Predicting accelerometer-based physical activity in physical education and total physical activity : The Self-determination Theory approach

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/ Predicting accelerometer-based physical activity in physical education and total physical activity : The Self-determination Theory approach © 2019University of Alicante Published version Gråstén, Arto; Yli-Piipari, Sami; Huhtiniemi, Mikko; Salin, Kasper; Seppälä, Sanni; Lahti, Jukka; Hakonen, Harto; Jaakkola, Timo Gråstén, A., Yli-Piipari, S., Huhtiniemi, M., Salin, K., Seppälä, S., Lahti, J., Hakonen, H., & Jaakkola, T. (2019). Predicting accelerometer-based physical activity in physical education and total physical activity : The Self-determination Theory approach. Journal of Human Sport and Exercise, 14(4), 757-771. https://doi.org/10.14198/jhse.2019.144.05 2019 VOLUME -- | ISSUE - | 2019 | 1 Predicting accelerometer-based physical activity in physical education and total physical activity: The Self-determination Theory approach ARTO GRÅSTÉN 1 , SAMI YLI-PIIPARI2, MIKKO HUHTINIEMI3, KASPER SALIN3, SANNI SEPPÄLÄ3, JUKKA LAHTI3, HARTO HAKONEN4, TIMO JAAKKOLA3 1School of Education, University of Tasmania, Launceston, Australia 2College of Education, University of Georgia, Athens, United States of America 3Faculty of Sport and Health Sciences, University of Jyväskylä, Jyväskylä, Finland 4LIKES Research Center, Jyväskylä, Finland ABSTRACT The present study tested the motivational model of physical education (PE) including needs for competence, autonomy, social relatedness, intrinsic and extrinsic motivation, in-class moderate to vigorous physical activity (MVPA), and total MVPA. Participants were 490 (264 girls, 226 boys) Finnish elementary school students. The data were collected using accelerometers and questionnaires for a seven-day period during the fall semester 2017. The key findings were that 1) social relatedness associated with total MVPA via inclass MVPA in girls, whereas competence was linked to in-class MVPA through extrinsic motivation in boys, 2) competence was positively linked to extrinsic motivation in a similar way in both girls and boys, 3) social relatedness and in-class MVPA were positively associated with total MVPA in both girls and boys, 4) competence, autonomy, and relatedness were positively linked to intrinsic motivation in girls when only competence and autonomy were related to intrinsic motivation in boys, and 5) in-class MVPA contributed 36% of total weekly MVPA minutes in the present sample. Although the indirect relationships between study variables did not fully support the existing PE motivational model, the direct associations showed that needs 1 Corresponding author. School of Education, University of Tasmania, Locked Bag 1307, Launceston, Tasmania, Australia. https://orcid.org/0000-0002-9598-7949 E-mail: arto.graste[email protected] Submitted for publication October 2018 Accepted for publication December 2018 Published in press January 2019 JOURNAL OF HUMAN SPORT & EXERCISE ISSN 1988-5202 © Faculty of Education. University of Alicante doi:10.14198/jhse.2019.144.05 Original Article Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE 2 | 2019 | ISSUE - | VOLUME -- © 2019 University of Alicante of competence, autonomy, relatedness could be promoted in PE classes to support intrinsic motivation, and total MVPA. Keywords: Psychological needs; Exercise motivation; School; Structural equation modelling. Cite this article as: Gråstén, A., Yli-Piipari, S., Huhtiniemi, M., Salin, K., Seppälä, S., Lahti, J., Hakonen, H., & Jaakkola, T. (2019). Predicting accelerometer-based physical activity in physical education and total physical activity: The Self-determination Theory approach. Journal of Human Sport and Exercise, in press. doi:https://doi.org/10.14198/jhse.2019.144.05 Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE VOLUME -- | ISSUE - | 2019 | 3 INTRODUCTION Technological advances of modern society have contributed to a lifestyle that has changed children’s physical activity behaviour from that of 20 years ago (Institute of Medicine, 2013). Many children are not participating in sufficient physical activity to maintain good health and well-being, as based on the recent international comparison only 30% of children and youth meet the current guidelines of daily 60 minutes of moderate to vigorous physical activity (MVPA) (Tremblay et al., 2016). Since school-aged students spend almost half of their waking hours in compulsory education, school physical education (PE) has been suggested to be one of the most cost-effective venues to increase children’s MVPA engagement (Bassett et al., 2013; McKenzie & Lounsbery, 2009). Existing evidence indicates that positive motivational experiences children gain from PE may play the central role when they adopt patterns of regular physical activity (Nupponen et al., 2010; Vanttaja et al., 2017). The present study utilized the motivational model of PE (Vallerand & Lalande, 2011) grounded in the Self-determination Theory (SDT; Deci & Ryan, 2000, 2008) to examine the associations among students’ motivational experiences, in-class MVPA, and total MVPA. School PE is defined as the planned and progressive learning that takes place in school curriculum timetabled time and which is delivered to all students and involves both “learning to move” (i.e. becoming more physically competent) and “moving to learn” (i.e. learning through movement) elements (Harris, 2015). The amount of MVPA during PE classes depends on many factors, such as student skill level (Fairclough & Stratton, 2006), lesson theme (Slingerland, 2014), gender (Slingerland, 2014), lesson context (Mersh & Fairclough, 2010), resources (Levin et al., 2001), teacher specialization (Sallis et al., 1997), and class size (McKenzie & Marshall, 2000). Therefore, a few previous studies reported in the PE context showed that there is a large variation in the amount of objective MVPA that PE classes contributes to total MVPA, ranging from 8% based on accelerometer scores (Dale, Corbin, & Dale, 2000) to 34% obtained using fitness trackers (Gråstén et al., 2015). For instance, previous findings in the sample of Finnish middle school students revealed that girls met up to 26% and boys 34% of their weekly objectively measured MVPA for two 45-minute PE classes (Gråstén et al., 2015). However, motivational experiences in school PE and its contributions to actual in-class MVPA and overall MVPA is largely unknown (Ntoumanis & Standage, 2009; Vanttaja et al., 2017). SDT (Deci & Ryan, 2000) addresses the relationships between social context and motivation. For example, SDT can be used to illuminate the relationships between intrinsic and extrinsic motivation in PE, in-class MVPA, and overall MVPA. Intrinsic motivation represents the situations when a student is motivated to engage in an activity because it is personally rewarding, i.e. finding the activity interesting or enjoyable without external contingencies (Deci & Ryan, 2008). Previous PE research has indicated that intrinsic motivation is linked with higher activity levels in PE classes (Gråstén & Watt, 2017; Lonsdale et al., 2009) and daily self-reported MVPA (Gråstén & Watt, 2017; Kalaja, 2012). Extrinsic motivation occurs when the student is motivated to perform a behaviour or engage in an activity to earn a reward, i.e. participating in PE activities to get higher grades or to win awards (Deci & Ryan, 2008). Extrinsic motivation has been shown to trigger negative behavioral outcomes, such as low physical activity in PE classes (Gao, 2012; Gao, Podlog, & Huang, 2013). The development of intrinsic and extrinsic motivation is specified using the concept of basic psychological needs, namely competence, autonomy and social relatedness (Deci & Ryan, 2008). Competence refers to beliefs about the ability to be successful in an achievement domain (Deci & Ryan, 2008). Previous findings have consistently showed that physical competence is positively linked with intrinsic motivation (Deci & Ryan, 2000; Gråstén & Watt, 2017; Standage, Duda & Ntoumanis, 2005) and self-reported MVPA (Cox, Smith, & Williams, 2008; Yli-Piipari, 2011). Autonomy is a need to self-determine behavior and to achieve consistency Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE 4 | 2019 | ISSUE - | VOLUME -- © 2019 University of Alicante between the particular activity and sense of self-control (Deci & Ryan, 2008). Previous studies grounded in SDT have provided evidence that higher perception of autonomy is associated with higher intrinsic motivation and MVPA levels during PE classes (Lonsdale et al., 2009) and self-reported total MVPA (Chatzisarantis & Hagger, 2009; Gråstén & Watt, 2017). Social relatedness is the need to feel connected and to perceive acceptance from other students (Deci & Ryan, 2008). Past studies have shown that PE classes that support perceptions of social relatedness predicted self-reported MVPA participation (Cox, Smith, & Williams, 2008; Standage, Duda, & Ntoumanis, 2005; Taylor et al., 2010). Previous PE studies have also found competence to be the strongest predictor of intrinsic motivation compared to autonomy and relatedness (Standage, Duda, & Ntoumanis, 2005; Taylor et al., 2010). Applied to a PE context, when a student feels physically competent, experiences freedom to make choices or influence on PE activities, and has a pleasant and supportive learning environment created by the teacher and peers, she or he will likely to be more intrinsically motivated in PE, leading them toward being more physically active (Slingerland, 2014; Weiss,Corbin, & Pangrazi, 2000). Figure 1. The theorized motivational model of PE including accelerometer-based in-class and total MVPA An evident limitation in respect to past research in the field is that only a limited number of studies (e.g. Dale, Corbin, & Dale, 2000; Gråstén et al., 2015) examined the amount of objectively measured MVPA that PE classes contributes to total MVPA. Although an impressive body of studies investigating PE motivation and its relations with overall MVPA have been reported, past studies have exclusively relied on self-reports (e.g. Cox, Williams, & Smith, 2007; Fairclough et al., 2012; Gråstén & Watt, 2017; Haerens et al., 2010; Slingerland, 2014; Taylor et al., 2010; Yli-Piipari, 2011). Considering this, a lack of research examining the motivational processes underlying actual in-class MVPA behaviour and its contributions to overall MVPA warrants further studies. Previously, boys have been shown to be more physically active (Tremblay et al., 2016) and to score higher than girls on physical competence (Fairclough, 2003), autonomy (Soini et al., 2007), and intrinsic motivation (Yli-Piipari, 2011), whereas girls have been reported to show higher social relatedness than boys (Gråstén & Watt, 2017). On this basis, to examine gender differences in the motivational processes, the present study adopted a multi-group analysis through structural equation modelling providing parameter estimates and interpretation of parameters to both groups (Hu & Bentler, Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE VOLUME -- | ISSUE - | 2019 | 5 1999). The present study extended previous findings by examining the motivational model of PE including psychological needs, motivation, objective in-class MVPA, and total MVPA (Fig. 1). The key aim was to test the motivational model of PE including needs of competence, autonomy, relatedness, intrinsic and extrinsic motivation, objective in-class MVPA, and total MVPA in school-aged girls and boys. A presumption of the present model was that competence, autonomy, and relatedness would be positively linked with overall MVPA (Cox, Smith, & Williams, 2008; Taylor et al., 2010; Yli-Piipari, 2011) through intrinsic motivation (Lonsdale et al., 2009; Standage, Duda & Ntoumanis, 2005; Vallerand & Lalande, 2011) and inclass MVPA (Gråstén & Watt, 2017; Lonsdale et al., 2009), whereas extrinsic motivation was expected to have negative links to both MVPA scores (Gao, 2012; Gao, Podlog, & Huang, 2013). The second aim was to analyse potential differences between girls and boys in competence, autonomy, relatedness, intrinsic and extrinsic motivation, in-class MVPA and total MVPA. Finally, the contribution of in-class MVPA to total weekly MVPA minutes was determined. METHODS Participants A total sample comprised 490 (264 girls, 226 boys) elementary school children aged between 10to 12-years (M = 11.25 ± .31 years). Children were recruited from 35 public schools across Finland. All Grade 5 students were invited to participate through the direct contact with the school principals. All students who returned signed guardian consents were permitted to participate in the study. Participants were ineligible if they had a medical condition or physical injuries as informed by parents before the commencement of the study. Participation was voluntary and no extra credit was awarded for the participation. Study approval was also obtained from the ethics committee of the local university. Measures Basic Psychological Needs The Finnish version of the Basic Psychological Needs in Physical Education Scale (BPN-PE; Vlachopoulos et al., 2011a) was used to assess the satisfaction of competence, autonomy, and social relatedness in PE. The scale consisted of 12 items referring three subscales: competence (e.g. I am able to do well even in the lessons considered difficult by most kids in my class), relatedness (e.g. My relationships with the other kids in my class are friendly), and autonomy (e.g. We do things in class that interest me). The scale had an item stem “Generally in PE class I feel?” All subscales were measured on five-point scales from (1) totally disagree to (5) totally agree. Vlachopoulos et al. (2011a) reported that the scale had an acceptable model fit in the sample of Greek elementary school students. Intrinsic and Extrinsic Motivation Contextual intrinsic motivation was measured using the Perceived Locus of Causality in Physical Education Scale (PLOC-R; Vlachopoulos et al., 2011b) which was modified for the Finnish school context. The measure consists of two subscales, comprising intrinsic motivation (e.g. I enjoy learning new skills) and extrinsic motivation (e.g. In this way I will not get a low grade). The scale comprised fourteen items and had the item stem “I take part in PE, because?” Each item was rated on a five-point Likert-scale ranging from strongly disagree (1) to strongly agree (5). Previously, the confirmatory factor analysis results supported the acceptable construct validity of the scale in a sample of Greek elementary school students (Vlachopoulos et al., 2011b). Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE 6 | 2019 | ISSUE - | VOLUME -- © 2019 University of Alicante In-class and Total MVPA In-class and total MVPA were measured using triaxial Actigraph GT3X+ accelerometers. Total MVPA was collected across a period of seven days and in-class MVPA was separated from the total MVPA minutes. Students participated in two PE classes (2 x 45 minutes) during the measurement period following the local curriculum. The activity monitors were light, small, easy to use, and were worn on the right waist. The devices collected the data at fifteen-second intervals and summarized minutes spent in MVPA. Only daytime activity (7:00 to 23:00) was included into the analyses apart from engaging in water-based activities. Total MVPA minutes were considered valid if a child had at least three days including one weekend day with at least 500 minutes of activity recorded per day, following the protocol adopted in the earlier studies of Bergh et al. (2011) and Gråstén and Watt (2016). Periods of 30 minutes of consecutive zero counts were defined as non-wearing time (Heil, Brage, & Rothney, 2012). The lower threshold for moderate-intensity was 2296 counts per minute (Evenson et al., 2008). The current device has previously been calibrated for young people in laboratory and free-living conditions (Martinez-Gomez et al., 2012). Procedure The objective MVPA data collection was completed across seven weeks, starting in August and continuing until October 2017. Students concurrently responded to the online questionnaires under the supervision of the teachers in the computer labs. Students were encouraged to answer honestly and assured that their responses were confidential. Children were advised to ask help if needed and they were told that their involvement was optional, and they could terminate their participation at any time without any consequences. Children represented 67 study groups taught by classroom teachers. All teachers were specialized in elementary school education with previous PE teaching experience. All classes focused primarily on developing functional movement skills, i.e. running, jumping, and throwing in different forms of activities (Finnish National Agency for Education, 2014). Data Analysis Prior main analyses, normal distribution, outliers, and missing values were detected. Study variables were approximately normally distributed and did not contain significant outliers based on the standardized values (± 3.00) (Tabachnick & Fidell, 2012). A total of 3.7% missing values in study variables were detected, as 96 students did not provide fully completed questionnaires. Missing completely at random (MCAR) test (  = 26.570, df = 24, p = .325) showed that data with and without missing information were similar, and therefore missing values were expected to be missing completely at random (Little & Rubin, 2002). Correlation coefficients, means, standard deviations, and Cronbach alphas were determined for each variable. The equality of means between girls and boys were tested using the Wald’s test of parameter equality, as a goodness of test is that it only requires the estimation of one model (Williams, 2015). Since this was the first study using the Finnish versions of the BPN-PE and PLOC-R scales, related confirmatory factor analyses were provided to test the factor stuctures of the scales. To test the associations between needs for competence, autonomy, relatedness, intrinsic and extrinsic motivation, in-class MVPA and total MVPA, a path model was implemented. Indirect effects between basic psychological needs and total MVPA variables were tested by setting intrinsic and extrinsic motivation and in-class MVPA as mediators into the model. Gender differences in loadings between variables were examined using two-group tests following the protocol of Muthén and Muthén (2013), in which two nested models can be tested by setting the parameters to be equal. Chi-square test (χ2) was used as a test of the models’ overall goodness-of-fit to the data. A non-significant difference between observed and theoretical distribution had an acceptable fit to the data (Hu & Bentler, Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE VOLUME -- | ISSUE - | 2019 | 7 1999). To determine the appropriateness of the model the standardized root mean square residual (SRMR) and the root mean square error of approximation (RMSEA), the comparative fit index (CFI), the Tucker-Lewis index (TLI) were examined (Hu & Bentler, 1999). A value of .05 or less for SRMR indicate excellent and .08 or less an acceptable magnitude of a varying quantity, and a value of .08 or less for the RMSEA indicate an acceptable fit of the model. The CFI and TLI indices greater than .95 are indicative for an excellent model fit. The MCAR -test for missing values was performed using SPSS Version 22.0 and path models using Mplus Version 8.0. RESULTS Descriptive Statistics First, the correlation coefficients, means, standard deviations, and Cronbach alphas of the study variables were examined (Table 1). The correlations between variables ranged from weak to strong. The strongest positive correlation was found between competence and relatedness in both girls and boys. The Cronbach alphas of the scales were acceptable in both groups. Mean scores indicated that perceptions of competence, relatedness, and intrinsic motivation were relatively high. A series of Wald’s tests revealed that boys had higher mean scores in competence (p < .001), relatedness (p < .01), intrinsic motivation (p < .01), in-class MVPA (p < .05), and total MVPA (p < .001) when compared to girls. Children’s in-class MVPA ranged from 0.25 to 68.75 minutes per weekly PE classes in both girls and boys, whereas girls’ total MVPA varied between 15.51 and 127.89 minutes and boys’ total MVPA between 19.35 and 150.18 minutes per week. In-class MVPA contributed approximately 36% of total weekly MVPA minutes in both girls and boys. Table 1. Correlations, means, standard deviations, and Cronbach alphas of the study variables 1 2 3 4 5 6 7 M (SD) α 1Competence - .45*** .53*** .44*** .30*** .07 .23*** 3.34 (.88) .86 2 Autonomy .50*** - .61*** .61*** .09 .07 .07 2.96 (.82) .73 3 Relatedness .62*** .52*** - .53*** .15* .19** .23*** 3.63 (.84) .80 4 Intrinsic motivation .50*** .54*** .46*** - .17** .03 .13 3.81 (.88) .78 5 Extrinsic motivation .26*** .08 .07 .14* - .03 .02 2.67 (.63) .74 6 In-class MVPA .05 -.04 .09 .03 -.15* - .21*** 19.90 (13.01) - 7 Total MVPA .12 .50 .17* -.02 -.06 .27*** - 55.11 (20.64) - M (SD) 3.64 (.81) 3.08 (.78) 3.84 (.73) 4.05 (.80) 2.74 (.74) 22.64 (14.57) 64.49 (23.65) α .81 .66 .78 .74 .77 - - Note 1. Correlations, means, standard deviations, and Cronbach alphas for girls (n = 264) are presented above and for boys (n = 226) below the diagonal. Note 2. ***p < .001, **p < .01, *p < .05. Confirmatory Factor Analysis Confirmatory factor analyses were conducted to test the factor stuctures of the Finnish versions of the BPNPE and PLOC-R scales. The construct validity of the the BPN-PE scale (χ2(50) = 106.59, p < .001, CFI = .97, TLI = .96, RMSEA = .048, SRMR = .035) was confirmed. However, the fit indices revealed a poor factor structure for the PLOC-R scale (χ2(139) = 476.91, p < .001, CFI = .85, TLI = .82, RMSEA = .071, SRMR = Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE 8 | 2019 | ISSUE - | VOLUME -- © 2019 University of Alicante .063). Based on the modification indices, the residual variances of the individual items “Because it is important to me to do well in PE” and “Because it is important to me to be good in the sports we practice in PE”, “Because I would feel bad about myself if I didn’t” and “Because it would bother me if I didn’t”, and finally “Because I would feel bad if the teacher thought that I am not good at PE” and “Because in this way I will not get a low grade” were allowed to correlate. The modified model had the acceptable factor structure (χ2(135) = 299.85, p < .001, CFI = .93, TLI = .91, RMSEA = .050, SRMR = .056). In addition, Cronbach alphas of both intrinsic and extrinsic subscales were acceptale (Table 1). Based on this, the present scales provided reliable results for the path model development. The Motivational Model of Physical Education First, to examine the associations between need for competence, autonomy, relatedness, intrinsic and extrinsic motivation, in-class and total MVPA participation, both girls and boys were included in the same model. The theorized model revealed an excellent model fit (χ2(1) = 2.57, p = .109, CFI = .99, TLI = .92, RMSEA = .057, 90% CI [.00, .15], SRMR = .011). The standardized model results showed direct paths between competence → intrinsic motivation (β = .19, p < .001), autonomy → intrinsic motivation (β = .39, p < .001), social relatedness → intrinsic motivation (β = .18, p < .001), competence → extrinsic motivation (β = .33, p < .001), in-class MVPA → relatedness (β = .21, p < .001), in-class MVPA → total MVPA (β = .21, p < .001), competence → total MVPA (β = .19, p < .001), autonomy → total MVPA (β = -.16, p < .05), and relatedness → total MVPA (β = .17, p < .01). An indirect path from relatedness to total MVPA via in-class MVPA was detected (β = .04, p < .01). Figure 2. The standardized parameter estimates of the motivational model for girls (n = 264) To find gender differences in associations between competence, autonomy, relatedness, intrinsic and extrinsic motivation, in-class and total MVPA participation, the model was separately tested in girls and boys. The model showed an excellent model fit for the data of girls (χ2(1) = 1.803, p = .179, CFI = .99, TLI = .93, RMSEA = .055, 90% CI [.00, .19], SRMR = .012) (Fig. 2). The standardized results showed direct paths between competence → intrinsic motivation, autonomy → intrinsic motivation, social relatedness → intrinsic Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE VOLUME -- | ISSUE - | 2019 | 15 Layne, T. (2014). Competition within physical education: Using sport education and other recommendations to create a productive, competitive environment. Journal for Sport and Physical Educators 27(6), 3−7. https://doi.org/10.1080/08924562.2014.960124 Levin, S., McKenzie, T., Hussey, J., Kelder, S. & Lytle, L. (2001). Variability of physical activity during physical education lessons across elementary school grades. Measurement in Physical Education and Exercise Science 5(4), 207−218. https://doi.org/10.1207/S15327841MPEE0504_02 Little, R., & Rubin, D. (2002). Statistical analysis with missing data. New York, NY: Wiley. https://doi.org/10.1002/9781119013563 Lonsdale, C., Sabiston, C., Raedeke, T., Ha, A., & Sum, R. (2009). Self-determined motivation and students' physical activity during structured physical education classes and free choice periods. Preventive Medicine 48(1), 69−73. https://doi.org/10.1016/j.ypmed.2008.09.013 Martinez-Gomez, D., Gomez-Martinez, S., Ruiz, J., Diaz, L., Ortega, F., Widhalm, K., Cuenca-Garcia, M., et al. (2012). Objectively-measured and self-reported physical activity and fitness in relation to inflammatory markers in European adolescents: The HELENA Study. Atherosclerosis 221(1), 260−267. https://doi.org/10.1016/j.atherosclerosis.2011.12.032 McKenzie, T., & Lounsbery, M. (2009). School physical education: The pill not taken. American Journal of Lifestyle Medicine 3(3), 219–225. https://doi.org/10.1177/1559827609331562 McKenzie, T., & Marshall, S. (2000). Student activity levels, lesson context, and teacher behavior during middle school physical education. Research Quarterly for Exercise and Sport 71(3), 249–259. https://doi.org/10.1080/02701367.2000.10608905 Mersh, R. & Fairclough, S. (2010). Physical activity, lesson context and teacher behaviours within the revised English National Curriculum for Physical Education: A case study of one school. European Physical Education Review 16(1), 29–45. https://doi.org/10.1177/1356336X10369199 Metcalf, B., Henley, W., & Wilkin, T. (2012). Effectiveness of intervention on physical activity of children: systematic review and meta-analysis of controlled trials with objectively measured outcomes. British Medical Journal 345, 1−11. https://doi.org/10.1136/bmj.e5888 Ministry of Education and Culture (2012). Finnish education in a nutshell − Education in Finland. Espoo: Kopijyvä. Ntoumanis, N., & Standage, M. (2009). Motivation in physical education classes: A self-determination theory perspective. Theory and Research in Education 7(2), 194–202. Nupponen, H., Penttinen, S., Pehkonen, M., Kalari, J., & Palosaari, A-M. (2010). The impact of school physical education study: starting points for the study, methods and description of data. (in Finnish). Turku: University of Turku. Ryan, R. & Deci, E. (2017). Sport, physical activity, and physical education. In R. Ryan & E. Deci (Eds.) Self-determination Theory: Basic Psychological Needs in Motivation, Development, and Wellness (pp. 481–507). New York, NY: The Guildford Press. Sallis, McKenzie, T., Alcaraz, J., Kolody, B., Faucette, N., & Hovell, M. (1997). The effects of a 2-year physical education program (SPARK) on physical activity and fitness in elementary school students. American Journal of Public Health 87(8), 1328–1334. https://doi.org/10.2105/AJPH.87.8.1328 Siedentop, D. & Tannehill, D. (2000). Developing teaching skills in physical education. Mountainview, CA: Mayfield. Slingerland, M. (2014). Physical Education's Contribution to Levels of Physical Activity in Children and Adolescents. Dissertation. Maastricht: Maastricht University. Soini, M., Liukkonen, J., Jaakkola, T., Leskinen, E., & Rantanen, P. (2007). Motivaatioilmasto ja viihtyminen koululiikunnassa [Motivational climate and enjoyment of physical education in school]. Liikunta & Tiede 44(1), 45−51. Gråstén et al. / Accelerometer-based physical activity JOURNAL OF HUMAN SPORT & EXERCISE 16 | 2019 | ISSUE - | VOLUME -- © 2019 University of Alicante Standage, M., Duda, J., & Ntoumanis, N. (2005). A test of self-determination theory in school physical education. British Journal of Educational Psychology 75(3), 411–433. https://doi.org/10.1348/000709904X22359 Tabachnick, B. & Fidell, L. (2007). Using Multivariate Statistics. Boston: Allyn and Bacon. Tammelin, T., Laine, K., & Turpeinen, S. (2013). Physical activity of school-aged children. Research Reports on Sport and Health 272. Jyväskylä: LIKES. Taylor, I., Ntoumanis, N., Standage, M., & Spray, C. (2010). Motivational predictors of physical education students' effort, exercise intentions, and leisure-time physical activity: A multilevel linear growth analysis. Journal of Sport and Exercise Psychology 32(1), 99−120. https://doi.org/10.1123/jsep.32.1.99 Tremblay M, Barnes J, González S, Katzmarzyk P, Onywera V, Reilly J, Tomkinson G, and the Global Matrix 2.0 Research Team. (2016). Global Matrix 2.0: Report card grades on the physical activity of children and youth comparing 38 countries. Journal of Physical Activity and Health 13(2), 343−366. https://doi.org/10.1123/jpah.2016-0594 Vallerand, R. & Lalande, D. (2011). The MPIC Model: The perspective of the hierarchical model of intrinsic and extrinsic motivation. Psychological Inquiry 22(1), 45−51. https://doi.org/10.1080/1047840X.2011.545366 Vanttaja, M., Tähtinen, J., Zacheus, T., & Koski, P. (2017). Tracking of inactivity: Longitudinal study on the changes of physical activity relationship (PAR) among young people who are not very active (in Finnish). Nuorisotutkimusverkosto: Nuorisotutkimusseuta verkkojulkaisuja 115. Vlachopoulos, S., Katartzi, E., & Kontou, M. et al. (2011a). The basic psychological needs in physical education scale. Journal of Teaching in Physical Education 30(3), 263−280. https://doi.org/10.1123/jtpe.30.3.263 Vlachopoulos, S., Katartzi, E., Kontou, M., Moustaka, F., & Goudas, M. (2011b). The revised perceived locus of causality in physical education scale: Psychometric evaluation among youth. Psychology of Sport and Exercise 12(6), 583−592. https://doi.org/10.1016/j.psychsport.2011.07.003 Wallhead, T., Garn, A., & Vidoni, C. (2013). Sport education and social goals in physical education: relationships with enjoyment, relatedness, and leisure-time physical activity. Physical Education and Sport Pedagogy 18(4), 427−441. https://doi.org/10.1080/17408989.2012.690377 Wallhead, T. & Ntoumanis, N. (2004). Effects of a sport education intervention on students' motivational responses in physical education. Journal of Teaching in Physical Education 23(1), 4−18. https://doi.org/10.1123/jtpe.23.1.4 Weiss, M., Corbin, C., & Pangrazi, B. (2000). Motivating kids in physical activity. Research Digest President's Council on Physical Fitness and Sports 3(11). Williams, R. (2015). Imposing and testing equality constraints in models. Available at: https://www3.nd.edu/~rwilliam/stats2/l42.pdf Yli-Piipari, S. (2011). The development of students' physical education motivation and physical activity: A 3.5-year longitudinal study across Grades 6 to 9. Doctoral thesis, University of Jyväskylä. This work is licensed under a Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).