Factors contributing to the uptake and maintenance of regular exercise behaviour in emerging adults

Abstract

Objective:

To identify the influence of parental autonomy support, basic need satisfaction and motivation on emerging adults’ physical activity level and exercise behaviours.

Design:

Cross-sectional survey.

Setting:

This study convenience-sampled approximately 435 college students identified as emerging adults – aged 18–25 years, who did not have a child, own a home, or have sufficient income to be fully independent.

Methods:

Survey responses were used in a path model to investigate how parental autonomy support, psychological mediators and motivational processes influenced emerging adults’ exercise behaviour.

Results:

The hypothesised model was supported with minor modifications. Most notable was the influence of parental physical activity level and autonomy support on psychological mediators, motivational processes and exercise behaviour.

Conclusion:

Results indicate that parents influence their children both directly and indirectly. The impact of autonomy and competence support was found to promote emerging adults’ intrinsic motivation, which consequently influenced actual physical activity and behaviour.

Keywords

College students motivational factors parental autonomy support physical activity self-determination theory USA

Introduction

Many adult health behaviours are established during the period from late adolescence to early adulthood, a time of great change when the effects of exercise and physical activity (PA) are needed the most (Boyle and LaRose, 2008; Buckworth and Nigg, 2004). During the transition from childhood to adulthood, young people establish patterns of behaviour and make lifestyle choices that affect both their current and future health; some also struggle to adopt behaviours that could decrease their risk of developing chronic diseases in adulthood (Gray, 2011). Thus, understanding why individuals participate in and adhere to exercise has attracted considerable attention in exercise science research (Vlachopoulos and Michailidou, 2006; Wilson et al., 2006). The benefits of regular exercise for reducing obesity and related chronic diseases are well known, yet many individuals interested in exercise do not follow through with their actions that lead to adherence (Ryan et al., 2008). Epidemiological evidence indicates the level of PA actually declines from high school to college; to be specific, only 38% of college students participate in regular vigorous activity and only 20% participate in regular moderate activity (United States Department of Health and Human Services [USDHHS], 2010). Therefore, further research on influences on PA and exercise are needed to provide information that may be useful in more effectively promoting active lifestyles in individuals or groups of young people.

Public health researchers remain most interested in social influences because it is believed that these can be manipulated through intervention to change behaviour – adherence is known to improve significantly when exercisers have strong encouragement and support (Daley and Duda, 2006; Wilson and Rogers, 2008). Individuals are also more likely to adopt values and behaviours promoted by those whom they feel connected and whom they trust, where conditions supporting an individuals’ experience of autonomy and competence foster the most volitional and high-quality forms of motivation and engagement in activity (Welk et al., 2003). In order to advance our understanding of the motivational determinants of exercise behaviour change, it is important to note that initiation and adherence to exercise have been conceptualised as multi-dimensional and dynamic (Farmanbar et al., 2009).

As a form of social influence, parents or parent-like figures can significantly impact exercise behaviours through a variety of mechanisms which include parental encouragement, beliefs, attitudes and behaviours (Weiss, 2000). Parents who afford autonomy and support competence are shown to increase autonomy-related outcomes of PA and exercise, including self-esteem and self-regulation of behaviour (Edwardson and Gorley, 2010). However, the degree to which any of these psychological needs is unsupported or thwarted within a given social context may have a detrimental impact on behaviour in that setting (Ryan and Deci, 2000). Theorists suggest that cognitive and physical developmental changes may further effect parent socialisation and motivation for engagement in PA and exercise; unfortunately, the majority of research on family socialisation has not yet adequately considered developmental changes that occur within the emerging adult and parent relationship (Gruber, 2008).

The need for more parent-based exercise interventions is frequently cited in the literature (Scarpa and Nart, 2012). Evidence suggests that theories of motivation and behaviour change, such as self-determination theory (SDT) and transtheoretical model (TTM), are at the forefront of research investigating the psychological antecedents and bases for intervention in exercise settings (Ryan et al., 2009).

Even with advances in research concerning these theoretical frameworks, it has been difficult to determine the importance of various sources of psychological, social, environmental and cultural influences on the emerging adult population up to this point. Epidemiological evidence indicates a fuller understanding of health-related behaviours in emerging adulthood is necessary for the development of strategies that encourage lifetime adherence to exercise (Kim, 2006). The adoption and maintenance of healthy behaviour in this period is critical. In fact, it has been established that exercise adherence improves significantly when exercisers receive strong parental encouragement (Wilson et al., 2003). However, no such research has examined the relationship between parental behaviour, encouragement and involvement in exercise in emerging adulthood – a time when the benefits of regular exercise and PA are needed the most.

The purpose of this study was to use path modelling techniques to determine the influence of SDT- and TTM-based constructs on the uptake and maintenance of regular exercise behaviour in emerging adults, more specifically, the influence of parental autonomy support, basic need satisfaction and motivation on PA behaviour and individual stage-of-change (SOC) for exercise behaviour. Studying the constructs within these theories together may help to better explain exercise behaviours within the emerging adult population.

In this study, social factors contributing to the uptake and maintenance of regular exercise behaviour in emerging adults were examined to test the following hypotheses: (1) high levels of perceived parental autonomy support and warmth would predict emerging adults’ exercise behaviour and meeting PA guidelines; (2) having a parental figure currently meeting PA guidelines would predict emerging adults’ exercise behaviour and meeting PA guidelines – failing to meet guidelines would have implications for the increased prevalence of obesity and related chronic diseases (USDHHS, 2008). Other psychological and motivational factors contributing to the uptake and maintenance of regular exercise behaviour in emerging adults were also examined to test the hypotheses: (3) high levels of autonomy and perceived competence would predict emerging adults’ exercise behaviour and meeting PA guidelines; (4) high levels of intrinsic motivation, identified and introjected regulation would predict emerging adults’ exercise behaviour; and (5) high levels of intrinsic motivation and introjected regulation would predict emerging adults’ meeting or not meeting PA guidelines. An illustration of the hypothesised model is contained in Figure 1.

Figure 1.

Hypothesised model predicting uptake and maintenance of regular exercise behaviour in emerging adults.

Methods

Sample

Utilising a cross-sectional design, we obtained a convenience sample of US college students aged between 18 and 25 years, who did not have a child, own a home, or have sufficient income to be fully independent. Participants were enrolled in a Healthful Living education course at a southeastern US university either online or face-to-face. Participation was voluntary, with no incentive or punishment given. Although participants were recruited through this one class, the demographics of the sample closely match those of the university as a whole and the course was required of all students enrolled in the university. Of the 1,147 students enrolled, 435 participated, giving a 38% response rate. Table 1 details the sample demographic characteristics.

Table 1.

Frequency and percentiles of demographic characteristics of student participants.

Variable	n	%	Variable	n	%
Gender (n = 435)			Age (n = 434), years
Male	195	44.8	18	237	54.6
Female	240	55.2	19	130	29.9
Race (n = 435)			20	52	12.0
White	275	63.2	21–25	25	5.8
Black	118	27.1	Parental influence (n = 410)
Hispanic	9	2.1	Mother	233	56.8
Asian	8	1.8	Father	147	35.9
Bi-racial/other	25	5.7	Equal	11	2.7
			No Influence	19	4.6

Measures

The perceptions of parents scale (POPS) was used to measure the extent to which parental figures afford autonomy and relatedness satisfaction in college students (Grolnick et al., 1997). This scale determined emerging adult perceptions of their parental figure’s influence on their overall health and well-being. The scale has 21 items for participants to describe their thoughts about their influential parental figure using a 7-point Likert scale ranging from 1 ‘not at all true’ to 7 ‘very true’. The scale has been validated in a sample of college students (Robbins, 1994). From these items, three subscale scores were calculated: autonomy support, involvement and warmth. The POPS measure returned a Cronbach’s alpha level of involvement, .84; autonomy support, .87; and warmth, .88.

The Basic Psychological Needs Satisfaction in Exercise Scale (BPNES) was used to assess the extent to which the psychological needs of participants were fulfilled in an exercise setting (Vlachopoulos et al., 2010). The scale, which has been found to be valid and reliable in an adult population, comprises 12 items divided into three subscales, with four items per subscale, to assess: autonomy (e.g. ‘The way I exercise is in agreement with my choices and interests’), competence (e.g. ‘I feel I perform successfully the activities of my exercise programme’) and relatedness (e.g. My relationships with the people I exercise with are close’). Responses were provided on a 5-point Likert scale ranging from 1 ‘don’t agree at all’ to 5 ‘completely agree’. Scores were calculated by finding the sum of responses for each subscale and dividing that number by the number of questions represented. The scale was used to refer to an individual’s overall experiences in exercise as opposed to any particular situation. The BPNES measure returned a Cronbach’s alpha level of autonomy, .78; competence, .86; and relatedness, .89.

In addition to the BPNES, the perceived competence scale (PCS) for regular exercise was used to identify participants’ overall perceived competence to exercise (Williams et al., 1998). This scale was used to supplement the BPNES because of the differences in measurement instruments and research designs that have been used in the past; it has been difficult to determine the importance of various sources of perceived competence in activity. The PCS is a short four-item questionnaire that was used to assess the degree to which emerging adults felt confident about being able to begin or permanently maintain a regular exercise regimen. This questionnaire asks participants to indicate the extent to which each statement was true for them using a 7-point Likert scale ranging from 1 ‘not at all true’ to 7 ‘very true’. An individual’s score is simply the average of his or her responses to the four PCS items. The PCS measure returned a Cronbach’s alpha level of .95. Previous studies have shown good validity and reliability among samples of adolescents and young adults (Williams and Deci, 1996).

The exercise self-regulation questionnaire (SRQ-E) addressed the reasons why a person exercises regularly (Williams et al., 1996). It has been previously validated in a sample of older adults. In the SRQ-E, participants were asked a total of 12 statements/reasons for exercising that required a response using a 7-point Likert scale with 1 ‘not at all true’ and 7 ‘very true’. Participants provided responses that represented external regulation, introjected regulation, identified regulation and intrinsic motivation. This questionnaire does not address amotivation and therefore does not account for a lack of motivation being indicative of regular exercise, a potential limitation to this particular measurement. However, the SRQ-E was used in this study to identify individual differences in the types of motivation or regulation. Each question asked why the respondent participates in a behaviour (or class of behaviour) and then provided several possible reasons that have been preselected to represent the different styles of regulation or motivation. Each participant received a score on each subscale by averaging responses to each of the items that make up that subscale – for example, the average of all items representing introjected regulation would represent the score for that subscale. The SRQ-E measure returned a Cronbach’s alpha level: external regulation, .76; introjected regulation, .60; identified regulation, .81; and intrinsic motivation, .86.

The SOCs model adapted for exercise behaviour suggests progression through six SOCs: precontemplation (no intention to exercise), contemplation (some intention to exercise), preparation (exercise some, but not regularly), action (exercise regularly, but for less than 6 months), maintenance (exercise regularly for longer than 6 months) and termination (exercise regularly for longer than 5 years) representing the sixth stage and exit from the cycle (Prochaska et al., 1994). Regular exercise for this scale was defined as participating in voluntary PA 3 or more days a week for 30 minutes or longer each time, at a level that increases heart rate and/or breathing rate. Participants were asked to place themselves into the stage corresponding to the one item they endorsed most strongly; participants were allowed only one response to represent their current SOC for exercise behaviour. The SOC model was additionally used in this study by having participants categorise their reported influential parental figure into an SOC for exercise behaviour.

The PA guidelines were derived from an evidence-based report on the health benefits of PA, written by the PA Guidelines Advisory Committee for the USDHHS (2008). This includes engaging in at least 150 minutes of moderate intensity or 75 minutes of vigorous intensity PA per week, using a variety of aerobic and muscle-strengthening activities. Muscle-strengthening activities are recommended to be completed on 2 or more days per week.

Statistical analysis

Frequency counts, mean values and standard deviations were used to gather statistics on demographic characteristics such as age, gender and ethnicity in both emerging adults and reported parental figure. After checking for any violation of statistical assumptions, correlations were run to determine potential relationships between variables to be loaded into the path model. SPSS AMOS v. 21 (SPSS, Inc.) was used to analyse the fit and relationships within the hypothesised model (Thompson and Green, 2006). Using maximum likelihood estimates, several fit indices were chosen to assess model fit: chi-square statistic/degrees of freedom (DF) (Satorra and Bentler, 1994), goodness-of-fit index (GFI) (Tanaka and Huba, 1989), root mean square error of approximation (RMSEA) and adjusted goodness-of-fit index (AGFI) (Schumaker and Lomax, 2004). For GFI, good fit was determined using .95 as a cutoff score. RMSEA was used to determine the model’s closeness to reality, with a cutoff score of .05 indicating good fit. Finally, AGFI determined model fit based on DF, with a cutoff of .9 indicating good fit. Parameter estimates were also examined to determine significance, whereby values exceeding 1.96 were considered significant (Hoyle and Panter, 1995). Although only the standardised parameter estimates are reported, both unstandardised and standardised parameter estimates were examined. Direct and indirect effects were evaluated in addition to R² values.

Results

Descriptive statistics and scale reliability

The ratio of participants to variables for the model was acceptable at approximately 23:1 (Bentler and Chou, 1987). The results of the path model analysis indicated poor fit (χ²(56, N = 273) = 1,057.67, p < .001; GFI = .66, RMSEA = .26, AGFI = .36). The original model was then modified using the reported modification indices and examination of pertinent literature. Initially, paths were drawn from perceptions of parental PA to all three basic needs. Modification indices suggested removing the path to relatedness in addition to parental warmth’s path to relatedness. In addition, perceptions of parental stage of change and involvement were completely removed from the model. Other deletions included the path from relatedness to all forms of regulation; the path from perceived competence to extrinsic, introjected and identified regulation; the path from autonomy to introjected regulation; as well as the path from extrinsic motivation to stage of change and PA behaviour. While related to stage of change, identified regulation was not significantly related to PA behaviour and was also removed. Additional paths were suggested between parental warmth and autonomy support, parental PA and PA behaviour, autonomy and competence, autonomy and relatedness, identified regulation and intrinsic motivation, introjected regulation and identified regulation, PA behaviour and stage of change, as well as a direct path from competence to PA behaviour. These modifications produced a much better fitting model (χ²(36, N = 273) = 34.80, p = .53; GFI = .98, RMSEA = .00, AGFI = .96).

The new model, located in Figure 2, confirmed the hypothesis that parental PA, warmth and autonomy support were significantly and directly related to autonomy (γ = .18, .19, .49, respectively) and perceived competence (γ = .11, .18, .28, respectively). In addition, autonomy was significantly and directly related to identified regulation (γ = .31) and intrinsic motivation (γ = .13) as was perceived competence to intrinsic motivation (γ = .22). Identified regulation and intrinsic motivation were, in turn, directly related to both stage of change (γ = .17 and .21, respectively), with intrinsic motivation also being directly related to PA behaviour (γ = .17). The hypothesis that parental PA, warmth and autonomy support were connected to relatedness was disconfirmed, as was the hypothesis that parental stage of change and involvement were related to any items in the model. Relatedness was also not directly related to any forms of self-regulation. It was also interesting to note that autonomy and perceived competence were not related to extrinsic and introjected regulation. Extrinsic motivation in this model was not related to either PA behaviour or stage of change, while identified regulation was not related to PA behaviour.

Figure 2.

Path model predicting uptake and maintenance of regular exercise behaviour in emerging adults. All paths included in the figure are significant at p < .05. Congruent with assumptions, all variables were allowed to correlate. Errors are not given in the model for sake of clarity.

Indirect relationships were also found in the model, including parental autonomy support to relatedness, extrinsic motivation, introjected and identified regulation, intrinsic motivation, PA behaviour and stage of change (Table 2). Parental warmth was similarly related to those items, with the exception of relatedness. Parental PA behaviour was also indirectly related to all constructs listed above. Individuals’ feelings of autonomy were indirectly related to extrinsic motivation, introjected and identified regulation, intrinsic motivation, PA behaviour and stage of change. These relationships were also seen via perceived competence and other items listed above, with the exception of introjected regulation. These described relationships and their standardised parameter estimates can be found in Figure 2. Variance accounted for in the model was as follows: warmth, R² = .64; autonomy, R² = .15; perceived competence, R² = .33; relatedness, R² = .36; extrinsic motivation, R² = .43; introjected regulation, R² = .04; identified regulation, R² = .58; intrinsic motivation, R² = .58; PA behaviour, R² = .29; and stage of change, R² = .45.

Table 2.

Indirect effects.

	1	2	3	4	5	8	9	10
1. Parental autonomy support
2. Parental warmth
3. Parental PA behaviour
4. Autonomy
5. Competence
6. Relatedness	.19**		.16*
7. Extrinsic motivation	.14**	.10**	.15**	.19**	.16**
8. Introjected regulation	.07**	.07**	.09**	.14**
9. Identified regulation	.32**	.26**	.35**	.37**	.05**
10. Intrinsic motivation	.35**	.29**	.39**	.70***	.26**	.16**
11. PA behaviour	.06**	.05**	.07**	.13**	.04**	.01**	.04**
12. Stage of change	.23**	.20**	.39**	.54**	.24**	.16**	.11**	.04*

PA: physical activity.

p < .05, **p < .01, ***p < .001.

Comment

This study was designed to extend the current literature by determining the degree of association between SDT- and TTM-based constructs on factors contributing to the uptake and maintenance of regular exercise behaviour in emerging adults. This study sought to determine the influence of parental autonomy support, basic need satisfaction and motivation on PA behaviour and individual stage of change for exercise behaviour using a path model.

Overall, deletion of specific paths in the model is supported in the literature. This includes the fact that in previous research, evidence supports a more significant link between autonomy and competence satisfaction than relatedness to both well-being and more self-determined motives for health-related behaviour (McEloy, 2002). In general, although many parental variables were significant predictors of emerging adults’ PA levels, the magnitude of parental influence to relatedness was generally small as was the emerging adults’ level of relatedness need satisfaction on motivation. This phenomenon was also demonstrated in the development process and initial evaluations of the BPNES, where the relationship with the motivational outcome was more strongly correlated with autonomy and competence rather than relatedness (Gunnell et al., 2013). Similar to recent findings, this study found emerging adults of active parents had higher overall scores on the parental influences measures and psychosocial correlates than inactive parents.

In the past, individuals experiencing higher levels of competence have been more likely to make and maintain healthy exercise behaviour change and to demonstrate positive health outcomes over time. Yet, perceived competence has only been shown to be a key factor in fostering intrinsic motivation – the most desirable and complete form of motivation for engagement in activity. Because extrinsic and introjected regulations are not strongly related to the process of integration, individuals’ motivation for engaging in exercise becomes less controlled and more autonomous over time (Van de Berghe et al., 2012). Therefore, it was expected in this study that higher self-determined regulations would be significantly associated with autonomy. Results indicate autonomy was more significantly associated with identified regulation of exercise behaviour than any other regulations, with intrinsic motivation and external regulation being equally associated.

Equally notable, researchers have found that regular exercisers endorse all other regulations more strongly than extrinsic motivation – that is, exercising due to outside or external pressures. It is apparent that regular exercisers tend to endorse self-determined regulations more strongly than controlled regulations; non-exercisers endorse less self-determined regulations (extrinsic and introjected regulations) to a higher degree than do exercisers and endorse self-determined regulations (identified and intrinsic) to a lesser degree than regular exercisers (Hagger and Chatzisarantis, 2007). This study confirmed previous findings by showing that those individuals who endorsed more self-determined regulations tended to meet the recommended PA levels, according to American College of Sports Medicine (ACSM) guidelines.

Additional paths were suggested between parental warmth and autonomy support to emerging adults’ exercise behaviour; these modifications produced a much better fitting model. A meta-analysis of construct associations revealed a climate in which parents are supportive may influence variables like basic psychological needs, motivation and enjoyment in exercise. Researchers found a warm climate positively predicted needs satisfaction which in turn positively predicted self-determined motivation to exercise; mother PA, father PA, overall support and attitude all showed positive relationships with overall PA (Landry and Solomon, 2004). While this study found that parental warmth did significantly predict autonomy and competence needs satisfaction in emerging adults, this relationship was not as strong as other social factors. Therefore, although a warm climate may have been significant in past examinations, this study suggests other social factors such as parental PA level, parental involvement and autonomy support may predict exercise behaviour more significantly.

Research further suggests, in accordance with SDT, that increasing an individual’s options and choices should increase their feelings of autonomy and competence for activity, including exercise (Fallon et al., 2005; Sheldon and Filak, 2008). Thus, when considering emerging adults living at or away from home, it may be important to consider whether or not they have a ‘choice’ in their daily activities. Because emerging adults are in a transitional period associated with breaking free from constraints and resisting coercion by being independent, autonomy may play a larger role in other areas of skill and life development through its association with perceived competence. Hence, social environments that afford autonomy are also more likely to support competence and enhance health-behavioural outcomes in emerging adults (Arnett, 2012). This study found that parents who afforded autonomy and support were also shown to increase autonomy-related outcomes including competence, relatedness, self-esteem and self-regulation of behaviour during the formative years. Justifiably, the path from autonomy to relatedness was deemed important and significant, and added to the path model.

Finally, the four different types of behavioural regulation are defined in terms of the degree to which the regulation of an extrinsically motivated activity has been internalised and integrated. In this study, investigators found that participants had strong identified regulation in addition to extrinsic motivation. Ultimately, these regulations produced a much better fitting model. It should be noted that self-determined motives are underpinned by valuing (identified) and enjoying (intrinsic) activities as opposed to performing them for reasons of self-imposed pressure (introjected) or separable rewards (extrinsic) (Ryan and Deci, 2007). Thus, this study wanted to display that individuals who report more self-determined motives also report more regular PA, as well as more positive physical and psychological outcomes of PA and exercise participation. This study assumed that individuals can move through a series of SOCs for exercise behaviour, beginning at living a sedentary lifestyle to regularly maintaining an active lifestyle.

Limitations

Previous research examining TTM- and SDT-based constructs in the exercise domain has been limited by use of self-report measures exclusively and inconsistent practices with regard to stage of exercise assessment. The results of this study, therefore, may have been subjected to self-report bias and unspecified amount of misclassification as to the participant’s actual exercise behaviour. In addition, the ability to generalise the current findings is limited by the sample of emerging adult participants. Through cross-sectional design, this study used convenience sampling of participants to help gather useful data and information that would have not been possible using probability sampling techniques. In addition, participants were recruited from courses within the university to achieve the large sample size in a relatively fast and inexpensive way. The sample was not chosen at random, and thus, the inherent bias of convenience sampling may undermine generalisation from the sample to the emerging adult population being studied.

Furthermore, some individuals may not have felt sufficiently competent at PAs, feeling either not physically fit enough or skilled enough to perform exercise properly; or they may have had health limitations that presented a barrier to exercise. Whether it be low interest or low perceived competence, research indicates many inactive individuals are either unmotivated (i.e. amotivated) to exercise, having no intention to exercise, or are insufficiently motivated in the face of other interests or demands on their time. As this study did not account for amotivation in the exercise self-regulation measurement, results may have not reflected those individuals associated with amotivation. Amotivation, or lack of motivation to exercise, may have influenced participant responses.

The possibility also exists that emerging adults’ perceptions of their parents’ exercise behaviour may have had a greater influence on their own behaviour than parents themselves. Despite growing evidence that parents are an important influence on emerging adult health, few longitudinal studies have explored the causal relationship between parental influence and exercise. Thus, there may have been design issues where the scales used in this study may have presented confusion regarding what each questionnaire was actually measuring; this may have increased interpretive errors for users who incorrectly perceive them as analogous – some of the wording used to depict certain exercise behaviours could have been inconsistent with the participant’s corresponding descriptions, and thus may have been misrepresentative of the intended content or purpose of the item. Participant responses may have been dependent upon an individual’s interpretation of test items, rather than the test stimuli alone. Therefore, inconsistencies in responses may have been due to methodological differences, making it difficult to compare results directly.

Conclusion

The final model confirmed hypothesis that parental PA, warmth and autonomy support were directly related to emerging adults’ experience of autonomy and perceived competence. Parental influence was also significantly related to attraction to PA and exercise, where parental PA, warmth and autonomy support were found to be a significant predictor for each of the intrapersonal correlate measures – autonomy and perceived competence. Conclusively, these findings are in line with previous research emphasising that needs specify the necessary conditions – the impact of autonomy and competence support was found to promote emerging adult’s intrinsic motivation, which consequently affected actual PA.

Collectively, our findings indicate the effects of parental influence upon emerging adults’ exercise behaviours are enduring. Emerging adults whose parents encourage them to exercise and model healthy exercise behaviour themselves are more likely to meet current PA guidelines. Thus, these findings serve to remind parents that they are role models for children and emerging adults and underscore the importance of parental involvement in promoting PA and exercise throughout emerging adulthood. This study also reminds parents that as emerging adults, their children are still watching and learning from observing their behaviours, both good and bad.

Footnotes

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

References

Arnett

(2012) Adolescence and Emerging Adulthood: A Cultural Approach. 5th ed. New York: Pearson Education.

Bentler

Chou

(1987) Practical issues in structural equation modeling. Sociological Methods & Research 16: 78–117.

Boyle

LaRose

(2008) Personal beliefs, the environment and college students’ exercise and eating behaviors. American Journal of Health Studies 23(4): 195–200.

Buckworth

Nigg

(2004) Physical activity, exercise, and sedentary behavior in college students. Journal of American College Health 53(1): 28–34.

Daley

Duda

(2006) Self-determination, stage of readiness to change for exercise, and frequency of physical activity in young people. European Journal of Sport Science 6: 231–243.

Edwardson

Gorley

(2010) Parental influences on different types and intensities of physical activity in youth: A systematic review. Psychology of Sport and Exercise 11: 522–535.

Fallon

Hausenblas

Nigg

(2005) The transtheoretical model of exercise adherence: Examining construct associations in later stages of change. Psychology of Sport and Exercise 6: 629–641.

Farmanbar

Niknami

Heydarnia

. (2009) Predicting exercise behavior among Iranian college students using the transtheoretical model and structural equation modeling. European Journal of Scientific Research 31(3): 355–365.

Gray

(2011) Theory guiding communication campaign praxis: A qualitative elicitation study comparing exercise beliefs of overweight and healthy weight college students. Qualitative Research Reports in Communication 12(1): 34–42.

10.

Grolnick

Deci

Ryan

(1997) Internalization within the family: The self-determination theory perspective. In: Grusec

Kuczynski

(eds) Parenting and Children’s Internalization of Values: A Handbook of Contemporary Theory. New York: Wiley, pp. 135–161.

11.

Gruber

(2008) Social support for exercise and dietary habits among college students. Adolescence 43(171): 557–575.

12.

Gunnell

Wilson

Zumbo

. (2013) Assessing psychological need satisfaction in exercise contexts: Issues of score invariance, item modification, and context. Measurement in Physical Education and Exercise Science 16(3): 219–236.

13.

Hagger

Chatzisarantis

(2007) Intrinsic Motivation and Self-Determination in Exercise and Sport. Champaign, IL: Human Kinetics.

14.

Hoyle

Panter

(1995) Structural Equation Modeling: Concepts, Issues, and Applications. Thousand Oaks, CA: SAGE.

15.

Kim

(2006) Application of the transtheoretical model to identify psychological constructs influencing exercise behavior: A questionnaire survey. International Journal of Nursing Studies 44: 936–944.

16.

Landry

Solomon

(2004) African American women’s self-determination across the stages of change for exercise. Journal of Sport & Exercise Psychology 26: 457–469.

17.

McEloy

(2002) Resistance to Exercise: A Social Analysis of Inactivity. New York: Human Kinetics.

18.

Prochaska

Norcross

DiClemente

(1994) Changing for Good: The Program That Explains the Six Stages of Change and Teaches You How to Free Yourself from Bad Habits. New York: Morrow.

19.

Robbins

(1994) An assessment of perceptions of parental autonomy support and control: Child and parent correlates. Unpublished Doctoral Dissertation, Department of Psychology, University of Rochester, Rochester, NY.

20.

Ryan

Deci

(2000) Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist 55: 68–78.

21.

Ryan

Deci

(2007) Active human nature: Self-determination theory and the promotion and maintenance of sport, exercise, and health. In: Hagger

Chatzisarantis

NLD

(eds) Intrinsic Motivation and Self-Determination in Exercise and Sport. Champaign, IL: Human Kinetics, pp. 1–19.

22.

Ryan

Patrick

Deci

. (2008) Facilitating health behavior change and its maintenance: Interventions based on self-determination theory. The European Health Psychologist 10: 2–5.

23.

Ryan

Williams

Patrick

. (2009) Self-determination theory and physical activity: The dynamics of motivation in development and wellness. Hellenic Journal of Psychology 6: 107–124.

24.

Satorra

Bentler

(1994) Corrections to test statistics and standard errors in covariance structure analysis. In: Von Eye

Clogg

(eds) Latent Variables Analysis: Applications for Developmental Research. Thousand Oaks, CA: SAGE, pp. 399–419.

25.

Scarpa

Nart

(2012) Influences of perceived sport competence on physical activity enjoyment in early adolescents. Social Behavior and Personality: An International Journal 40(2): 203–204.

26.

Schumaker

Lomax

(2004) A Beginner’s Guide to Structural Equation Modeling. Mahwah, NJ: Lawrence Erlbaum Associates.

27.

Sheldon

Filak

(2008) Manipulating autonomy, competence and relatedness support in a game-learning context: New evidence that all three needs matter. British Journal of Social Psychology 47: 267–283.

28.

Tanaka

Huba

(1989) A general coefficient of determination for covariance structure models under arbitrary GLS estimation. British Journal of Mathematical & Statistical Psychology 42: 233–239.

29.

Thompson

Green

(2006) Evaluating between-group differences in latent variable means. In: Serlin

(ed.) Structural Equation Modeling: A Second Course. Charlotte, NC: Information Age Publishing, pp. 119–169.

30.

United States Department of Health and Human Services (USDHHS) (2008) 2008 Physical Activity Guidelines for Americans. Washington, DC: US Government Printing Office. Available at: http://health.gov/paguidelines/pdf/paguide.pdf (accessed 7 June 2016).

31.

United States Department of Health and Human Services (USDHHS) (2010) Healthy People 2010. 2nd ed. Washington, DC: US Government Printing Office.

32.

Van de Berghe

Vansteenkiste

Cardon

. (2012) Research on self-determination in physical education: Key findings and proposals for future research. Physical Education and Sport Pedagogy 1: 97–121.

33.

Vlachopoulos

Michailidou

(2006) Development and initial validation of a measure of autonomy, competence, and relatedness in exercise: The basic psychological needs in exercise scale. Measurement in Physical Education and Exercise Science 10: 179–201.

34.

Vlachopoulos

Ntoumanis

Smith

(2010) The basic psychological needs in exercise scale: Translation and evidence for cross-cultural validity. International Journal of Sport and Exercise Psychology 8: 394–412.

35.

Weiss

(2000) Motivating kids in physical activity. Presidents’ Council on Physical Fitness and Sports Research Digest 3(4): 11.

36.

Welk

Wood

Morss

(2003) Parental influences on physical activity in children: An exploration of potential mechanisms. Pediatric Exercise Science 15: 19–33.

37.

Williams

Deci

(1996) Internalization of biopsychosocial values by medical students: A test of self-determination theory. Journal of Personality and Social Psychology 70: 767–779.

38.

Williams

Freedman

Deci

(1998) Supporting autonomy to motivate glucose control in patients with diabetes. Diabetes Care 21: 1644–1651.

39.

Williams

Grow

Freedman

. (1996) Motivational predictors of weight loss and weight-loss maintenance. Journal of Personality and Social Psychology 70(1): 115–126.

40.

Wilson

Rogers

(2008) Examining relationships between perceived psychological need satisfaction and behavioral regulations in exercise. Journal of Applied Biobehavioral Research 13: 119–142.

41.

Wilson

Rodgers

Blanchard

. (2003) The relationship between psychological needs, self-determined motivation, exercise attitudes, and physical fitness. Journal of Applied Social Psychology 33: 2373–2392.

42.

Wilson

Rogers

Rodgers

. (2006) The psychological need satisfaction in exercise scale. Journal of Sport & Exercise Psychology 28: 231–251.