A mixed method approach to the effect of family structure on young people’s physical and sedentary activities

A mixed method approach to the effect of family structure on young people’s physical and sedentary activities

ABSTRACT

The family is considered a stable bedrock and vital influence on young people’s physical activity and overall health. However, recent decades have witnessed increasing diversity in modern families resulting from a decline in marriage and rising divorce rates. This paper sets out to uncover the complexities of family structure and its impact on young people’s engagement in various types of activities. Adopting an ecological approach and mixed method design it draws on questionnaires (n = 381) and semi-structured interviews (n = 62) with young people from three inner city comprehensive schools in the United Kingdom. Quantitative results indicated that those in single parent families engaged in more sedentary activities (p<0.01) than their intact couple family counterparts. Those in intact couple families however reported more time in lifetime activities than children in single parent families (p<0.01). Qualitative results suggested that those in single parent families received less parental and logistical support due to a lack of time and additional parental responsibilities, creating an environment that encouraged sedentary pursuits and preventing engagement in lifetime activities with other family members. In contrast, those in two parent families had more opportunities to engage in lifetime activities individually and as part of whole family activities.

INTRODUCTION

Epidemiological studies indicate that physical activity declines around the onset of puberty (Brodersen et al., 2007; Sallis et al., 2000) and that within particular groups (i.e. low income families) young people are more susceptible to this decline than others. Around the same age, the family is identified as one of the most important influences on both young people’s physical activity and sedentary behaviours, with only “biological factors or children’s own cognitions, attitudes and affect about activity (themselves likely impacted by the family)… considered more proximal factors” (Saelens and Kerr, 2008; 267).

However, in many Westernized countries and in the United Kingdom in particular, a decline in marriage and rising divorce rates mean that families have become less stable with more young people experiencing transitions from one family structure to another (Kay, 2004; Quarmby and Dagkas, 2010). In England, approximately 1 in 4 dependant children are now living with single parents whilst more than 1 in 8 live in a stepfamily (McConnell and Wilson, 2007). Empirical research from Australia has even suggested that a decline in the conventional ‘nuclear’ family may in part explain low levels of physical activity among young people and such social transformations in families may be a key influence on their physical activity and sedentary behaviours (Bagley et al., 2006). This finding is echoed in a recent qualitative study (Quarmby and Dagkas, 2010), which indicated that the transition from one family structure to another can result in reduced physical activity and increased sedentary behaviour. 

From an epidemiological perspective, the associations between family structure and young people’s physical activity have produced conflicting results. Worldwide findings from developed countries including the United States (Sallis et al., 1999a; Sallis et al., 1999b), Canada (O’Loughlin et al., 1999), Australia (Bagley et al., 2006; Hesketh et al., 2008; Salmon et al., 2005) and England (Gorely et al., 2009) have reported no association between activity in single and two parent families. That said, three studies from the US (Duncan et al., 2004; Lindquist et al., 1999; Sallis et al., 1992) stated that young people in single parent families were more active. On the other hand, additional studies from Australia (Hesketh et al., 2006), South Africa (McVeigh et al., 2004) and Canada (Tremblay and Willms, 2003) found that those in two parent families were typically more active than those in single parent families. Interestingly, there is a much clearer picture that emerges from epidemiological studies when the association with sedentary behaviours is considered. In 2004, Gorely and colleagues conducted a review of the relevant literature and concluded that television viewing was inversely related to the number of parents in the home: fewer parents in the home (single parent families) meant more time spent watching television. These findings were similar to other studies from England (Gorely et al., 2009) and Australia (Bagley et al., 2006; Hesketh et al., 2006).  

Despite authors advocating that future research try to explore the impact of one-parent families on youth physical activity and sedentary behaviour (Saelens and Kerr, 2008), most statistical data fails to explain how and why variables such as family structure influence the lives of young people. Clearly, in order to fully understand the contexts in which such conflicting behaviours occur, further qualitative research is required that adds meaning to participation statistics.

AREA OF STUDY

In an effort to illuminate the influence of family structure this paper embraced both quantitative and qualitative elements. It addresses physical activity of young people from different family structures by adopting an ecological approach with a particular focus on the family as a social environment. Ecological approaches have at their core, the notion that physical activity behaviour is influenced by multiple facets of the intrapersonal, interpersonal and environmental (Gorely, 2005). Such models focus on individual influences as well as physical and social environmental factors that may facilitate or inhibit individual behaviour (Sallis and Owen, 1999). This approach is used to study the interrelations of the individual and their physical and social environment and suggests that individuals adopt their behaviour in response to changes in the physical and social environment, which provides a framework for explaining behaviour.

According to ecological models, multiple levels of influence are thought to determine individual behaviour which consists of the micro-system being as the most proximal setting to the individual and macro-system the most distal. Of these, the micro-system is the immediate setting within which individuals interact (i.e. the family or the school) and consists of both social (e.g., verbal support) and physical (e.g., the presence of a safe play spaces) characteristics. However, it is likely that more than one micro-system plays a role in understanding physical activity. Encompassing the micro-system is the meso-system,in which two or more micro-systems may interact to exert influence on physical activity behaviour (i.e. the family and the school). The exo-system is then the larger social system that the individual is a part of, whereas the macro-system refers to the larger socio-cultural context (e.g. cultural beliefs and values, social class structure) in which they reside. Each of these levels is proposed to facilitate or constrain certain behaviours though the most proximal levels (micro-system) are thought to have the most direct and explicit influence (Gorely, 2005).

The ecological approach therefore provides a framework for understanding the complex interplay between the many personal and environmental influences on behaviour. As such, this study explored psychosocial and environmental factors that contributed to children’s participation in physical and sedentary activities within the specific context of the family environment. Specifically, it looked at how different family structures (as unique micro-systems) affected young people’s time in various activities coupled with exploring the underlying explanations for such behaviour.

METHODOLOGY

A two phase sequential mixed methods approach was employed (each phase including instrument and data analysis is detailed below). Participants recruited for the study included young people (aged 11 – 14) from three inner city comprehensive schools (secondary state schools) from the Midlands region of England, United Kingdom. The age range of 11 – 14 was considered the prime age for parent child interaction and individual activity participation (Yang et al., 1996). Each school was selected from a low socioeconomic status (SES) area (to provide a greater diversity in family structures) based on their Index of Multiple Deprivation (IMD) score. The IMD is a Government measure of area deprivation that assesses income, employment, health and education (Noble et al., 2008). Using the school postcode, the IMD provided an overall rank of deprivation for each school from which subsequent participants were drawn.

As such, participants of three prominent family structures were sought from low SES schools including: Intact couple families – whereby “all children are the biological children of two non-divorced parents”; Single parent families – “families in which all children are the biological children of a non-married, non-cohabitating man or woman”; and Stepfamilies – “families in which the study child is the biological child of one partner but biologically unrelated to the other partner” (Wise, 2003; 21)

Phase 1 – Instrument and Procedure

Participants in Phase 1 (n = 381) completed a self report recall questionnaire to gauge time spent in a variety of activities. The questionnaire was based on Cale’s (1993) Four by One Day Physical Activity Recall which was designed to capture the type and the length of time children engaged in activities from the previous day only. A reliability of r = 0.62 (p<0.05) was obtained for the original questionnaire and was deemed a reliable measure of physical activity (Cale, 1994). Concurrent validity was assessed using heart rate monitoring and an observational method.

The resulting relationship for heart rate monitoring was r = 0.61 (p<0.01) whilst no significant difference was recorded between the recall and observational values (t = 0.72). Children aged 11-14 were therefore thought to be capable of accurately recalling time in activity for one previous day only (Cale, 1994). Based on feedback from participants after initial piloting the adapted questionnaire contained a list of 26 activities that could be categorised according to type (see Table I).  

Table 1. Questionnaire Activity Categories

Contenido disponible en el CD Colección Congresos nº 9

 

Questionnaires were delivered to randomly selected classes and students before Physical Education (P.E.) lessons on two separate days with the researcher and teacher present at all times, to distribute, deliver, collect and assist with any difficulties the students may have had with comprehension. All questions were read out to the students who were also asked to indicate their age and gender and to write down who they normally lived with at home and their relationship to them (to determine family structure). Students were then asked whether they had engaged in any of the 26 activities on the previous day (outside of school) and if so, to report their time in that activity to the nearest 15 minutes. Each individual activity was read out by the investigator before children responded. The questionnaire was delivered once on a Monday and once on a Wednesday capturing participants previous day activity (excluding school time) that could be compared across family structures.

Phase 2 – Instrument and Procedure

Since there is a need to understand, engage and empower young people with regards physical activity, listening to their voice, their own experiences, was paramount and in keeping with the mixed methods approach, semi structured interviews were employed in the second phase to do just that. Qualitative interviewing allowed the researcher to gather an understanding of the participant’s experiences and generate pictures by reconstructing events about a specific topic (Powney and Watts, 1987). The semi structured interview schedule was devised to enhance findings from the questionnaire after initial piloting with 12 children (aged 11 – 12). The resulting questions were specifically focused on the impact of family structure and their engagement in physical activities, ranging from: “What activities do you normally do with your family?”; “Do you enjoy doing physical activities with members of your family, why?” “Is there anything that prevents you taking part in physical activities more often?”
To ensure everyone understood the nature of the second phase, a school assembly was held in each school during which time the participants were informed of the nature of the interviews and invited to take part. Purposive sampling was then employed drawing on demographic data from the questionnaires before thirty small paired interviews where conducted in an open P.E. staff room (lasting 20 – 45 minutes) with participants from the same year, gender and family structure (n = 62). With permission, interviews were recorded and immediately transcribed verbatim so as to ensure a complete and accurate record of the data was obtained.

Phase 1 – Data Analysis

Data analysis was carried out on PASW Statistics 18.0, with weekend and weekday data analysed separately since there is greater discretionary time available at weekends, which may influence behaviour (Jago et al., 2005). All data were tested for approximation to the normal distribution by Kolmogorov-Smirnov test. Continuous minutes in each activity variables were found to be non-normally distributed, with significant skew and kurtosis evident in many variables. Attempts were made to correct the large positive skew of these variables using logarithmic transformations but the data remained non-normally distributed. Activity minute data were therefore analyzed using non-parametric statistics. Kruskal-Wallis tests were used to identify any effect of family structure on the time spent carrying out different activities during the week and at weekends. Where differences were found, Mann-Whitney U tests were performed to determine the location of any differences between family structures. Exact p values where calculated where possible but in the event of computational problems a Monte Carlo approximation based upon 200,000 samples was used to determine the p value. A Bonferroni correction was applied to the probability of a Type I error (a) when multiple comparisons were carried out (Field, 2009). In order to more easily present the non normal distributions of activity minutes data, durations spent for each type of activity were grouped in 30 minute intervals from 0 minutes.

Phase 2 – Data Analysis

Analysis was based on deductive and inductive procedures (LeCompte and Preissle, 1993), which involved scanning the data for common categories and themes (semi-structured interview schedule), developing working typologies on an examination of initial cases and then modifying and refining them on the basis of subsequent cases (Quarmby and Dagkas, 2010). Peer-debriefing and members check methods were used to enhance the trustworthiness of the study (Miles and Huberman, 1994): the lead research met with a sample of participant children to comment on interview transcripts and preliminary findings. The data analysis was also conducted by two researchers who independently coded responses before resulting categories from both were compared. Whilst it is acknowledged that each student’s experiences were unique, the quotations presented were chosen as they were consistently reported across the sample of participants.

RESULTS

Data reported in the results section is addressed in the same order that data were analysed before being interpreted together in the following discussion and conclusion. Phase 1 included participants of both genders (nmale = 218; nfemale = 163), aged 11–14 (M = 12.54; SD = 0.91), from the three family structures outlined earlier (Intact Couple Family = 201; Single Parent Family = 120; Step Family = 60). The second qualitative phase again included both genders (nmale = 34; nfemale = 28; M = 12.54; SD = 0.91) purposely selected to represent the aforementioned family types (Intact Couple Family = 19; Single Parent Family = 24; Step Family = 19).

Quantitative Results

For the first phase there were significant effects of family structure identified for sedentary activities in children during the week (H(2)=9.17, p<0.01, Figure 1) and at the weekend (H(2)=7.55, p=0.02, Figure 2). In addition, there was an effect of family structure on the time spent on lifetime activities in the week (H(2)=9.70, p<0.01, Figure 2) but not at the weekend (H(2)=1.70, p=0.43, Figure 2).
However, there were no significant effects reported for family structure on the distribution of time spent in domestic (H(2)=3.86, p<0.15, week; H(2)=4.63, p<0.10, weekend), partner (H(2)=0.25, p<0.89, week; H(2)=5.41, p<0.07, weekend), games (H(2)=5.91, p<0.05, week; H(2)=3.04., p<0.22, weekend), or other activities (H(2)=1.74, p<0.42, week; H(2)=2.16, p<0.34, weekend).

Sedentary Activities   
After correction for multiple comparisons further analyses (a=0.0167) showed that there were no significant differences between intact couple families and step families during the week (U=5871, z=-0.31, p=0.76, r=-0.02) or weekend (U=6023, p=0.99, z=-0.01, r=-0.00), or between single parent families and step families (U=2895, p=0.03, z=-2.14, r=-0.16, week; U=3048, p=0.09, z=-1.68, r=-0.13 weekend). A significant difference was found between intact couple families and single parent families during the week (U=9757, p=0.004, z=-2.87) and weekend (U=9875, p=0.006, z=-2.72). However, the effect sizes of week (r=-0.16) and weekend (r=-0.15) activity minutes were small. Mean ranks suggested that children from single parent families spent more time engaged in sedentary activities during both the week (150 c.f. 180; intact, single) and weekend (150 c.f. 179; intact, single).

Lifetime Activities

Further analyses (a=0.0167) of the lifetime activities data during the week showed significant differences between intact couple families and single parent families (U=9620, p<0.01, z=-3.09). Although this effect was weak (r=-0.17), the mean ranks suggested that children from single parent families spent less time in lifetime activities during the week (173 c.f. 141; intact, single). In fact, 29% of those in intact couple families engaged in lifetime activities for 90 minutes or more compared to just 14% (approximately half) of those in single parent families. There were no differences between intact couple families and stepfamilies during the week (U=5981, p=0.92, z=-0.01, r=-0.00) or single parent families and step families (U=2989, p=0.06, z=-1.90, r=-0.14).

Contenido disponible en el CD colección congresos nº 16

Contenido disponible en el CD colección congresos nº 16

Qualitative Results

Sedentary Activities

In conjunction with the findings from the first phase were findings from the second, qualitative phase that emphasised young people’s engagement in sedentary activities. Moreover, such findings predominantly related to young people from single parent families. As the examples below indicate, those in single parent families are often prone to a lack of parental support. For Jack, this was in the form of transportation, which prevented him engaging in physical activities and instead, left him to seek out sedentary alternatives.
My mum’s job [referring to barriers to activity], she gets back at like half five so if I want to go anywhere to do anything it’s normally too late so instead like I just play X box and stay in when she’s not back. (Jack, single parent family)
In addition, a lack of time on their parents’ part also contributed to a lack of joint family physical activities. This led to involvement in joint activities that were low intensity and sedentary in nature. Both Adam and Naomi (single parent families) argued that the only time they had to engage with their parents was whilst watching television together.

We watch TV, and have dinner and everything. Well not everything, she [mother] wants her own space and like wants to do everything that she wants to do rather than me going with her all the time but we normally watch TV together…
She’s normally [mother] really busy with like housework or other work so yeah… it’s quite hard like she just don’t have time I guess and she just wants some time to herself but sometimes we like… well yeah just TV really. (Adam, single parent family)
Apart from watching like TV… no not really [referring to family activities] … maybe used to do little things when they [parents] were together ages ago but she’s not got the time now so yeah just TV. (Naomi, single parent family)
In the previous examples, Adam suggested that his mother’s desire to use her free time, outside of her domestic chores, for herself, restricted their ability to interact and engage in anything other than low intensity sedentary pursuits. Similarly, Naomi pointed towards time as a barrier to engaging in anything other than sedentary pursuits with her mother. Interestingly though, she indicated that this wasn’t always the case when both her parents were together.

At the weekend, similar findings emerged that focused on parent’s lack of time. Often the participants reported visiting their other biological parent but again, a lack of time on their part prevented anything other than sedentary activities being engaged in together. 
Erm… weekends my mums always out, while I’m up at me Nan’s or Dad’s and during the week, like Monday to Friday she’s at work… Erm, one week I go to my Nan and then the other week to my dad but they’re always busy so we don’t really do much just watch TV. (Lauren, single parent family)
It was certainly clear that a common theme amongst responses from children of single parent families was the limited time available for physical activity pursuits that contributed to the development of sedentary behaviours.

Lifetime Activities

In contrast to single parent families, having two parents (especially biological parents) at home seemed to more frequently encourage and support engagement in lifetime activities. Given that lifetime activities tend to be readily accessible and low cost (e.g. running, cycling, swimming) (Green, 2004) young people from intact couple families reported participating in such activities after school and in the evenings. Sam, for example highlighted that whilst he engaged in activity with his father, his mother ensured all other family responsibilities were met.

Yeah, bike rides, parks, picnics, cinema, town, things like that, shopping… I go riding with my dad and sometimes go for a run with her [mother] and like sometimes she comes cycling too… [but usually] she does housework and cooking and stuff so it’s all ready for when we get in. (Sam, intact couple family)
It was certainly evident in some responses how two parents could work together so that if one was preoccupied or busy with work or household duties, then the other could provide support for engagement in lifetime activities.
Well you know I’m quite lucky really coz like… unlike Gemma [her friend from a single parent family] when dads busy doing something like at work or whatever, mum always takes me swimming with my sisters or when mum’s cooking dad will do something like take us for a short bike ride (Claire, intact couple family)

There was a clear contrast between young people from two parent families and those from single parent families whose parents faced barriers that prevented them from supporting their child’s engagement in lifetime activities during the week. In fact, the extract below further demonstrates the difficulties that young people from single parent families faced in trying to engage in such activities.
…sometimes we go on a bike ride and I play … football with my brother. We used to play in the garden quite a lot… [but] not as much… he’s never in any more so now I just stay in a bit more and watch a bit of TV with mum when she’s back from work and after she’s finished doing the house stuff…. (Sarah, single parent family)
The home social environment in two parent families therefore appeared more conducive to physical activity during the week as two parents were able to manage daily domestic duties better. This subsequently allowed them more time to engage in easily accessible lifetime activities (such as running or cycling) with their children. In contrast, parents in single parent families were more prone to the pressures of time and work which prevented joint engagement in such activities.   

DISCUSSION

This study sought to explore the influence of family structure on the types of activities that young people engage in. In doing so, it adopted a mixed methods design and drew from an ecological approach for data analysis in an attempt to help understand the influence of unique micro-systems on individual behaviour. The two main findings from the data centred around sedentary and lifetime activities. For the purposes of this paper, sedentary activities included those activities that may be considered inactive (Marshall and Welk, 2008) as outlined in table 1 earlier. Lifetime activities (also categorised in table 1) were activities that could be freely undertaken when and how individuals chose, with whom and wherever they wanted and were recreational, often with a health and fitness orientation (Green, 2004). In line with previous epidemiological studies (Bagley et al., 2006; Gorely et al., 2009; Lindquist et al., 1999; Hesketh et al., 2006), young people from single parent families spent more time in sedentary activities than their two parent counterparts.

Using a mixed method approach also allowed qualitative data to enhance quantitative findings, indicating that those from single parent families faced more barriers during the week and at the weekend to the amount of time they could spend with their parents, ultimately effecting their availability to engage in joint activities other than those most accessible (e.g. television viewing). These young people had parents with numerous additional responsibilities such as caring for other children and household duties, which made managing their child’s physical activity difficult due to logistical constraints such as limited transport and a lack of time. Consequently, this created a home environment that promoted sedentary behaviours as the only joint activity

Unlike previous studies that produced conflicting results for the relationship between family structure and physical activity, total physical activity was not assessed here. However, types of physical activity were considered and quantitative results indicated differences for lifetime activities when split by family structure. A significant difference was reported between those in intact couple families and those in single parent families with young people from intact couple families spending more time in lifetime activities. Lifetime activities are thought to be health and fitness orientated and possess a greater carry-over value than team or game activities as they are often more readily available and so, if children engage in lifetime activities then there is a greater possibility that they will follow a physically active lifestyle during adulthood (Fairclough et al., 2002). Qualitative data suggested that during the week, these activities were more readily engaged in by young people from intact families, who were likely to participate with one or both of their parents. In their reports, whilst one parent took care of household duties, the other was able to engage in easily accessible activities with them. Hence, if young people from both types of two parent families are exposed to and spend more time in lifetime activities in youth, then there exists the possibility that they will continue to engage in similar activities in adulthood.

Similar to a previous report (Quarmby and Dagkas, 2010), children from two parent families received more physical and social support to engage in activities after school. Baranowski argues that (1997; 183) the “family is something more than the sum of its parts” suggesting two parents are better suited to manage the family environment more easily than single parent families. It was evident here that children from single parent families faced more barriers and lead more sedentary lifestyles than their intact couple family counterparts. Here, children in single parent families received less parental support due to their parent’s lack of free time, workload and catering for younger siblings, which in turn impacted on family logistics. Furthermore, many young people reported that at weekends, they had to travel to see their other biological parent (often father) who in some cases did not live nearby, restricting time to engage in anything other than sedentary activities. Importantly, ecological approaches acknowledge that behaviour is both influenced by, and influences the environment in a reciprocal reaction and thus, young people growing up in families with limited opportunities for physical activity but preferences for sedentary activity may ultimately shape their behaviour to suit the environment.

This study is unique in that it draws on both quantitative and qualitative data, adopts an ecological approach and discusses differences in the types of activities that young people from certain families engage in. Moreover, by drawing on families from low income areas, the potential influence of socioeconomic factors was minimized allowing for greater comparability between family structures. However, whilst the use of the IMD may not accurately reflect the actual level of deprivation of the families from which these participants were drawn (since children may not live in the same geographical ward to that of their school) it did provide reasonable indication of the deprivation of families living within the area.

CONCLUSION

Drawing from an ecological approach, this research suggested that the nature of the family structure restricted the availability of joint family activities and impacted on the amount of time children spent in certain physical activities. For some, their opportunity for joint activity was limited to sedentary pursuits with their single parent as this is the most accessible activity. In contrast, young people from intact couple families spent more time during the week engaged in accessible lifetime activities, as a joint or whole family activity, thanks to a greater ability to manage the family workload.

Understanding the types of activity young people have opportunities to engage in may make it easier to design and target specific interventions. In fact, the results of this research suggest future designs of family based interventions look to address correlates of physical activity and sedentary behaviour for young people, particularly in deprived areas. It may be that lifetime activities performed together have a more lasting effect and since single parent families struggle to find time to do this, future interventions could try to accommodate the competing demands that single parents face. It may also be important to consider what parents themselves do and that reducing sedentary behaviour in youth may require actions to change the behaviour of adults particularly among single parent families. Finally, if we are to truly understand the physical activity choices of young people from a variety of backgrounds then models that integrate, social, cultural and physical aspects, such as these simply can not be ignored.

REFERENCES

Bagley, S., Salmon, J. and Crawford, D. (2006) Family Structure and Children’s Television Viewing and Physical Activity. Medicine and Science in Sport and Exercise, 38 (5): 910-918.

Baranowski, T. (1997) Families and Health Actions, In Gochman, D. (ed.) Handbook of Health Behaviour Research: Personal and Social Determinants. New York, Plenum Press. 1: 179-206.

Brodersen, N., Steptoe, A., Boniface, D., et al. (2007) Trends in physical activity and sedentary behaviour in adolescence: ethnic and socioeconomic differences. British Journal of Sports Medicine, 41: 140-144.

Cale, L. (1993) The Four by One-Day Physical Activity Recall Questionnaire Protocol and Instruction Manual. Loughborough, Loughborough University.

Cale, L. (1994) Self report measures of children’s physical activity: recommendations for future development and a new alternative measure. Health Education Journal, 53 (4): 439-453.

Duncan, S., Duncan, T., Strycker, L., et al. (2004) A Multilevel Analysis of Sibling Physical Activity. Journal of Sport and Exercise Psychology, 26 (1): 57-68.

Fairclough, J., Stratton, G. and Baldwin, G. (2002) The contribution of secondary school physical education to lifetime physical activity. European Physical Education Review, 8 (1): 69-84.

Field, A. (2009) Discovering Statistics Using SPSS. London, Sage.

Gorely, T. (2005) The Determinants of Physical Activity and Inactivity in Young People, In Cale, L. and Harris, J. (ed.) Exercise and Young People: Issues, Implications and Initiatives. Basingstoke, Palgrave Macmillan: 81-102.

Gorely, T., Atkin, K., Biddle, S., et al. (2009) Family circumstance, sedentary behaviour and physical activity in adolescents living in England: Project STIL. International Journal of Behavioral Nutrition and Physical Activity, 6 (33): 1-8.

Green, K. (2004) Physical education, lifelong participation and ‘the couch potato society’. Physical Education and Sport Pedagogy, 9 (1): 73-86.

Green, K., Smith, A. and Roberts, K. (2005) Young people and lifelong participation in sport and physical activity: a sociological perspective on contemporary physical education programmes in England and Wales. Leisure Studies, 24 (1): 27-43.

Hesketh, K., Crawford, D. and Salmon, J. (2006) Children’s television viewing and objectively measured physical activity: associations with family circumstance. International Journal of Behavioral Nutrition and Physical Activity, 3 (36): 1-10.

Hesketh, K., Graham, M. and Waters, E. (2008) Children’s After-School Activity: Associations with Weight Status and Family Circumstance. Pediatric Exercise Science, 20 (1): 84-94.

Jago, R., Anderson, C., Baranowski, T., et al. (2005) Adolescent Patterns of Physical Activity: Differences by Gender, Day, and Time of Day. American Journal of Preventative Medicine, 28 (5): 447-452.

Kay, T. (2004) The Family Factor in Sport: A review of family factors affecting sports participation, In England, S. (ed.) Driving up participation: The challenge for sport. London, Sport England: 39-60.

LeCompte, M. and Preissle (1993) Ethnography and Qualitative Design in Educational Research. San Diego, Academic Press.
Lindquist, C., Reynolds, K. and Goran, M. (1999) Sociocultural Determinants of Physical Activity among Children. Preventive Medicine, 29: 305-312.

Marshall, S. and Welk, G. (2008) Definitions and Measurement, In Smith, A. and Biddle, S. (ed.) Youth Physical Activity and Sedentary Behaviour. Leeds, Human Kinetics: 3-30.

McConnell, H. and Wilson, B. (2007) Families, In Smallwood, S. and Wilson, B. (ed.) Focus on Families. Basingstoke, Palgrave Macmillan: 1-19.

McVeigh, J., Norris, S. and de Wet, T. (2004) The relationship between socio-economic status and physical activity patterns in South African children. Acta Paediatr, 93 (7): 982-988.

Miles, M. and Huberman, M. (1994) Qualitative Data Analysis. New York, Sage.

Noble, M., mcLennan, D., Wilkinson, K., et al. (2008) The English Indices of Deprivation 2007. London, Department for Communities and Local Government.

O’Loughlin, J., Paradis, G., Kishchuk, N., et al. (1999) Prevalence and Correlates of Physical Activity Behaviors among Elementary Schoolchildren in Multiethnic, Low Income, Inner-City Neighborhoods in Montreal, Canada. Annals of Epidemiology, 9 (7): 397-407.

Powney, J. and Watts, M. (1987) Interviewing in Educational Research. London, Routledge.

Quarmby, T. and Dagkas, S. (2010) Children’s engagement in leisure time physical activity: exploring family structure as a determinant. Leisure Studies, 29 (1): 53-66.

Saelens, B. and Kerr, J. (2008) The Family, In Smith, A. and Biddle, S. (ed.) Youth Physical Activity and Sedentary Behaviour. Leeds, Human Kinetics: 267-294.

Sallis, J., Alcaraz, J., McKenzie, T., et al. (1999a) Predictors of Change in Children’s Physical Activity Over 20 Months. American Journal of Preventative Medicine, 16 (3): 222-229.

Sallis, J., Alcaraz, J., McKenzie, T., et al. (1992) Parental Behavior in Relation to Physical Activity and Fitness in 9-Year-Old Children. American Journal of Diseases of Children, 146 (11): 1383-1388.

Sallis, J. and Owen, N. (1999) Physical Activity and Behavioural Medicine. London, Sage Publications.

Sallis, J., Prochaska, J. and Taylor, W. (2000) A review of correlates of physical activity of children and adolescents. Medicine and Science in Sport and Exercise, 32 (5): 963-975.

Sallis, J., Prochaska, J., Taylor, W., et al. (1999b) Correlates of Physical Activity in a National Sample of Girls and Boys in Grades 4 Through 12. Health Psychology, 18 (4): 410-415.

Salmon, J., Timperio, A., Telford, A., et al. (2005) Association of Family Environment with Children’s Television Viewing and with Low Level Physical Activity. Obesity Research, 13 (11): 1939-1951.

Tremblay, M. and Willms, J. (2003) Is the Canadian childhood obesity epidemic related to physical inactivity? International Journal of Obesity, 27 (9): 1100-1105.

Wise, S. (2003) Family structure, child outcomes and environmental mediators: An overview of the Development in Diverse Families study. Melbourne, Australian Institute of Family Studies.

Yang, X., Telema, R. and Laakso, L. (1996) Parents Physical Activity, Socioeconomic Status and Education as Predictors of Physical Activity and Sport Among Children and Youths – A 12 Year Follow up Study. International Review for the Sociology of Sport, 31 (3): 274-289.