular school-based sports pro-
side school. Participation in extracurricular school-based grams reach almost two third of the children who do not
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 12 of 15
http://www.ijbnpa.org/content/11/1/48

engage in community sports, it would seem important that might also be the case that a number of boys, by partici-
effort is put into offering extracurricular school-based pating in extracurricular school-based sports, develop
sports in more elementary schools. more autonomous motivation towards sports. Earlier re-
A second aim of the present study was to assess whether search by Hagger et al. [27] indicates that autonomous
extracurricular school-based sports participants were more motives in a PE context are related to autonomous mo-
physically active in daily life than children who do not tives in a leisure-time context in secondary school children
participate in extracurricular school-based sports, while from diverse countries. In other words, children who are
controlling for their participation in community sports. autonomously motivated to participate in PE are more au-
The results indicate that children who participate in extra- tonomously motivated to participate in sports than chil-
curricular school-based sports are more physically active in dren who are less or not motivated to participate in PE,
daily life than their non-participating peers. This difference suggesting that transfer across contexts can take place.
in PA attributed to participation in extracurricular school- Given its cross-sectional nature, the results of the present
based sports is even larger than the increase in PA due to study might therefore also indicate that participation in
participation in community sports. Furthermore, the aver- extracurricular school-based sports stimulates the devel-
age amount of weekly time spent in extracurricular school- opment of autonomous motivation towards sports be-
based sports by the participants is only a fraction of the cause pupils start to value and enjoy activities through
difference in weekly PA between extracurricular school- participating in it. However, this reasoning would only
based sports participants and non-participants, suggesting hold for boys and not for girls, as for girls levels of autono-
that children participating in extracurricular school-based mous motivation were overall more alike across groups,
sports have an overall more active lifestyle than their coun- independent of whether they engaged in extracurricular
terparts, independent of whether they engage in in com- school-based sports or not.
munity sports or not. These results are in line with the The results further indicate that extracurricular
findings of Sallis et al. [19] and Haerens et al. [20] who school-based sports participation is not related to pupils’
found that the provision of PA organized at school is posi- controlled motivation or amotivation towards sports.
tively related to children’s total PA-levels. It is possible that The combination of children’s overall low levels of amo-
participation in extracurricular school-based sports en- tivation and high levels of autonomous motivation appear
courages children to be physically active in leisure time but to confirm the generally positive attitude that children of
not necessarily to participate in organized community this age group have towards sports.
sports. More specifically, the acquaintance with new sports
or skill improvement in familiar sports can stimulate chil- Practical implications
dren to engage in sports activities outside the school con- As children who do not engage in extracurricular school-
text, such as playing soccer with friends on a grass field in based sports or in community sports have lower PA-levels
the neighborhood [44]. On the other hand, given the and a significantly lower autonomous motivation towards
cross-sectional character of the study, it is also possible sports, the question arises how these children can be stim-
that children who are more generally inclined to be physic- ulated and motivated to engage in sports or in PA more
ally active in leisure time are more engaged in extracurric- generally. One possibility would be to adapt extracurricu-
ular school-based sports participation. lar school-based sports programs to the needs of this spe-
The third aim of the present study was to evaluate cific target group so that these programs become more
whether children who participate in extracurricular attractive and thus more accessible for this group of chil-
school-based sports, compared to those who do not, are dren. Another target group that requires special attention
better (i.e., more autonomously) or relatively worse (i.e., are girls since they have lower participation rates in extra-
more controlled) motivated or amotivated to participate curricular school-based sports than boys. Furthermore,
in sports in general. It was found that boys not engaging girls are overall less physically active and, with the excep-
in community sports but attending extracurricular school- tion of the group that does not participate in extracurricular
based sports programs are more autonomously motivated school-based sports or community sports who displayed
towards sports than boys not engaging in community nor low levels of autonomous motivation overall, they are less
extracurricular school-based sports. It is thus indeed pos- autonomously motivated towards sports than boys. The
sible that there exists a group of boys who are autono- type of sports offered in extracurricular school-based sports
mously motivated towards sports, who found their way to programs might be one of the reasons why girls are less
extracurricular school-based sports but not to community attracted to extracurricular school-based sports than boys.
sports. For this group of autonomously motivated boys In most schools, much of the extracurricular school-based
who might not get the chance to participate in community sports time is dedicated to traditional team sports (e.g., soc-
sports, extracurricular school-based sports could serve as cer, netball and hockey), which are typically more popular
a perfect opportunity to play sports. On the other hand, it among boys than girls. This emphasis on traditional ball
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 13 of 15
http://www.ijbnpa.org/content/11/1/48

games is similar to the focus in the extracurricular school- study were validated for the target population and a
based sports offer in schools in England and Wales [11] researcher guided the questioning procedure.
and the United States [45]. Even though swimming and The present study had also some considerable strengths.
athletics, which are considered attractive sports for both A first strength of the present study was the sample size
boys and girls, are part of most schools’ extracurricular (N = 1049), obtained from 60 different classes out of 30 di-
school-based sports offer, they are often only once-a-year- verse elementary schools, which were located in different
offered activities and might therefore not contribute to con- areas (urban, suburban and rural). This relatively large
tinued extracurricular school-based sports participation. To sample allowed us to presume that the results of the
attract more girls to participate in extracurricular school- current study are representative for all Flemish children of
based sports it might be effective to more frequently offer the last two grades of elementary school, attending schools
those sports that are popular among girls (e.g. swimming, with an extracurricular school-based sports offer.
dance, rope skipping, gymnastics and aerobics) [46-48]. Another strength was the procedure followed to collect
From a practical point of view, it is also important to data. All the questions were read out loud and children
keep in mind that extracurricular school-based sports are had the possibility to ask further explanation if questions
just one of various potential mediums to motivate children were unclear. This method has undoubtedly contributed
to an active lifestyle. The positive relationship between to a reliable data set without missing values.
participation in extracurricular school-based sports and Third, multilevel regression analyses were employed to
overall PA-levels makes extracurricular school-based sports analyze the hierarchical data. With regard to the frequency
one of many promising settings to stimulate inactive and the content of the offer, extracurricular school-based
children to a more active lifestyle. sports can vary at school-level. By using multilevel ana-
lyses, this variance was automatically taken into account.
Strengths and limitations
Since the current study was cross-sectional, it is not Conclusions
possible to determine the direction of the relationship Extracurricular school-based sports programs are often
between extracurricular school-based sports participation seen as ideal channels to promote sports participation and
on the one hand and PA and motivation towards sports PA among children who are not active in community
on the other hand, which is a limitation of the present sports. The results of the present study indicate that in
study. The question therefore remains if participation in elementary schools offering extracurricular school-based
extracurricular school-based sports triggers a more active sports, many children engage in those activities. The
lifestyle and a more autonomous motivation towards extracurricular school-based sports programs attract not
sports or if it is rather vice versa and children who already only children already engaging in community sports, but
have high PA-levels and a high autonomous motivation to also children not yet participating in community sports.
participate in sports are more attracted to extracurricular
school-based sports than their peers. In this respect, it Additional file
would be interesting to investigate the direction of these
relationships by implementing extracurricular school- Additional file 1: Methodology of the FPAQ validity study.
based sports in elementary schools not offering extracur-
ricular school-based sports at the moment. Intervention Abbreviations
studies based on such implementations or longitudinal BRPEQ: Behavioral regulations in physical education questionnaire;
FPAQ: Flemish physical activity questionnaire; PA: Physical activity; PE: Physical
studies could provide more clarity on the direction of the education; SDT: Self-determination theory; SES: Socioeconomic status.
relationships.
Besides the cross-sectional design of the study, a poten- Competing interests
The authors declare that they have no competing interests.
tial limitation is the way pupils were classified with regard
to extracurricular school-based sports participation. As Authors’ contributions
participation in extracurricular school-based sports was ADM contributed to the questionnaire design, data collection, data
management, statistical analysis, interpretation of results and drafting the
questioned by a simple ‘yes’ or ‘no’-question, pupils only manuscript. NA contributed to the statistical analysis, interpretation of results
participating once in extracurricular school-based sports and critical review of the manuscript. GC contributed to the study design,
during the current school year were categorized as extra- questionnaire design and critical review of the manuscript. IDB contributed
to the study design and critical review of the manuscript. LH contributed to
curricular school-based sports participants. Consequently, the study design, questionnaire design, statistical analysis, interpretation of
the overall high percentage of participants may give a results, and critical review of the manuscript. All authors have read and
slightly distorted view. A third limitation is the possibility approved the final manuscript.
of over- or underestimation of certain variables (e.g. the
Acknowledgements
weekly PA), especially given the young age of the partici- This study was supported by the Policy Research Centre on Sports and is
pants. However, all questionnaires used in the present funded by the Flemish Government.
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 14 of 15
http://www.ijbnpa.org/content/11/1/48

Author details 20. Haerens L, Craeynest M, Deforche B, Maes L, Cardon G, De Bourdeaudhuij I:

1
Department of Movement and Sports Sciences, Ghent University, The contribution of home, neighbourhood and school environmental
Watersportlaan 2, 9000 Ghent, Belgium. 2The Policy Research Centre on factors in explaining physical activity among adolescents. J Environ Public
Sports, funded by the Flemish Government, Brussels, Belgium. 3Department Health 2009, 2009:10.
of Developmental, Personality and Social Psychology, Ghent University, 21. Trost SG, Rosenkranz RR, Dzewaltowski D: Physical activity levels among
Watersportlaan 2, 9000 Ghent, Belgium. 4The Flemish Research Foundation, children attending after-school programs. Med Sci Sports Exerc 2008,
Brussels, Belgium. 40(4):622–629.
22. Beets MW, Beighle A, Erwin HE, Huberty JL: After-school program impact
Received: 12 August 2013 Accepted: 28 March 2014 on physical activity and fitness a meta-analysis. Am J Prev Med 2009,
Published: 7 April 2014 36(6):527–537.
23. Eime RM, Harvey JT, Craike MJ, Symons CM, Payne WR: Family support and
ease of access link socio-economic status and sports club membership
References in adolescent girls: a mediation study. Int J Behav Nutr Phys Act 2013,
1. Brettschneider WD, Naul R: Obesity in Europe: young people’s physical 10(50):doi:10.1186/1479-5868-10-50.
activity and sedentary lifestyles. In Sport Sciences International. Volume 4. 24. Deci EL, Ryan RM: The “what” and “why” of goal pursuits: human needs
Edited by Brettschneider WD, Naul R. Frankfurt am Main: Peter Lang; 2007:7–26. and the self-determination of behavior. Psychol Inq 2000, 11(4):227–268.
2. Nuviala Nuviala A, Tamayo Fajardo JA, Nuviala Nuviala R: Perceived quality 25. Guay F, Ratelle CF, Chanal J: Optimal learning in optimal contexts: the
of school sports as predictor of sports dropout in adolescents. Revista role of self-determination in education. Can Psychol 2008, 49(3):233–240.
Internacional de Medicina y Ciencias de la Actividad Física y del Deporte 2012, 26. Vansteenkiste M, Niemiec C, Soenens B: The development of the five mini-
12(47):389–404. theories of self-determination theory: an historical overview, emerging
3. Currie C, Gabhainn SN, Godeau E, Roberts C, Smith R, Currie D, Picket W, trends, and future directions. In Advances in Motivation and Achievement.
Richter M, Morgan A, Barnekow V (Eds): Inequalities in Young people’s Health: Volume 16 (The Decade Ahead). Edited by Urdan T, Karabenick S. UK:
HBSC International Report from the 2005/2006 Survey. Health Policy for Emerald Publishing; 2010:105–166.
Children and Adolescents, (No. 5). Copenhagen: WHO Regional Office for 27. Hagger MS, Chatzisarantis NLD, Barkoukis V, Wang CKJ, Baranowski J:
Europe; 2008. Perceived autonomy support in physical education and leisure-time
4. Omorou YA, Erpelding ML, Ascalon H, Vuillemin A: Contribution of taking physical activity: a cross-cultural evaluation of the trans-contextual
part in sport to the association between physical activity and quality of model. J Educ Psychol 2005, 97(3):376–390.
life. Qual Life Res 2013, 8(22):2021–2029. 28. Petlichkoff LM: The dropout dilemma in youth sports. In The Child and
5. Stevenson A: Oxford Dictionary of English. 3rd edition. Oxford: Oxford Adolescent Athlete: Encyclopedia of Sports Medicine. Volume 6. Edited by
University Press; 2010. Bar-Or O. Oxford: Blackwell Science; 1996:418–432.
6. Cale L, Harris J: School-based physical activity interventions: effectiveness, 29. Biddle SJ, Gorely T, Stensel DJ: Health-enhancing physical activity and
trends, issues, implications and recommendations for practice. Sport Educ sedentary behaviour in children and adolescents. J Sports Sci 2004, 22:679–701.
Soc 2006, 11(4):401–420. 30. Hollingshead AA: Four-Factor Index of Social Status. New Haven, CT: Yale University;
7. Fox KR, Cooper A, McKenna J: The school and the promotion of children’s 1975. http://psy6023.alliant.wikispaces.net/file/view/hollingshead+ses.pdf.
health-enhancing physical activity: perspectives from the United Kingdom. 31. Philippaerts RM, Matton L, Wijndaele K, Balduck AL, De Bourdeaudhuij I,
J Teach Phys Educ 2004, 23(4):338–358. Lefevre J: Validity of a physical activity computer questionnaire in 12-to
8. Pate RR, O’Neill JR: After-school interventions to increase physical activity 18-year-old boys and girls. Int J Sports Med 2006, 27(2):131–136.
among youth. Br J Sports Med 2009, 43(1):14–18. 32. Verstraete SJM, Cardon GM, Trost SG, De Bourdeaudhuij IMM: Reliability
9. Green K, Smith A, Roberts K: Young people and lifelong participation in and validity of a questionnaire to measure usual physical activity in
physical education: a sociological perspective on contemporary physical children with and without parental assistance. In The Effectiveness of an
education programmes in England and Wales. Leis Stud 2005, 24:27–43. Intervention Promoting Physical Activity in Elementary School Children. PhD
10. Haerens L, Kirk D, Cardon G, De Bourdeaudhuij I, Vansteenkiste M: thesis. Ghent University; 2006.
Motivational profiles for secondary school physical education and its 33. Haerens L, De Bourdeaudhuij I, Maes L, Cardon G, Deforche B: School-
relationship to the adoption of a physically active lifestyle among based randomized controlled trial of a physical activity intervention
university students. Eur Phys Educ Rev 2010, 16(2):117–139. among adolescents. J Adolesc Health 2007, 40(3):258–265.
11. Pühse U, Gerber M: International Comparison of Physical Education. Oxford: 34. Cardon G, Haerens L, Verstraete S, De Bourdeaudhuij I: Perceptions of a
Meyer & Meyer Sport (UK) Ltd; 2005. school-based self-management program promoting an active lifestyle
12. Leatherdale S, Manske S, Faulkner G, Arbour K, Bredin C: A multi-level among elementary schoolchildren, teachers, and parents. J Teach Phys
examination of school programs, policies and resources associated with Educ 2009, 28(2):141–154.
physical activity among elementary school youth in the PLAY-ON study. 35. D’Hondt E, Deforche B, Gentier I, De Bourdeaudhuij I, Vaeyens R,
Int J Behav Nutr Phys Activ 2010, 7(6):doi:10.1186/1479-5868-7-6. Philippaerts R, Lenoir M: A longitudinal analysis of gross motor
13. Penney D, Harris J: Extra-curricular physical education: more of the same coordination in overweight and obese children versus normal-weight
for the more able? Sport Educ Soc 2007, 2(1):41–54. peers. Int J Obes 2012, 37(1):61–67.
14. Powers HS, Conway TL, McKenzie TL, Sallis JF, Marshall SJ: Participation in 36. Aelterman N, Vansteenkiste M, Van Keer H, Van den Berghe L, De Meyer J,
extracurricular physical activity programs at middle schools. Res Q Exerc Haerens L: Students’ objectively measured physical activity levels and
Sport 2002, 73(2):187–192. engagement as a function of between-class and between-student
15. Lubans D, Morgan P: Impact of an extra-curricular school sport differences in motivation towards physical education. J Sport Exerc
programme on determinants of objectively measured physical activity Psychol 2012, 34(4):457–480.
among adolescents. Health Educ J 2008, 67(4):305–320. 37. Markland D, Tobin V: A modification to the behavioural regulation in
16. Van Acker R, De Bourdeaudhuij I, De Martelaer K, Seghers J, Kirk D, Haerens L, exercise questionnaire to include an assessment of amotivation.
De Cocker K, Cardon G: A framework for physical activity programs within J Sport Exerc Psychol 2004, 26(2):191–196.
school-community partnerships. Quest 2011, 63(3):300–320. 38. Maas C, Hox J: Sufficient sample size for multilevel modeling.
17. Pot N, Van Hilvoorde I: Generalising the effects of school sports: Methodology 2005, 1(3):85–91.
comparing the cultural contexts of school sports in the Netherlands and 39. Greenland S: Introduction to regression modeling. In Modern Epidemiology.
the USA. Sport Soc 2013, 16(9):1164–1175. 2nd edition. Edited by Rothman KJ, Greenland S. Philadelphia: Lippincott-
18. Park RJ: Sport, sex and society in a transatlantic Victorian perspective. Raven; 1998:chapter 21.
Int J Hist Sport 2007, 24(12):1570–1603. 40. Jones K, Subramanian SV: Developing Multilevel Models for Analysing
19. Sallis JF, McKenzie TL, Conway TL, Elder JP, Prochaska JJ, Brown M, Zive MM, Contextuality, Heterogeneity and Change Using MLwiN 2.2, Volume 1. Centre
Marshall SJ, Alcarez JE: Environmental interventions for eating and for Multilevel Modelling: University of Bristol; 2012.
physical activity. A randomized controlled trial in middle schools. 41. Rasbash J, Steele F, Browne WJ, Goldstein H: A User’s Guide to MLwiN, v2.10.
Am J Prev Med 2003, 24(3):209–217. Centre for Multilevel Modelling: University of Bristol; 2009.
De Meester et al. International Journal of Behavioral Nutrition and Physical Activity 2014, 11:48 Page 15 of 15
http://www.ijbnpa.org/content/11/1/48

42. Kirk D, Carlson T, O’Connor A, Burke P, Davis K, Glover S: The economic

impact on families of children’s participation in junior sport. Aust J Sci
Med Sport 1997, 29(2):27–33.
43. Van Acker R, De Bourdeaudhuij I, De Martelaer K, Seghers J, Cardon G: Sport
en beweging binnen een Brede School. In Sport voor allen. Strategieën
voor laagdrempelig bewegen en sporten in Vlaanderen. Volume 4. Edited by
Phillipaerts R. Leuven: Acco; 2011:105–122.
44. Jago R, Baranowski T: Non-curricular approaches for increasing physical
activity in youth: a review. Prev Med 2004, 39(1):157–163.
45. Kanters MA, Bocarro JN, Edwards MB, Casper JM, Floyd MF: School sport
participation under two school sport policies: comparisons by race/ethnicity,
sex, and socioeconomic status. Ann Behav Med 2013, 45(S1):S113–S121.
46. Jago R, Davis L, McNeill J, Sebire SJ, Haase A, Powell J, Cooper AR:
Adolescent girls’ and parents’ views on recruiting and retaining girls into
an after-school dance intervention: implications for extra-curricular
physical activity provision. Int J Behav Nutr Phys Activ 2011,
8(91):doi:10.1186/1479-5868-8-91.
47. Oliver KL, Hamzeh M: ‘The boys won’t let us play’: 5th grade mestizas
publicly challenge physical activity discourse at school. Res Q Exerc Sport
2010, 81(1):39–51.
48. Oliver KL, Hamzeh M, McCaughtry N: ‘Girly girls can play games/Las Niñas
Pueden Jugar Tambien’: co-creating a curriculum of possibilities with 5th
grade girls. J Teach Phys Educ 2009, 28(1):90–110.

doi:10.1186/1479-5868-11-48
Cite this article as: De Meester et al.: Extracurricular school-based sports
as a motivating vehicle for sports participation in youth: a cross-sectional
study. International Journal of Behavioral Nutrition and Physical Activity
2014 11:48.

Submit your next manuscript to BioMed Central

and take full advantage of:

• Convenient online submission

• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution

Submit your manuscript at

www.biomedcentral.com/submit