Abstract(resumo):

The objective of this study was analyze longitudinally the association between types of social support provided by different sources and physical activity and assess the moderation of this relationship by the sex and age of adolescents. This is observational longitudinal study (2014 to 2017), with a yearly data collection involving 330 adolescents (50.6% girls), aged between 10 and 13 years in 2014 (starting year), from João Pessoa, Paraíba state, Brazil. Social support was measured using a 5-item scale for each support source (father, mother, friends) that produced emotional and instrumental social support scores. Physical activity was measured by a questionnaire and the time (minutes/week) spent on sports, physical exercise, recreational activities and active displacement was determined. The emotional and instrumental social support provided by the father, mother and friends were positively and significantly associated with time spent on sports, physical exercises, recreational activities, and with active displacement when provided by the mother. Instrumental support provided by friends exhibited a greater magnitude of association with time spent on sports and recreational activities (p ? 0.05) when compared to its emotional counterpart. Adolescents who received greater social, emotional and/or instrumental social support from their parents and friends spent more time on sports, physical exercises and recreational activities over the years.

Keywords(palavra-chave):

Determinants; Social Support; Motor Activity; Students.

Content(conteúdo):

Introduction
Social support, a construct that is part of Sociocognitive Theory1 and Socioecological Models2, can be provided by the father, mother, friends, siblings and teachers3, among other groups, and be emotional (encourage, watch, comment), instrumental (practice together, invite) and/or informational (advice, conversations)4,5. In adolescents, emotional and instrumental support provided by their parents and friends are the most widely investigated5-7.
Cross-sectional6-8 and longitudinal studies9-11 revealed that social support is associated with higher physical activity levels in adolescents. However, most studies analyzed the relationship between the global social support score (combination of the father’s and mother’s support) and total time spent on moderate-to-vigorous physical activity, by domains (leisure, displacement) or their combinations (leisure and displacement)6-11, and few studies were longitudinal9-11.
However, these ways of operationalizing social support and physical activity, and the use of a cross-sectional approach preclude determining the relationship between social support types, and sources vary with the physical activity practiced by adolescents over time. Leisure and displacement physical activities in general explain part of the variation in total time adolescents spend on physical activity12,13, and in leisure, they are involved in diversified activities such as sports, recreation, and physical exercises12. These activities require specific demands in terms of infrastructure, equipment, materials, displacement, motivation and incentive to engage in them14-16. Sport activities may require instrumental and to a lesser extent emotional support14,15, while physical exercises17 and recreational activities16 need emotional support16,17. Active displacement has often been used to move from one place to another in order to meet daily demands and may not need a specific type of support18,19.
In addition, adolescents´ sex and age can be moderated in the relationship between types of social support and physical activity 20. Studies have identified differences in the types of physical activities practiced between the sexes12,21 and age range12 of adolescents. Differences in the types, levels and sources of support between these adolescent subgroups should be considered, given the differences in interpersonal relationships in the physical activity practices of young people21.
Male adolescents are raised with more freedom (running, roughhousing, playing in the street), are considered more skillful and endowed with greater physical ability20 when compared to girls, who are seen as more fragile and encouraged to engage in less intense activities (household chores, and supervised domestic games)20,22. In regard to age, older adolescents spend more time in the company of friends, seek independence from their parents and participate less in collective and sport activities23. By contrast, younger individuals receive more support from their families since they are closer to them than their friends16 and because they are the main source of encouragement to engage in healthy behaviors at this stage of life23.
In this respect, the present study is the first to longitudinally analyze the relationship between types of social support and different physical activities in Brazilian adolescents. As such, the aim of this study was to analyze longitudinally the association between the types of social support provided by different sources and physical activity and assess the moderation of this relationship by the adolescent’s sex and age.

Methods
This is a longitudinal observational study that used LONCAAFS data – Physical Activity, Diet and Adolescent Health. LONCAAFS study aims at analyzing the interrelationships between physical activity level, sedentary behavior, food habits, quality of life, health indicators, and the psychosocial and environmental factors related to the physical activities and sedentary behavior of adolescents aged between 10 and 13 years, studied between 2014 and 2017, with yearly data collections. This study was approved by the Human Research Ethics Committee of the Health Sciences Center of the Federal University of Paraíba (Protocol no 240/13). The mother/father/ legal guardian signed a consent form authorizing the adolescents to take part in the study.
The target population consisted of adolescents aged 10 to 13 years in grade 6 of elementary school at municipal and state schools of João Pessoa, Paraíba in 2014. In order to calculate the sample size, the following aspects were considered: reference population size of 9,520 grade 6 elementary school students in 2014; 50% prevalence of outcomes, 95% confidence interval, maximum acceptable error of 4%; design effect (deff) of two; and 40% increase in sample size to allow for possible losses and refusals, resulting in a sample of 1,582 adolescents.
The sampling process consisted of single-stage conglomerates, where 28 schools (14 state and 14 municipal) were systematically selected, distributed proportionally by geographic region in the municipality of João Pessoa (north, south, east and west zones) and number of grade 6 students. All the students enrolled in grade 6 were invited to participate. The sampling process is described in the flowchart – Figure 1.
LONCAAFS data collection was conducted between February and December, at the school during class time, in all the years of the study (2014 to 2017). A previously trained team following a standardized protocol applied a questionnaire during a 20-minute face-to-face interview and measured height and weight.
The sociodemographic variables were sex, age, class schedule (morning and afternoon) and mother’s education (incomplete elementary, complete elementary, complete secondary or more). In order to determine economic class, the Brazilian Association of Research Companies (ABEP) criteria were followed: presence of material goods, number of domestic employees and the head of the family’s education, classifying the students into high (A/B) and middle/low class (C/D/E)24.
Weight (kg) and height (cm) were measured in triplicate by the same rater, using a digital balance and portable stadiometer, respectively25, and their average value considered as final result. Nutritional status was determined using body mass index (BMI = body mass [kg] / height² [meters]), and classified according to the World Health Organization: no excess body weight (low weight: ? -2 Z scores and normal weight: > -2 to ? +1 z scores) and with excess body weight (overweight: > + 1 to < +2 Z scores and obesity ? +2 Z scores Z)26.
Social support was measured using the Social Support Scale (SSS), which exhibits satisfactory reproducibility, validity and internal consistency levels27. The scale consists of 15 items, five for each support source (father, mother, friends). The adolescents reported the frequency (never = 1, rarely = 2, often = 3 or always = 4) with which their father, mother and friends provide the following forms of support during a normal week: encourage, practice together, transport (for the father or mother), invite (for friends), watch and comment. To follow is an example of one of the scale items: “During a normal week how often do your father/ mother/friends encourage you to engage in physical activity”? Adolescents who were not living with their biological father and/or mother were instructed to consider their grandfather/grandmother, uncle/aunt or older brother/sister if they had assumed the role of parent.
For analysis purposes, emotional and instrumental support scores were produced for each support source: emotional support (encourage, watch and comment), varying from three to 12 points; instrumental support (practice together and transport – father’s and mother’s support – friends’ support), varying from two to eight points. Higher values indicate greater social support.
In order to classify emotional and instrumental support levels in each support source, tertiles were created using the baseline collection results of the study (2014) and applied to the other years (2015, 2016 and 2017). The values observed in the 1st, 2nd and 3rd tertile, respectively were: emotional support – father: 0 to 6.0; 6.1 to 9.0; 9.1 to 12 points; mother 0 to 6.0; 6.1 to 9.0; 9.1 to 12 points; friends: 0 to 7.0; 7.1 to 9.0; 9.1 to 12 points; instrumental support – father: 0 to 3.0; 3.1 to 5.0; 5.1 to 8 points; mother: 0 to 2.0; 2.1 to 4.0; 4.1 to 8 points; friends: 0 to 2.0; 2.1 to 4.0; 4.1 to 8 points. Based on the values found, the adolescents were allocated to two groups: lowest (1st tertile) and highest level of social support (2nd and 3rd tertiles) for emotional and instrumental support scores, in all the years of the study.
Physical activity was measured using the Physical Activity Questionnaire for Adolescents (PAQA), which showed good reproducibility and concurrent validity28. This instrument consisted of 19 physical activities in the leisure and active displacement domains. The adolescents reported if they practiced each of the activities for at least 10 minutes (yes vs. no), as well as the frequency (days/week) and duration (minutes/day) in the week before data collection. Physical activity scores (minutes/week) were produced by adding the products of the frequency x duration of each activity that belonged to the respective sports, as follows: sports (basketball, handball, volleyball, beach volleyball, swimming, soccer, beach soccer, futsal, martial arts, Olympic or rhythmic gymnastics), physical exercises (gym or aerobic exercises, walking, running, weight raining), recreational activities (dancing, cycling, games) and active displacement (cycling or walking to school and other locations).
Excluded from the study were adolescents not within the age range of interest in 2014 (younger than 10 and older than 13 years); interviewees with less than 9 or more than 15 months of data collection from one year to another; those with any disability that prevented or limited their responses to the questionnaire and/or engaging in physical activity, and no response on the social support scale items and/or physical activity, in at least one of the study years.
At the beginning of each data collection year, losses were adolescents who were not located at the school after three visits from the study team, consultations with the administrators and telephone calls to the parents and/or guardians; those that moved to a different city; and those that transferred to difficult-to-access private or public schools.
In the analysis of the present study, adolescents whose variable of interest data were complete during the four years of study were considered. In addition, an analysis was conducted with adolescents whose data were complete at least in 2014 for the variables analyzed (sensitivity analysis).
The variables between the included and excluded adolescents were compared using the Chi-squared and Mann-Whitney tests for qualitative and quantitative variables, respectively (data with non-normal distribution – Shapiro-Wilk test: p ? 0.05).
Unadjusted binary logistic regression was used to analyze the pattern of missing data over the study years. To that end, a dichotomous variable was created: adolescents with complete (missing = 0) and incomplete data in at least 1 year (missing = 1). This variable was related to the first year (2014) for the analyzed variables.
In descriptive analysis, the mean and standard deviation were used for quantitative variables and absolute and relative frequency distribution for their qualitative counterparts. Generalized Estimating Equations (GEE) were used to longitudinally analyze the association between types of emotional and instrumental support provided by the father, mother and friends (independent variables) and duration of different physical activities (dependent variables). Models were created for each social support type and source, in each physical activity. The following parameters were established in all GEE analyses: Gamma family (to distribute independent variable data), Identity link function and Exchangeable correlation matrix. This matrix was used because it is more suitable when the correlation between repeated measures is constant between the subjects (observations equally spaced over the years), and for exhibiting the lowest Quasi-likelihood Under Independence Model Criterion (QIC) in the models tested29. The regression coefficient (?) expresses the average change per year in the dependent variable (duration of each physical activity) for each increase or decline in the independent variable (emotional and instrumental social support). For example, a ? coefficient of 15.28 indicates that for every 1-point gain in emotional social support, an average increase of 15.28 minutes/week is expected in the duration of sports activities per year.
The potential confounding factors analyzed were sex, age, economic class, mother’s education, class schedule and adolescent’s nutritional status. The variables were selected using the Forward method and the final model was composed of those exhibiting statistical significance (p ? 0.05) or changes of at least 10% in the regression coefficients of dependent variables. In order to determine possible differences in the magnitudes of the associations between social support types and sources for the same physical activities, the intersection of the confidence intervals in the associations was considered.
To assess moderation between the types of social support and physical activity practices according to the adolescents´ sex and age, interaction terms were created for all support types and sources, such as sex*father’s social support. Analyses were conducted in Stata 16.0 and a significance level of p ? 0.05 was used.

Results
Of the 1,475 adolescents enrolled in the study in 2014, at the end of the follow-up four years, 330 had complete data for the variables of the present study – Figure 1. It was found that the missing data were classified as random (MAR – Missing at Random) – Supplement I. Excluded from analyses were a larger proportion of 12-13-year-olds from the middle-low economic class, whose mothers had incomplete elementary education, no excess body weight and longer active displacement time – Table 1.
In the baseline data (2014), most of the adolescents were girls (50.6%), 10-11 years old (70.3%) and from the middle-low economic class (52.1%) – Table 1. Over the years, most of the adolescents reported receiving higher emotional and instrumental support from their parents, and greater emotional support from their friends, spending more time on sports, followed by recreational activities – Table 2. Average time spent on all physical activities declined, with a greater magnitude in recreational activities (from 171.41 in 2014 to 71.16 minutes/week in 2017) and lower in active displacement (from 119.76 in 2014 to 107.07 minutes/week in 2017) – Table 2.
In both unadjusted and adjusted analysis, the emotional and instrumental support provided by the father, mother and friends was positively and significantly associated with more time spent on sports, physical exercises and recreational activities. Both types of support provided by the mother were associated with greater active displacement – Table 3.
Significant higher associations were observed only between instrumental support provided by friends and time spent on sports (instrumental support – ? = 21.69; 95%CI: 16.79; 26.58 vs. emotional support – ? = 12.13; 95%CI: 8.58; 15.68) and recreational activities (instrumental support – ? = 19.27; 95%CI: 14.78; 23.76 vs. emotional support – ? = 9.69; 95%CI: 6.07; 13.30) when compared to emotional support. There were no significant differences in the magnitudes of the associations between each support type and physical activity between the different social support sources – Table 3.
Significant interactions were identified between sex and the social support provided by the father for the associations with duration of physical exercises, and by friends with time spent on sport activities; and between sex and instrumental support provided by the mother for the associations with duration of active displacement, and by friends with time spent on recreational activities – Table 3. The longer times male adolescents spent on physical exercises and sports when compared to their female counterparts when they received emotional support provided by their father and friends, respectively. The girls spent more time on active displacement and recreational activities than the boys when they received instrumental support provided by their mother and friends, respectively – Figure 2. On the other hand, associations between the different types and types of social support and physical activity practices were not moderated by age (p > 0.05) – Table 3.
In sensitivity analysis, despite the slight variations in the magnitude of adjusted and raw associations and the moderations between the results of the analyses of adolescents who had complete data in all the years of the study and those with complete data at least in 2014 for the variables of interest, there was no overlap between the confidence intervals – Supplement I.

Discussion
Over the years, adolescents who received more emotional and/or instrumental social support from their parents and friends showed a greater amount of time engaging in sports, physical exercise, and recreational activities, as well as active commuting when the support was provided by the mother. The influence of types of social support on physical activity varied according to the adolescents' sex. Emotional support from the father and friends had a greater influence on the time boys spent engaging in physical exercise and sports, respectively, while instrumental support from the mother and friends had a greater influence on girls´ active commuting and recreational activities, respectively.
Adolescents who received more support from their father, mother, and friends, regardless of the type, spent more time engaging in different physical activities over the years. These results corroborate the findings from longitudinal studies on adolescents9,10, demonstrating that emotional and instrumental social support are determinants of the time spent on different physical activities. Although emotional support consists of encouragement, affection, and understanding5, and instrumental support involves assistance with or providing materials for physical activities, both positively influence different types of physical activity, indicating that social support is a key determinant of physical activity in adolescents.
The hypothesis of the present study, which suggested that types of social support influence physical activity practices differently, was not confirmed. This finding may be due to the social support scale used, which does not measure other forms of instrumental support (paying fees or purchasing materials/equipment) that may be important for the physical activity. As such, its possible influence on different types of physical activity may not have been identified. Additionally, the instrument used to measure physical activity does not consider the context (school vs. out-of-school) in which the activity was performed.
The physical activities practiced by adolescents may have been predominantly performed at school, including physical education classes. Since the adolescents in this study belong to a low-income class, they may have fewer opportunities to engage in physical activities that require transportation to locations, financial resources, and adequate infrastructure to practice outside of school. Additionally, active commuting may be these adolescents´ primary means of travel to and from school. These factors may have contributed to the absence of differences in the magnitude of associations between the types of emotional and instrumental support and the time spent engaging in different physical activities.
Some physical activities performed outside the school setting may require greater social support32. For example, engaging in physical exercise and sports often demands more instrumental (purchasing equipment, paying fees, and transportation to training locations) and emotional support (encouragement and motivation). On the other hand, engaging in recreational activities relies more on invitations and joint participation, which are forms of instrumental support. Thus, considering that these adolescents have lower socioeconomic conditions, they may be less involved in sports and physical exercise due to the demands and resources required for these activities.
If the present study had included adolescents attending private schools, who generally have better socioeconomic conditions, there would have occurred greater variation in contexts and types of physical activities practiced, especially in sports (swimming, basketball, tennis, and volleyball)14 and physical exercises12, which require more instrumental support. These factors could have contributed to identifying differences in the influence of social support types according to the physical activities practiced by adolescents.
Confirming one of the hypotheses of this study, differences were found in the relationships between the types of social support and physical activity practices, depending on the source of support (father, mother, friends). Instrumental support provided by friends had a greater influence on the time spent engaged in sports and recreational activities when compared to its emotional couterpart, but both types provided by the mother influenced active commuting time. The stronger influence of instrumental support from friends on sports and recreational activities may be due to the fact that these activities are mostly collective, and their participation often requires invitations and joint engagement from friends, which are the most common forms of instrumental support they provide6,13. Cheng et al.8 found that physically active friends play an important role in influencing their peers to engage in more physical activity, through behavioral modeling (serving as role models) and/or by providing social support8.
Emotional and instrumental support from the mother was associated with greater active commuting in adolescents, which may be explained by their lower socioeconomic status. For this reason, walking or cycling may be their only means of transportation to school and other locations18. This practice may be influenced by mothers' encouragement (emotional support) for their children to commute actively (on foot or cycling), since it is beneficial to their health, but primarily because these adolescents live close to school and/or lack the financial resources to pay for transportation or do not have access to a motor vehicle31.
The associations between the types and sources of social support and physical activity practices varied in magnitude according to the adolescents' sex. Emotional support from the father and friends had a greater influence on exercise and sports participation in boys than in girls. Exercise and sports are the most commonly practiced activities among male adolescents12,13, and fathers and friends may be the greatest sources of support for these activities. Physical exercise can require significant physical effort10,16, and motivation and encouragement have been identified as forms of emotional support associated with engagement in this practice14,16. Reimers et al. found that emotional support from fathers was associated with more time spent engaging in moderate-to-vigorous physical activity in boys. Sports are collective activities that are common in adolescents' daily lives, especially among boys, who tend to prefer more competitive activities14 and are encouraged by their friends to participate as a way to support one another6.
It was also observed that instrumental support provided by the mother and friends had a greater influence, respectively, on active commuting and recreational activities among girls compared to boys. This may be due to girls´ preference for outdoor activities that do not have fixed rules and do not require specific skills or techniques12. The fact that instrumental support from the mother was more strongly related to active commuting in girls may be explained by mothers' greater concern for their daughters' safety during their commutes. Mothers tend to be more protective of their daughters than their sons34, often accompanying them to places such as parks, squares, and churches35,36. For recreational activities, which are collective in nature and typically take place among the young people themselves, the invitation and/or companionship of friends33 are the most relevant forms of instrumental support they provide.
Adolescents' age did not moderate the relationship between emotional and instrumental support from parents and friends and the physical activities analyzed, demonstrating that both types of support seem to exert the same level of influence on engagement in physical activities, irrespective of age. These findings may, in part, be explained by the small age difference between the adolescents in this study (10 to 13 years old in 2014 and 13 to 16 years old in 2017), which would result in only slight changes in social relationships as well as similarities in attitudes and physical activity preferences.
The results of this study suggest that interventions to increase physical activity, regardless of the type, should consider planning and developing actions to increase the emotional and instrumental support from fathers, mothers, and friends. However, further longitudinal studies are needed to determine potential variations in the relationship between types of social support and different forms of physical activity (sports, physical exercise, recreational activities, and active commuting) in adolescents across a broader age range (10 to 19 years old) and different socioeconomic levels, followed over prolonged periods. Additionally, it is important to validate social support scales that include more items assessing emotional, instrumental, and informational support, provided separately by different sources (fathers, mothers, friends, and physical education teachers).
Among the limitations of this study, the following stand out: potential selection bias, given that adolescents with higher active commuting times were excluded from the analyses; however, only social support from the mother was associated with active commuting time. Additionally, the social support scale did not include items on informational support, and the questionnaire used to measure physical activity did not distinguish between activities performed during and outside of physical education classes. Furthermore, only students from public schools were included, who generally come from families with lower socioeconomic status. This may have influenced the social support provided by parents and friends and the types of physical activities practiced, impacting the relationships between these variables.
Although the study had a high attrition rate (with a response rate of 22.5% after four years), significant differences were observed in age, economic class, mother´s schooling level, and nutritional status. However, these losses were classified as random (Supplement I) and were not associated with instrumental and emotional social support scores. Sensitivity analysis showed that the results were similar to those obtained with the study sample, suggesting that there was no bias and/or that the sample size was sufficient to test the hypotheses. Finally, the attrition rate was similar to that of other longitudinal studies with adolescents in the same age range9,37.
The main strengths of this study were the use of a longitudinal design to assess possible changes in the influence of different sources of support on physical activity practices; the analysis of different types of social support (emotional and instrumental) and support sources (father, mother, friends) with different types of physical activity; the use of scales with satisfactory levels of validity and reproducibility; and the standardized protocol followed by a trained team during data collection.
Adolescents who perceive higher levels of emotional and/or instrumental support from their father, mother, and friends tend to engage more in sports, physical exercise, and recreational activities over the years. The time spent on active commuting increased when adolescents received social support from their mothers, regardless of the type. Instrumental support from friends contributed more to the increase in time spent on sports and recreational activities. Boys engaged in physical exercise for longer when receiving emotional support from their father and spent more time playing sports when support came from their friends, when compared to girls. Girls engaged in more active commuting when they received instrumental support from their mother and spent more time on recreational activities when receiving support from friends.





Conflicts of Interest
The authors declare no conflicts of interest.

Contributors
Moura IRD, Cabral TG, Mendonça G, and Farias Júnior JC participated in all stages of the study, including data collection and analysis, manuscript writing, and revision. Florindo AA, Rech CR, and Camargo EM contributed to manuscript writing and revision.

Funding Source
This study was funded by the National Council for Scientific and Technological Development (CNPq) (protocol: 486306/2012-7) and the Research Support Foundation of the State of Paraíba (FAPESQ) (protocol: 460887/2014-9).

Acknowledgments
We thank the State Department of Education of Paraíba and the Municipal Department of Education and Culture of João Pessoa for their authorization, and the public school students who participated in the LONCAAFS Study. We also thank the National Council for Scientific and Technological Development (CNPq) and the Research Foundation of the State of Paraíba (FAPESQ) for providing financial resources for the LONCAAFS Study and the Research Productivity Fellowship awarded to Professor José Cazuza de Farias Júnior. Alex Antônio Florindo is grateful for the funding received from the National Council for Scientific and Technological Development (CNPq) (protocol: 309301/2020-3).




References
1. Plotnikoff RC, Costigan SA, Karunamuni N, Lubans DR. Social cognitive theories used to explain physical activity behavior in adolescents: a systematic review and meta-analysis. Int J Prev Med. 2013;56(5):245-53.
2. Bauman AE, Reis RS, Sallis JF, Wells JC, Loos RJ, Martin BW. Correlates of physical activity: why are some people physically active and others not? Lancet. 2012;380(9838):258-71.
3. Uijtdewilligen L, Nauta J, Singh AS, et al. Determinants of physical activity and sedentary behaviour in young people: a review and quality synthesis of prospective studies. Br J Sports Med. 2011;45(11):896-905.
4. Laird Y, Fawkner S, Kelly P, McNamee L, Niven A. The role of social support on physical activity behaviour in adolescent girls: a systematic review and meta-analysis. Int J Behav Nutr Phys Act. 2016;13(1):79.
5. Duncan SC, Duncan TE, Strycker LA. Sources and types of social support in youth physical activity. Health Psychol. 2005;24(1):3.
6. Mendonça G, Farias Júnior JC. Physical activity and social support in adolescents: analysis of different types and sources of social support. J Sports Sci. 2015;33(18):1942-51.
7. Piola TS, Bacil EDA, Silva MP, Pacifico AB, Campos W. Associação entre apoio social e nível de atividade física em adolescentes. Rev Bras Ativ Fís Saúde. 2018;23:1-10.
8. Cheng LA, Mendonça G, Farias Júnior JC. Physical activity in adolescents: analysis of the social influence of parents and friends. J Pediatr. 2014;90(1):35-41.
9. Kirby J, Levin KA, Inchley J. Parental and peer influences on physical activity among Scottish adolescents: a longitudinal study. J Phys Act Health. 2011;8(6):785-93.
10. Lau E, Faulkner G, Qian W, Leatherdale S. Longitudinal associations of parental and peer influences with physical activity during adolescence: findings from the COMPASS study. Am J Health Promot. 2016;36(11):235.
11. Dowda M, Dishman RK, Pfeiffer KA, Pate RR. Family support for physical activity in girls from 8th to 12th grade in South Carolina. Prev Med. 2007;44(2):153-9.
12. Mendonça G, Cheng LA, Farias Júnior JC. Padrões de prática de atividade física em adolescentes de um município da região Nordeste do Brasil. Cien Saude Colet. 2018;23:2443-51.
13. Lopes MVV, da Silva KS, Bertuol C, da Silva JA, Nahas MV. Tipos e quantidades de atividades físicas praticadas por adolescentes do sul do Brasil. Pensar Prát. 2018;21(3).
14. Silva S, Knuth A, Duca G, et al. Individual and collective sports practice and associated factors among adolescents belonging to a birth cohort study. Rev Bras Educ Fís Esporte. 2009;23(3):263-74.
15. Junior MRDA, Araújo CLP, Pereira FM. Atividades físicas e esportivas na adolescência: mudanças de preferências ao longo das últimas décadas. Rev Bras Educ Fís Esporte. 2006;20(1):51-8.
16. Bengoechea EG, Sabiston CM, Ahmed R, Farnoush M. Exploring links to unorganized and organized physical activity during adolescence: The role of gender, socioeconomic status, weight status, and enjoyment of physical education. Res Q Exerc Sport. 2010;81(1):7-16.
17. Gonçalves VO, Martínez JP. Imagem corporal de adolescentes: um estudo sobre as relações de gênero e influência da mídia. Comun & Inf. 2014;17(2):139-54.
18. Camargo EM, Santos MPM, Ribeiro AGP, Mota J, Campos W. Interação dos fatores sociodemográficos na associação entre fatores psicossociais e transporte ativo para a escola. Cad Saude Publica. 2020;36:e00102719.
19. Carlson JA, Sallis JF, Kerr J, et al. Built environment characteristics and parent active transportation are associated with active travel to school in youth age 12–15. Br J Sports Med. 2014;48(22):1634-9.
20. Gonçalves H, Hallal PC, Amorim TC, Araújo CL, Menezes A. Fatores socioculturais e nível de atividade física no início da adolescência. Rev Panam Salud Publica. 2007;22(4):246-53.
21. Cowley ES, Watson PM, Foweather L, et al. “Girls Aren’t Meant to Exercise”: perceived influences on physical activity among adolescent girls—The HERizon Project. J Children. 2021;8(1):31.
22. Venturini GRO, Guerra VH, Rodrigues BM, et al. Gênero e Educação Física Escolar. J EF Deportes-Revista Digital, Buenos Aires. 2010;15(147).
23. Seabra AF, Mendonça DM, Thomis MA, Anjos LA, Maia JA. Determinantes biológicos e sócio-culturais associados à prática de atividade física de adolescentes. Cad Saúde Pública. 2008;24:721-36.
24. ABEP – Associação Brasileira das Empresas de Pesquisa. Critério de Classificação Econômica Brasil. 2013.
25. Lohman TG, Roche AF, Martorell R. Anthropometric Standardization Reference Manual. Champaign: Human Kinetics. 1988
26. WHO. Global strategy on diet, physical activity and health. World Health Organization. 2004.
27. Mendonça G, Prazeres Filho A, Crochemore-Silva I, Farias Júnior JCd. Reprodutibilidade, validade e consistência interna das escalas de apoio social e autoeficácia para atividade física em adolescentes de 10 a 14 anos de idade. Rev Paul Pediatr. 2021;40.
28. Prazeres Filho A, Barbosa AO, Mendonça G, Farias Júnior JC. Reproducibility and concurrent validity of the Physical Activity Questionnaire for Adolescents (QAFA) aged 10-14 years. Rev Bras Cineantropom Desempenho Hum. 2017;19(3):270-82.
29. Cui J. QIC program and model selection in GEE analyses. The Stata Journal, 7 (2), 209–220. 2007.
30. Due P, Holstein B, Lund R, Modvig J, Avlund K. Social relations: network, support and relational strain. Soc sci med. 1999;48(5):661-73.
31. Thuany M, Santos FK, Almeida MB, Gomes TNQF. Deslocamento ativo em adolescentes: prevalência e preditores associados ao trajeto casa-escola. Rev Bras Ativ Fís Saúde. 2021; 26:1-8.
32. Reimers AK, Schmidt SC, Demetriou Y, Marzi I, Woll A. Parental and peer support and modelling in relation to domain-specific physical activity participation in boys and girls from Germany. J PLoS One. 2019;14(10): e0223928.
33. Prado CV, Lima AV, Fermino RC, Añez CRR, Reis RS. Apoio social e prática de atividade física em adolescentes da rede pública de ensino: qual a importância da família e dos amigos? Cad Saude Publica. 2014; 30:827-38.
34. Christian HE, Klinker CD, Villanueva K, et al. The effect of the social and physical environment on children’s independent mobility to neighborhood destinations. J Phys Act Health. 2015;12(s1): S84-S93.
35. Esteban-Cornejo I, Carlson JA, Conway TL, et al. Parental and Adolescent Perceptions of Neighborhood Safety Related to Adolescents' Physical Activity in Their Neighborhood. Res Q Exerc Sport. 2016;87(2):191-9.
36. Silva A, Fermino RC, Souza CA, Lima AV, Rodriguez-Añez CR, Reis RS. Socioeconomic status moderates the association between perceived environment and active commuting to school. Revista de saude publica. 2018; 52:93.
37. Barnett LM, Morgan PJ, van Beurden E, Beard JR. Perceived sports competence mediates the relationship between childhood motor skill proficiency and adolescent physical activity and fitness: a longitudinal assessment. Int J Behav Nutr Phys Act. 2008;5(1):1-12.