GAME CHANGERS THE IMPACT OF MAJOR LIFE EVENTS ON SPORT PARTICIPATION Jasper van Houten
Game changers The impact of major life events on sport participation Jasper van Houten
The studies presented in this dissertation were performed at Radboud Social Cultural Research (RSCR) in collaboration with the HAN School of Sport and Exercise, and financially supported through a grant from the HAN University of Applied Sciences. Design and layout by Lotte Makkinje Printed by Ipskamp Printing ISBN 978-94-6421-768-1 © J.M.A. van Houten, 2022 All rights reserved. Save exceptions stated by the law, no part of this publication may be reproduced, stored in a retrieval system of any nature, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the author. School of Sport and Exercise
Game changers The impact of major life events on sport participation Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus prof. dr. J.H.J.M. van Krieken, volgens besluit van het college van promoties in het openbaar te verdedigen op dinsdag 28 juni 2022 om 10.30 uur precies door Jasper Machiel Antoine van Houten geboren op 17 juli 1986 te Nijmegen
Promotor Prof. dr. G.L.M. Kraaykamp Copromotor Dr. M. Visser Manuscriptcommissie Prof. dr. C.M.C. Verbakel Prof. dr. B. Steenbergen Dr. M. Waardenburg (Universiteit Utrecht)
CONTENTS Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Synthesis A new life stage, a new sport activity? An event history analysis of the impact of major life events on starting a sport When do young adults stop practising a sport? An event history analysis of the impact of four major life events The transition to adulthood: A game changer!? A longitudinal analysis of the impact of five major life events on sport participation Career, family, and sport participation: a simultaneous exhibition? A study of narratives on the impact of major life events during the transition to adulthood Acknowledgements (dankwoord) Summary in Dutch (samenvatting) Appendix Data management References About the author 6 42 70 96 124 168 172 177 224 225 234
SYNTHESIS CHAPTER 1
8 | Chapter 1 GENERAL INTRODUCTION AND AIM OF THIS DISSERTATION The social significance of sport has surged in recent decades. Its public value draws more and more on a vision beyond the intrinsic value of sport (the aim of “sport for sports sake”) and more towards its external merits (“sport for the greater good”) (Brookes &Wiggan, 2009). This appreciation of sport’s functional power stems from its perceived advantageous consequences (Schlesinger & Nagel, 2015; Waardenburg & Van Bottenburg, 2013), for example in improving physical, psychological and social health (Coenders et al., 2017; Eime et al., 2013; Miles, 2007; Oja et al., 2015), in nurturing social capital and integration and thereby fostering social networks (Knoppers, 2006; Putnam, 1995; Seippel, 2006), and in spurring economic growth (European Commission, 2013; Van der Meulen et al., 2012). It is therefore understandable that creating possibilities for safe and nearby sport participation for everyone (i.e., “sports for all”) across the life span (i.e., “lifelong sport participation”) have become a priority in governmental policies and campaigns related to sports, physical activity and health (Council of Europe, 2001; Pilgaard, 2013; Van Tuyckom, 2011; Waardenburg & Van Bottenburg, 2013) such as, in the Netherlands, the Ministry of Health, Welfare and Sport (Van Ark, 2020; Van Ark & Blokhuis, 2020), the Dutch Olympic Committee & Dutch Sports Federation (NOC*NSF, 2017) and the Association of Sports and Municipalities (VSG, 2018)). From an individual’s perspective, practising sport is, for many, an important leisure time activity and present in everyday life. Time spend on active sport participation more than doubled from 1.5% of the available leisure time in 1975, to 3.6% in 2011 (Tiessen-Raaphorst et al., 2014), up to 3.9% in 2016 (Pulles & Tiessen-Raaphorst, 2018). According to most recent measurements in 2016, Dutchmen spend on average 1.7 of the weekly available 43.8 hours of leisure time on active sport participation, and the ones that participated weekly did so for 3.5 hours a week (Pulles & Tiessen-Raaphorst, 2018; Roeters, 2019). However, sport participation of individuals is intrinsically dynamic over the life course. The stability (also called “tracking”) of sport participation is low (Hirvensalo & Lintunen, 2011; Malina, 2001; Telama, 2009; Vanreusel et al., 1997). Individual sport careers are shaped by alternations between active sport episodes and episodes of inactivity (Engel & Nagel, 2011) and characterised by changes in activity level and sort of activity (Butcher et al., 2002; Lunn, 2010; Scheerder et al., 2006; Seippel, 2005). So, from a life-course perspective,
1 | 9 Synthesis besides interpersonal disparities (i.e., differences between individuals) in sport participation, intrapersonal differences (i.e., individual changes) in sport participation during the life course are an important research topic (Engel & Nagel, 2011; Pilgaard, 2013; Pulles & Wendel-Vos, 2018; Tiessen-Raaphorst et al., 2010). This raises the question: why do individuals change their sport participation during their life course? To get an answer to this question, it is important to consider that people experience various major life events during their life, which are expected to affect sport participation and thereby might explain differences and changes in sport participation over the life course. For example, on a typical day in the Netherlands 4891 people move houses, 322 young adults start living on their own and 250 start to cohabit with their partner, 176 couples get married, 54 couples enter into a civil partnership, and 462 women give birth to a child (Statistics Netherlands, 2021c). In addition, according to the most recent data on labour market characteristics, 104,830 people obtained a bachelor’s or master’s degree in the academic year 2016/2017, of which 82,560 entered the labour market after graduation, and 76,420 of them started working within a year after graduating (Statistics Netherlands, 2021a). Furthermore, over 94 thousand employees retired in 2020 (Statistics Netherlands, 2021b). However, little is known about whether such major life events have particular consequences for participation in sport over the life course. I address this question in this dissertation by examining to what extent major life events affect the sport participation of individuals in the Netherlands. To this end, I conduct four empirical studies that examine the influence of major life events on individual sport participation from a life-course perspective. I pay specific attention to the transition to adulthood, as sustaining sport participation during this transition seems particularly challenging (Hirvensalo & Lintunen, 2011; Malina, 2001; Telama, 2009; Vanreusel et al., 1997). This period in life is often associated with a low stability of and sharp decline in individual sport participation (Engel & Nagel, 2011; Hirvensalo & Lintunen, 2011; Leslie et al., 2001; Van Tuyckom, 2011), and a high likelihood of dropping out, especially of club-organised sport (Borgers et al., 2016b; Eime et al., 2016; European Commission, 2018; Lunn, 2010; Pilgaard, 2013; Scheerder et al., 2006; Vandermeerschen et al., 2016). Moreover, the transition to adulthood is for many people a “socially critical period” of key individual life events, which might induce differences between and changes within individual sport careers and introduce breaches in the continuity of sport participation: they move out
10 | Chapter 1 of the parental home and start living on their own, graduate and find their first paid job, start cohabiting, get married and become parents (Allender et al., 2008; Arnett, 2007; Scheerder et al., 2006; Zarrett & Eccles, 2006). Therefore, the transition to adulthood is a very busy and turbulent period in the lives of most people, and potentially crucial when it comes to promoting “sports for all” and a “lifelong sport participation”. Policy makers and sport providers and their initiatives tend to overlook the (young) adults, as they mostly focus on groups that traditionally lagged in sport participation, such as the disabled, children in poverty and the elderly (Hoekman, 2018; Hoekman & Breedveld, 2013). The empirical studies in this dissertation investigating changes and differences in sport participation over the life course, in particular in relation to the occurrence of major life events and especially during the transition to adulthood, may expand knowledge needed to target and encourage people to get and stay involved in sport and sustain sport participation over the course of their lives. First, I examine whether four major life events marking the transition to adulthood (beginning to work, moving out to live on one’s own, starting to cohabit or getting married and the birth of one’s first child) and two events that occur later in life (children leaving the parental home and retirement) affect the likelihood of starting a sport over the life course (chapter 2). Second, I investigate if young adults are more or less likely to stop practising a sport when they begin to work, move out to live on their own, start to cohabit or get married and become a parent (chapter 3). In both studies, I research starting/stopping a sport in general, as well as starting/stopping a club sport in particular. Third, I study if leaving fulltime education, beginning to work, engaging in a relationship, cohabitation or marriage, and becoming a parent affect the number of sports practised by individuals, their sport frequency, and the sport setting in which they practise sport (chapter 4). Finally, I investigate why and how people change their sport participation when major life events occur that mark the transition to adulthood within two life domains: the professional career and the family domain (chapter 5).
1 | 11 Synthesis PREVIOUS RESEARCH: MAIN FINDINGS AND SHORTCOMINGS Sport participation in the Netherlands According to the latest measurements, 73% of the Dutch population aged 6 years and older participates in sport at least 12 times a year (The Netherlands Institute for Social Research (SCP) & Statistics Netherlands, 2018b), and 53.8% of the population aged 12 years and older practises sport at least once a week (Statistics Netherlands & RIVM, 2020). With these relatively high rates, the Netherlands occupies the fourth place in the general European ranking of sport participation, after Sweden, Denmark and Finland, according to the latest data of the European Commission (2018). As shown in figure 1.1, most Dutch sport participation rates are quite stable over recent years, but there are a couple of trends visible. Firstly, there is a small gradual increase in the proportion of people who never participates in sport, from 28% in 2009 (not shown in the figure), to 29% in 2013, up to 31% in 2017. This trend, possibly caused by population ageing, is even more pronounced across Europe (European Commission, 2010, 2014, 2018; Van Stam & Van den Dool, 2018). Secondly, the figures illustrate that sport participation has become diversified and de-traditionalised in recent decades, due to societal processes of individualisation, modernisation and informalisation (see, e.g., Borgers et al., 2018; Heij, 2018; Klostermann & Nagel, 2014; Van Ingen & Dekker, 2011). Although the Netherlands has the highest club membership rate in Europe (European Commission, 2018), with 28% of the Dutch population aged 6 years and older being a club member, this proportion has declined with 4% between 2012 and 2018 (The Netherlands Institute for Social Research (SCP) & Statistics Netherlands, 2018a). Additionally, measurements of NOC*NSF (2019) reveal that between 2013 and 2019, the absolute number of club members decreased by 220,200 (-4.9%) to a total of 4,270,000 members and the total number of memberships dropped with 271,000 to 5,108,000 (-5.0%). Between 2019 and 2020 an additional decline of 35,600 members and 34,900 memberships was measured (NOC*NSF, 2020). Contrastingly, practising sport in commercial settings (e.g., at health or fitness centres), in spontaneous club settings (e.g., an informal group/club consisting of family, friends or relatives) and practising sport alone became more popular in the Netherlands (European Commission,
12 | Chapter 1 2018; Van den Dool, 2019). For example, in 2018, 23% of the Dutch population aged 6 years and older participated in a commercial setting, compared to 19% in 2012 (The Netherlands Institute for Social Research (SCP) & Statistics Netherlands, 2018a). As a result, the market share of club sport participation declined, from 51.4% in 2013 to 41.0% in 2019 (NOC*NSF, 2019). 2012 2013 2014 2015 2016 2017 2018 2019 2020 80% 70% 60% 50% 40% 30% 20% 10% 0% SPORT PARTICIPATION 2012 73.0% 53.5% 32.0% 19.0% 2013 52.7% 29.0% 2014 72.0% 51.4% 32.0% 20.0% 2015 52.1% 2016 70.0% 51.0% 31.0% 21.0% 2017 53.9% 31.0% 2018 73.0% 52.8% 28.0% 23.0% 2019 52.8% 2020 53.8% At least 12 times a week(a) At least once a week(b) Club membership(a) In a commercial setting(a) Never participate in sport(c) Figure 1.1 Sport participation in the Netherlands throughout the years Sources: (a) Vrijetijdsomnibus (VTO) 2012-2018 (The Netherlands Institute for Social Research (SCP) & Statistics Netherlands, 2018b), (b) Gezondheidsenquête/Leefstijlmonitor 2012-2020 (Statistics Netherlands & RIVM, 2020), (c) Sport and physical activity – Special Eurobarometer 412 (European Commission, 2014) and 472 (European Commission, 2018). The previous sections show that practising sport is a common form of leisure time (physical) activity in the Netherlands; many people spend quite some time doing it, in a variety of ways and forms. Yet not everyone is participating, and many people do not sustain participation over the life course, even despite the governmental “sports for all” and “lifelong sport” ambitions and efforts. This is problematic, as participating in sport contributes to various valuable aspects of the lives of human beings, such as health, social relations and playfulness, as well as being considered a meaningful end in itself by many people (Coenders
1 | 13 Synthesis et al., 2017; Eime et al., 2013; Reiner et al., 2013; Seippel, 2006). This makes sport participation an important social phenomenon. Not surprisingly, inclusion and inequality in sport participation is a central theme in the field of sport sociology for some time now (Breuer & Wicker, 2008). Answering questions related to this theme, such as “who practises sport and who doesn’t?”, “when do changes in sport participation within the life course occur?” and “how is sport participation affected by major life events?”, is both scientifically and socially relevant. It is meaningful from a sociological perspective to gain more insight in this social phenomenon. Furthermore, it is important from a practical point of view, for evidence based political advice, sport policy development and sport development planning. Individual determinants of sport participation Prior research shows a wide range of individual determinants of sport participation (for an overview see, for example, Breuer et al., 2010; Foster et al., 2005; Humphreys & Ruseski, 2010). Generally, men are more likely to participate in sport than women (Farrell & Shields, 2002; Hartmann-Tews, 2006; Van Tuyckom, 2011). The Netherlands, however, is one of the few countries where this is not the case (Hartmann-Tews, 2006; Scheerder & Breedveld, 2004; Tiessen-Raaphorst et al., 2014, 2019). According to some studies, Dutch woman are even more likely to participate in sport than Dutch men (Hovemann & Wicker, 2009; Van Tuyckom et al., 2010). Yet, some aspects of sport involvement are in favour of Dutch men, such as duration and time spent on sport, participation in team and competitive sports and sport club membership (Scheerder & Breedveld, 2004; Tiessen-Raaphorst, 2015; TiessenRaaphorst et al., 2014, 2019). Studies also point at a positive effect of human capital and socio-economic status (in terms of educational attainment, household income and/or occupation), and to a negative effect of migration background (or ethnicity) on sport participation. In the Netherlands, individuals who are higher educated, have a higher income and those who are native citizens, are more likely to practise sport than their counterparts who are lower educated, have a lower income, and those with a migration background, respectively (e.g., TiessenRaaphorst, 2015; Tiessen-Raaphorst et al., 2014, 2019; Van der Poel et al., 2018). These findings are also valid for many other countries in Europe (Breuer & Wicker, 2008; Downward, 2007; European Commission, 2018; Hovemann &
14 | Chapter 1 Wicker, 2009; Van Tuyckom & Scheerder, 2014) and beyond (e.g., in the U.S. and Canada, see: Humphreys & Ruseski, 2009, 2010). A limitation of most of these studies on determinants of sport participation, however, is that they lack theoretical rigour. As stated by Breuer & Wicker (2008), mostly, statistical trial and error procedures lead to significant determinants that have little theoretical basis and, consequently, do not explain why these factors influence sport participationmeaningfully. This often results in excessive or inadequately complex models, which are difficult to interpret and to use in practice, for example to develop sport policy and sport programmes or interventions. From age to life events. Sport participation over the life course Age is another important individual factor for sport participation, according to earlier research. In the Netherlands, as in other countries, overall sport participation rates are lower in older age groups than in younger age groups (European Commission, 2018; Pulles & Wendel-Vos, 2018). Practising sport at least 12 times a year and sport club membership are most common among children of 6 through 11 years old, with participation rate of 92% and 80%, respectively. In the age groups that follow, these rates decline (The Netherlands Institute for Social Research (SCP) & Statistics Netherlands, 2018b), as shown in Table 1.1. This Table also shows that 69.9% of Dutch youngsters aged 12 through 17 practise sport on a weekly basis, which is relatively more that the group of children aged 4 through 11. However, again, the rates drop in the following age groups (Statistics Netherlands & RIVM, 2020). Many cross-sectional studies still affirmthe traditional assumptionof decreasing sport activity with increasing age (e.g., Breuer & Wicker, 2009; European Commission, 2018; Hovemann & Wicker, 2009; Schoenborn & Barnes, 2002). A major drawback of these cross-sectional studies is that developments over the life course of individuals are not measured, and that it remains unclear if the identified differences in sport activity between age groups can really be ascribed to age effects (i.e., the effects of increasing age; getting older) (Breuer et al., 2010). Studies with a life-course perspective, employing more advanced data and using longitudinal or cohort sequence analysis, are much scarcer and reveal different and sometimes contradictory findings (Breuer et al., 2010; Curtis et al., 2000; Hovemann & Wicker, 2009; Jenkin et al., 2017). For example, Curtis et al. (2000) found positive correlations between age and sport activity in general, whereas Breuer & Wicker (2009) found positive correlations
1 | 15 Synthesis only for women and negative correlations for men, and (Shaw et al., 2010) found that, on average, rates of leisure-time physical activity increased within younger adults (between age 25 and 41) and decreased within middleaged (between age 45 and 61) and older adults (between age 65 and 81). Additionally, results for longitudinal and cohort sequence analyses showed that declining sport participation by age may to a significant degree reflect a cohort effect rather than an age effect (Breuer & Wicker, 2009; Lunn, 2010). Table 1.1 Differences in sport participation rates by age group AT LEAST 12 TIMES A YEAR(a) Age group 6 to 11 years 12 to 19 years 20 to 34 years 35 to 54 years 55 to 64 years 65 to 79 years 80 years and older % 92 87 77 74 69 63 35 CLUB MEMBERSHIP(a) Age group 6 to 11 years 12 to 19 years 20 to 34 years 35 to 54 years 55 to 64 years 65 years and older % 80 56 27 22 18 17 AT LEAST ONCE A WEEK(b) Age group % 4 to 11 years 12 to 17 years 18 to 64 years 65 years and older 64,5 69,9 56,9 37,8 Sources: (a) Vrijetijdsomnibus (VTO) 2018 (The Netherlands Institute for Social Research (SCP) & Statistics Netherlands, 2018b), (b) Gezondheidsenquête/Leefstijlmonitor 2020 (Statistics Netherlands & RIVM, 2020) So, although age is recognised as a crucial determinant, especially from a lifecourse perspective (Breuer & Pawlowski, 2011; Hovemann & Wicker, 2009), the relationship between age and sport participation is not as evident and much more ambiguous as suggested by the abundance of cross-sectional studies. In addition, age as a determinant is better understood by seeing it as a proxy variable, indicating all potential age-related influences on sport participation (Breuer & Pawlowski, 2011; Breuer & Wicker, 2009; Hovemann & Wicker, 2009). From a socioeconomical approach, four different age-related factors can be differentiated (Breuer & Pawlowski, 2011): changes in physical (decreasing health and fitness by age), mental (changing motivation by age), social (changing norms and support by age) and economic resources (agerelated changes in individual time and financial budgets). Nevertheless, most studies on sport, physical activity and aging still just focus on age as a proxy and contribute only by detecting correlations (Breuer & Pawlowski, 2011). To get a better understanding of sport participation from a life-course perspective, the proxy-variable age should be studied in more detail (Breuer & Wicker, 2008). A more substantive interpretation of age factors can be provided
16 | Chapter 1 by major life events, as they oftenmark important life transitions and represent physical, psychological, social and/or economical changes in a person’s living situation (Bartley et al., 1997; Heikkinen, 2010; Hirvensalo & Lintunen, 2011; Tiessen-Raaphorst et al., 2010). Therefore, looking at major life events can provide tangible explanations for the relationship between sport activity and age (Breuer & Pawlowski, 2011), and investigating their impact is of major importance for making sense of differences and changes in sport participation over the life course (Haycock & Smith, 2018; Hirvensalo & Lintunen, 2011). Earlier research indicates that life events indeed affect sport participation (Allender et al., 2008; Engberg et al., 2012; Gropper et al., 2020; Hirvensalo & Lintunen, 2011). Yet, most of the studies assess the impact of major life events on physical activity. Gropper et al. (2020) extensively and thoroughly reviewed these studies, concluding that events that are primarily associated with decreases in physical activity are starting cohabitation, getting married, pregnancy, parenthood, school transitions, and entry into the labour market, whereas retirement is associated with increases in physical activity. However, in these studies it remains unclear if the findings are accurate for sport participation in particular. This is because sport participation is part of but not distinguishable from the physical activity assessment or not even accounted for in the measurement of physical activity, as studies focused on physical activity domains that include sport participation (e.g., leisure-time physical activity and total physical activity) or investigated other physical activity domains excluding sport participation (e.g., occupational physical activity, active commuting and transport, and domestic activities). Additionally, because earlier research focused on a great variety of different physical activity domains and intensities, findings are inconclusive or mixed and it is difficult to synthesize results in a differentiated, yet concise fashion (Gropper et al., 2020). Few previous studies related major life events to sport participation in particular (Allender et al., 2008; Engberg et al., 2012; Haycock & Smith, 2018; Hirvensalo & Lintunen, 2011). Seippel (2005) found that important reasons for ending organised sport are related to life events, like moving house, family and education. Living on one’s own and having young children was negatively associated with sport participation (Broek et al., 2010; Tiessen-Raaphorst et al., 2010). Humphreys & Ruseski (2009), Nomaguchi & Bianchi (2004) and Ruseski et al. (2011) found that marriage and having children are negatively associated with the time spent on practising sport. In addition, some studies revealed that marriage and having children negatively affect the decision to participate
1 | 17 Synthesis in sport as well (Farrell & Shields, 2002; Nomaguchi & Bianchi, 2004; Ruseski et al., 2011), whereas (Humphreys & Ruseski, 2009) did not find this effect. In contrast, Lee and Bhargava (2004) found that married people participate more in sport activities than unwed people. Furthermore, Nomaguchi and Bianchi (2004) showed that paid work is negatively associated with time spent on sport. Results of the studies by Schönbach et al. (2017) and Sjösten et al. (2012) revealed that the transition to retirement led to an increase in sport participation, and Sjösten et al. (2012) also found that both men and women were more likely of engaging in sport activities in groups than alone after retirement. The state of research regarding the impact of major life events on sport participation is not only limited by the lack of studies specifically investigating the relationship between life events and sport participation. Another notable drawback is that most prior studies are cross-sectional. Generally, sport participation research with a life-course approach has focused on differences between age groups and sometimes on people with different household compositions and employment statuses. Few studies deal with the dynamics in individual sport careers (i.e., when exactly people start, stop or change sport participation) by employing life course data. This generally hindered the investigation of individual changes over time and the drawing of conclusions in terms of causes and effects (Kraaykamp et al., 2013; Lunn, 2010; Pilgaard, 2013; Pulles &Wendel-Vos, 2018). In addition, many studies had limited sample sizes and the potential of nonresponse bias, limiting the generalisation to larger populations (Richards et al., 2019). Further, few studies have examined multiple life events (Engberg et al., 2012). It has thus remained unclear which life events are most important (Hirvensalo & Lintunen, 2011). With this dissertation, I attempt to overcome these limitations. Firstly, by utilising a life course perspective and employing high-quality Dutch life course data. Secondly, by investigating the effects of multiple major life events in different life domains, namely: becoming a student, leaving full-time education, beginning work, moving out to live on one’s own, engaging in an intimate relationship starting to cohabit or getting married, becoming a parent, children leaving the parental home, and retirement. Thirdly, by employing a resource approach to develop a concrete theoretical framework for the impact of these life events. Lastly, by focussing on sport participation, in particular on the likelihood of starting (Chapter 2) and stopping (Chapter 3) a sport in general and in a club setting, differences and changes in sport frequency, number of
18 | Chapter 1 sports and sport setting (Chapter 4), and the underlying mechanisms via which major life events affect sport participation (Chapter 5). THE IMPACT OF MAJOR LIFE EVENTS ON SPORT PARTICIPATION. A LIFE-COURSE PERSPECTIVE AND RESOURCE APPROACH. In this dissertation, I study sport participation from a life-course perspective (Elder Jr. et al., 2003; Shanahan et al., 2016). This perspective concentrates on age-related transitions that are socially created, socially recognised and shared (Hagestad & Neugarten, 1985; Heikkinen, 2010). It has been introduced as a temporal research approach that canpotentially help tounderstanding changes in physical activity behaviours over the lifespan (Hirvensalo & Lintunen, 2011). From a life course approach, individual life courses can be defined as careers (like occupational careers, family careers, sport careers, etc.) that are viewed as a sequence of activities and events both in various life domains and in the attendant institutions and organisations (Engel & Nagel, 2011). From this perspective, the life course is seen as a sequence of life phases characterised by life events that change one’s accustomed pattern of life and daily routine (Holmes & Rahe, 1967). Such major life events (the term used in this dissertation), also known as life change events, transition events, or transitions, can be defined in reference to the MeSH term definition as “those occurrences, including social, psychological and environmental, which require an adjustment or effect a change in an individual’s pattern of living” (MeSH, 1977). According to the working definition put forward by Luhmann et al. (2012), life events “mark the beginning or the end of a specific status. A status is a nominal variable with at least two levels” (p. 4). Life events are thus singular occurrences that lead to a shift from one status to another (Gropper et al., 2020). For example, the transition to adulthood is marked by major life events such as leaving full-time education, beginning work, engaging in an intimate relationship, starting to cohabit, getting married, and birth of the first child (Arnett, 2007; Hirvensalo & Lintunen, 2011; Kilmartin, 2000; Zarrett & Eccles,
1 | 19 Synthesis 2006). These events change a person’s educational, employment, marital and parental status, from a status that is generally more associated with adolescence (e.g., being unemployed, single, childless) to a more adult status (e.g., being employed, married, a parent) (Bell & Lee, 2005). How major life events affect sport participation may be understood by looking at changes in the availability of resources associated with each event (Hirvensalo & Lintunen, 2011; Holmes & Rahe, 1967). In the neo-Weberian tradition of looking at social action from a resource perspective (Bourdieu, 1978; Breen, 2005; Coleman, 1990; Sugden & Tomlinson, 2000; Weber, 1978), life chances – i.e., opportunities and hindrances to access scarce and valued outcomes, like sport participation – are distributed according to the resources available to an individual (Kraaykamp et al., 2013). Major life events and the associated status changes bring new roles and responsibilities, alongside a new organisation of everyday life, altering the temporal, social, physical, mental and/ or economic resources a person can draw on (Bartley et al., 1997; Heikkinen, 2010; Hirvensalo & Lintunen, 2011; Tiessen-Raaphorst et al., 2010). These changes in, for example, leisure time, social capital and support, finances, and someone’s physical or mental state, may trigger periods of socioeconomical adaptation and readjustment and lead to behavioural changes (Gropper et al., 2020; Holmes & Rahe, 1967). Therefore, major life events are expected to affect a person’s opportunities and constraints for participating in sport (Engel & Nagel, 2011; Kraaykamp et al., 2013; Lunn, 2010; Lunn et al., 2013; TiessenRaaphorst et al., 2010), and changes in a person’s sport behaviour during the life course may be interpreted as an alternation of their disposition in the field of sport for adapting to a new configuration of resources (Borgers et al., 2016b; Engel & Nagel, 2011; Pilgaard, 2013). In the empirical chapters, I develop a theoretical rationale for the impact of the life events, by applying this resource reasoning to the specific major life events and sport participation aspects under investigation, using relevant theories like time budget theory and temporal organisation theory (Southerton, 2006) and social motivation theory (Hills et al., 2000). This results in a concrete theoretical framework and expectations per chapter, helping me to guide research processes, and enabling me to embed the empirical findings within the framework for interpretation, increasing scientific thoroughness, reducing the risk of jumping to hasty conclusions, and improving the comparability of the studies. This way I advance upon earlier research, which often lacks theoretical rigour (Breuer & Wicker, 2008; Gropper et al., 2020).
20 | Chapter 1 METHODOLOGY Essential to the methodology that I use in my dissertation, is that I adopt a life course approach and all empirical chapters have a clear longitudinal component or design. I employ a combination of quantitative and qualitative research methods, using either retrospective or prospective individual life course data containing information regarding experiencing major life events and the sport careers of Dutch individuals. This contributes to the literature, as it allows me to accurately consider the occurrence and timing of major life events and differences and changes in sport participation and study the impact of the events. Data InChapter 2and3, I employ individual life-coursedata fromthe SportersMonitor 2010 (Van den Dool, 2010). These data refer to respondents from the GfK ConsumerJury Panel, which adequately represents the non-institutionalised Dutch population aged 15 years and older. Respondents participated through computer-assisted web interviews (CAWI). One module in the online survey registered important aspects of an individual’s life course, such as the exact occurrence of major life events and a complete history of sport participation using retrospective questioning. Questions referred to personal behaviours that were relatively easy to remember, for example, the age at which a major life event occurred, ages of starting and stopping a sport, participation context and frequency. To prepare the SportersMonitor data for analyses, I constructed person-period files in which every record held information on a respondent for a particular year. For example, if the respondent began to work, moved out the parental home, started to cohabit or got married, and became a parent, as well as if he or she started or stopped a sport in that year. For the investigation of the effects of major life events on the likelihood of starting a sport during the life course (Chapter 2), I created person-year records extending from the 18th year of life up to and including the 65th. Respondents who were younger than 65 years at the time they were interviewed were included in the person-year file from their 18th through their age at that time. As respondents were “at risk” of starting a sport in general and starting sport in a competitive club setting in particular in all the years of the person-year file, the risk set used for the analyses comprised all of the records of the 2707 respondents in the person-
1 | 21 Synthesis year file (70,631 person-years in total). For the investigation of the impact of major life events on the likelihood of stopping a sport during the transition to adulthood (Chapter 3), I narrowed down the person-period file and created person-year records extending from the age of 18 up to and including age 35 or the age upon the interview when respondents were younger than 35 years at that time. The risk set used for the analysis of stopping participation in a sport in general consisted of 24,947 person-years in which 2272 respondents were at risk of stopping a sport, because they were practising at least one sport (regardless of the context). The risk set used for the analysis of ending a sport club membership is a subset of the risk set for stopping a sport, consisting of 13,955 person-years in which 1530 respondents were at risk of ending sport club membership, because they were practising one or more sports in a club context. In Chapter 4, I employ data from the 2009 and 2013 waves of the Netherlands Longitudinal Lifecourse Study (NELLS; Tolsma et al., 2014). This is a nationally representative large-scale panel survey of 15–45 year olds in the Netherlands. The fieldwork of the first wave has been done by Intomart GfK. The questionnaire of the first wave consisted of two parts: a face-to-face fully structured interview and a self-completion questionnaire. The fieldwork of the second wave has been done by Veldkamp in a mixed mode: all questions were either administered face-to-face or via a computer assisted web interview (CAWI). To facilitate the study of social dynamics from a life-course perspective, data was collected on a range of topics, like respondents’ life transitions and leisure activities, including sport. This makes NELLS especially suited for my investigation of changes in sport participation and the influences of life events that mark the transition to adulthood. I used information of the respondents who participated in both waves, resulting in a sample of 2317 respondents for the analyses of the number of sports and sport frequency. For the analysis of the likelihood to switch from a club setting to a “lighter” setting or not practising sport at all, I restricted the sample to the 522 individuals who practised sport (mostly) in a club setting in the first wave. In Chapter 5, I draw from primary narrative information from 46 Dutch adults, gathered in two rounds of interviews. In a first round, narrative interviews were held with 19 adults (10 women and 9 men) aged between 29 and 77. To reach saturation of the information with respect to sport participation and life events related to the transition to adulthood in particular, more data on this phase was obtained in a second round of interviews with 17 mothers and 10
22 | Chapter 1 fathers with at least one young child (aged 6 months through 8 years). For the individual narrative interviews, semi-structured guides were constructed based on its specific requirements for the application of a narrative method. The interview guides aimed at activating a storytelling mode in the participant and inviting them to share their insiders’ perspectives on how they experienced and dealt with the occurrence of life events in relation to sport participation, and the dynamics of actors, resources, and constraints regarding practising sport, particularly during the transition to adulthood. This resulted in a data set consisted of 1082 narratives providing insights the mechanisms underlying the effects of life events on sport participation during this transition. Measuring sport participation Sport participation is the key dependent variable in this dissertation. In all empirical chapter’s sport participation is operationalised as a dynamic variable that is measured over time. To get a broad and comprehensive understanding of the effects of major life events, sport participation is measured in various ways throughout this thesis: from starting a sport to stopping a sport, in general and in a club setting, and more subtle changes in between. In the first two empirical chapters, I utilise the retrospective life course data from the SportersMonitor 2010 to measure starting a sport in general and starting a competitive club sport (Chapter 2) and stopping a sport in general and ending a sport club membership (Chapter 3) in a given year. For starting a sport, the value of 1 was assigned in the years in which a respondent started a sport, and a 0 in all other years. For starting a competitive club sport, I assigned a value of 4 to the years in which a respondent started a sport in a competitive club setting. I assigned a value of 3, 2 and 1, respectively, to the years in which a respondent started a sport in a recreational club setting, some other formal setting and on an informal basis. In the years in which no sport was started, I assigned a 0. In the years in which respondents started two or more sports in different organisational forms, I assigned the value for the most intensive or “heavy” organisational form. Stopping a sport was similarly coded as starting a sport, scoring 1 in years in which a respondent stopped practising a sport, and 0 in years they did not stop practising sport. Ending a sport club membership was scored 1 for years in which a respondent ended membership in a sport club, and 0 in all other years. In Chapter 4, I turn to more subtle changes in sport participation, utilising the
1 | 23 Synthesis longitudinal panel data of NELLS to measure the number of sports practised by respondents, their frequency of sport participation and the setting in which they practised sport at two points in time with a four-year gap. The number of sports was measured as the sum of all the sports a respondent participated in, ranging for 0 to a maximum of 10 sports. To measure sport frequency, I first scored every sport by recoding the frequency categories “not at all” as 0, “less than once a month” as 0.5, “1 to 3 times per month” as 2, and “4 times or more per month” as 4 for every sport. Then I constructed a scale, from 0 to 40, adding the scores for all 10 sports. Sport setting is measured based on the setting in which respondents usually practised the sport they participated in most often. It consists of de following categories: (0) club setting, (1) commercial or alternative setting, (2) informal group setting, (3) individual setting, and (4) not practising sport at all. In the fifth chapter, I focus on the considerations, choices and changes that people make regarding their sport participation during the transition to adulthood, particularly when confronted with major life events and associated changes in resources. I turn to lived experiences based on a narrative approach, providing people’s insiders’ perspectives on their sport participation and how and why this was affected by major life events. This illuminates the underlying mechanisms via which major life events affect sport participation during the transition to adulthood, and in what way they do so, revealing differences and changes in people’s resources, their willingness to participate in sport, and ways found to do so.
24 | Chapter 1 EMPIRICAL STUDIES The following chapters of this dissertation present four empirical studies, each contributing to a better understanding of the impact of major life events on sport participation during the life course, especially during the transition to adulthood. As such, these chapters can be read as standalone articles. This does mean that there is some overlap in the presentation of the theoretical starting point, and similar major life events are investigated. However, not every empirical chapter covers exactly the same life events, due to employing different data sets and methods, and differences in the scope of the studies (in particular, chapter two has a wider scope compared to the other empirical chapters). Table 1.2 presents an overview of the data and method used in every empirical chapter, as well as the life events and sport participation aspects under investigation. A new stage of life, a new sport activity? Event history analyses of the impact of major life events on starting a sport (chapter 2) Chapter two is the first empirical chapter and examines if beginning to work, moving out to live on one’s own, starting to cohabit or getting married, the birth of one’s first child, children leaving the parental home and retirement play a role in picking up new sport activities and the organisational form of the chosen sport activity. The central research question is as follows: to what extent are starting a sport in general and starting a competitive club sport affected by major life events? To understand how major life events relate to sport participation, I employ a resource perspective considering that life events lead to increased or decreased resources and constraints to practise sport. I describe, in detail, a theoretical rationale regarding time constraints and social resources to explain how the six major life events under investigation could help or hinder starting a sport in general and starting a competitive club sport in particular. To test the hypotheses deduced from this theoretical framework, I use retrospective life-course data on 2707 adults from the Dutch Sporters Monitor 2010 of the Mulier Institute and NOC*NSF, which traced sport careers and additionally offers information regarding the age at which the six major life events occurred. Event history analysis shows that the odds of starting a new sport increased when people started a paid job, moved out to live on their own,
1 | 25 Synthesis Table 1.2 Overview of empirical studies MAJOR LIFE EVENTS Becoming a student Leaving full-time education Beginning work Moving out to live on one’s own Engaging in an intimate relationship Starting to cohabit or getting married Becoming a parent Children leaving the parental home Retirement Starting a sport in general and in a club setting (Chapter 2) SportersMonitor 2010 DATA METHOD SportersMonitor 2010 Netherlands Longitudinal Lifecourse Study (NELLS) 2009-2013 Primary narrative data (own collection drawn from interviews) Event history analyses Event history analyses Multilevel “withinbetween” modelling & Multinomial logistic regression analyses Micro-level analysis of lived experiences based on a narrative approach Stopping a sport and ending club membership (Chapter 3) Number of sports, sport frequency and sport settings (Chapter 4) Why and how sport is affected by major life events (Chapter 5) when their children left home and upon retirement. The odds of starting a sport decreased when their first child was born. Finally, becoming active in a competitive club sport was stimulated by moving out and retirement, but inhibited by cohabitation or marriage. Based on these results, I conclude that experiencing major life events has consequences for sport participation, in terms of increasing or decreasing the likelihood of starting a sport in general, and more particularly, starting a sport in a demanding organisational form, that is, a competitive club setting.
26 | Chapter 1 When do young adults stop practising a sport? An event history analysis on the impact of four major life events (chapter 3) The third chapter builds on the previous one by examining the relationship between major life events and dropping out of (instead of picking up) a sport in general and in a club setting in particular. So, after investigating if major life events help or hinder “desirable” sport behaviour (i.e., starting a sport; chapter 2), I now study the role of major life events on the risk of exhibiting “undesirable” sport behaviour (i.e., stopping a sport). Additionally, I focus on the transition to adulthood, since this is, for many, a period of key individual life events in which, typically, the stability (also called “tracking”) of physical and sport activity is found to be quite low and we see a sharp decline in sport participation. The main research question is: to what extent are stopping a sport and ending a sport club membership affected by major life events among young adults? I focus on beginning to work, starting to live on one’s own, starting to cohabit or gettingmarried, and the birth of one’s first child. These life events are specifically chosen because they are commonly experienced by young adults, and they represent major transitions in the process of becoming an adult in the life domains of employment, residency, personal relationships and parenting. Building on the theoretical rationale discussed in chapter two, I employ a theoretical framework in the neo-Weberian tradition of understanding social action from a resource perspective. Based on changes in temporal and social resources associated with the occurrence of the major life events, I argue that they increase a young adult’s risk to stop practising a sport in general and to end a sport club membership in particular. To test the theoretical expectations, I again employ detailed retrospective lifecourse data from the Dutch SportersMonitor 2010, now to reconstruct the sport careers histories and occurrence of the aforementioned life events between the ages of 18 and 35. This resulted in two risk sets. The first consists of the years in which 2272 respondents were at risk of stopping sport participation, because they participated in at least one sport (regardless of the context). The second consists of the years in which 1530 respondents were at risk of ending sport club membership, because they were practising at least one sport in a club context. Event history analyses indicate that the risk to stop practising a sport increases when young adults begin to work, move out to live on their own, and start cohabiting or get married, as expected. The risk of ending a sport club membership rises when young adults start to live on their own and when they cohabit or get married. The birth of the first child increases the risks of both
1 | 27 Synthesis stopping a sport and ending club membership for young women, but not for young men. The overall findings suggest these four major life events that mark the transition to adulthood play a role in explaining the sport participation of young adults, as their occurrence is associated with an increased likelihood of dropping out of a sport and/or ending a club membership. The transition to adulthood: A game changer!? A longitudinal analysis of the impact of five major life events on sport participation (chapter 4) Besides that the occurrence of major life events can help or hinder starting a sport over the life course (chapter 2) and increase the likelihood of stopping a sport and/or ending a club membership among young adults (chapter 3), they might also induce more subtle changes in sport participation. In this fourth chapter I examine if this is the case during the transition to adulthood, by addressing the following research question: to what extent do major life events that mark the transition to adulthood affect (1) the number of different sports practised by individuals, (2) the frequency of sport participation, and (3) the switch from practising sport (mostly) in a club setting to practising sport in other informal or organisational settings, or to not practising sport at all? The fivemajor life events under investigation are: leaving fulltime education, beginning to work, engaging in an intimate relationship, formalising a relationship through cohabitation or marriage and becoming a parent. I apply the same resource approach as in the previous empirical chapters to explain the effects of these life events. In line with this approach, I reason that a reduction of temporal resources and increase of professional and social obligations related to the occurrence of major life events induce people to re-evaluate and seek for alternative ways to continue sport participation during their transition to adulthood. This leads me to expect that the major life events reduce the sport frequency and number of sports practised and increase the likelihood of switching from a “heavy” club setting to a “lighter” setting (or dropping out of sport altogether) during the transition to adulthood. The expectations are empirically tested by employing panel data (2009 and 2013) from the Netherlands Longitudinal Lifecourse Study (NELLS). It provides longitudinal information on education, employment, relationship, civil/marital status, and parenthood for 2829 Dutch citizens (ages 15–45) and their sport behaviour, which makes it possible to distinguish between-differences ofwww.proefschriften.net