Social stratification of physical activity. An exploration into how logics of practice affect participation in movement culture

ABSTRACT

Background

Researchers have long studied the social stratification of physical activity patterns in terms of ‘determinants’ of physically active lifestyles. In this article, we set out to explore how Bourdieu’s concept logic of practice can be used as an intermediating analytical tool to promote understanding rather than the calculation of human participation in movement culture.

Theoretical framework

The logic of the practice is similar to the logic of the logician, but it differs in terms of temporality and directionality. Specifically, where the logician needs not take notice to such worldly matters, they are inevitably inherent in social life. Thus, logics of practice guide how practice is played out as well as, on the level of the unconscious, the attitude of its participants towards themselves, the practice, and each other. Different logics of practice, as previously theorised by Engström, may ‘taste’ differently to people depending on their habitus, that is, their embodied dispositions. In this article, such logics of practice of movement culture are investigated further.

Purpose

The purpose of the article is to investigate how gender and social position relate to participation in movement cultures that are guided by different logics of practice.

Method

In 1968, just over 2000 Swedish teenagers who were born in 1953, answered a questionnaire about their participation in movement culture. New data collections took place in 1973, 1978, 1983, 1994 and 2006/2007. The current paper is based on data collected from 846 individuals, now 63 years old, who chose to participate in a new data collection. Data about self-reported participation in movement culture, and the stated taste (or dis-taste) for certain logics of practice in movement culture, was calculated in relation to gender, level of education (cultural capital) and self-reported class affiliation (economic capital).

Results and conclusions

The statistical analysis indicates that while the logic of performing in this study indicates masculine and upper-class values, the logic of experiencing indicates feminine and academic values. The logic of improving seems to hold a middle-ground position. Importantly one of the logics, experiencing nature, which is actually also the one that most of the 63 year-olds express a taste for, seem not to be linked to either gender or social position. We conclude that this may have important societal and pedagogical implications, for example, that increasing the possibilities to experience nature could facilitate participation in movement culture at least among Swedish upper middle-aged people.

KEYWORDS:

• Physical activity
• social stratification
• movement culture
• logics of practice
• follow-up study

Introduction

Physical activity patterns among populations have long interested researchers (e.g. Banavali et al. 2020; Carballo-Fazanes et al. 2020; Caspersen, Pereira, and Curran 2000; Dishman, Sallis, and Orenstein 1985). A special interest has been directed towards the relationship between physical activity habits during childhood and the propensity to be physically active later in life (e.g. Cleland, Dwyer, and Venn 2012; Dennison et al. 1988). Similarly, a great deal of interest has been directed towards the relationship between gender, social background and physical activity habits (e.g. Kenyon and McPherson 1973; Scheerder and Vos 2011; Wilson 2002). In this research, the relationship between gender, social position and physically active lifestyles is frequently approached in terms of ‘determinants’ and ‘correlates’ (see, e.g. Bauman et al. 2012; Biddle, Mutrie, and Gorely 2007), which suggests a mechanical relationship similar to natural laws. We suggest that such a perspective risks missing out on important cultural and contextual aspects of social life, such as the ‘what, who, when and where’ of physical activity. Moreover, we also believe that mechanistic understandings may contribute to stigmatise sedentary lifestyles (cf. Stearns 2002). In this article, we further explore a cultural sociological approach to understanding physical activity patterns previously suggested by Engström (2008).

Nearly fifteen years ago, Engström (2008) outlined a more indirect, that is, non-mechanistic, approach to the question ‘who is physically active?’ His account was based on results from a 38-year follow-up study about cultural capital and sport participation from adolescence to middle age. Inspired by Pierre Bourdieu’s cultural sociological work (1978, 1988, 1990), Engström tentatively discussed participation in movement culture based on some thoughts about what he termed ‘sporting practices’. Sporting practices were ‘different principles or logic within the field of sport that give an action meaning to participants’ (Engström 2008, 322). Later, Engström endeavoured to develop his thinking about sporting practices, but now under the designation ‘logics of practice in movement culture’ (Engström, Redelius, and Larsson 2018). The main idea in that theoretical piece was to more clearly conceptualise the different logics of practice and hypothesise about how they might relate to the social stratification of movement culture in terms of gender and social position. The purpose of this article is to empirically investigate Engström’s ideas concerning a possible relationship between the social stratification of movement cultures (regarding gender and social position) and the different logics that guide practice within those movement cultures.

Participation in movement culture – a cultural sociological perspective

Much of the mechanistic research about physical activity patterns are rooted in the World Health Organization’s (WHO) definition of ‘physical activity’ as ‘any bodily movement produced by skeletal muscles that require energy expenditure’. Such a definition places focus on the function of physical activity, which may explain also the use of mechanistic concepts such as ‘determinants’ and ‘correlates’ (see, e.g. Bauman et al. 2012; Biddle, Mutrie, and Gorely 2007). Research about physical activity patterns has consistently reported that not all human beings develop a physically active lifestyle enough to support their health (Bauman et al. 2012). Based on the functionalist and mechanistic approach, researchers then typically emphasise that people ‘need to be more physically active’ or that they ‘need behavioural changes’ (e.g. French et al. 2014; Rhodes and Pfaeffli 2010).

The focus on ‘behaviour change’ is interesting given that few people, at least those who are living in industrialised countries, could have missed the message that physical activity is beneficial for their health. Arguably, most people, regardless of gender and social position, would be able to come up with a wide range of reasons why they would benefit from physical activity. Yet, differences in physical activity patterns seem to relate strongly to both gender and social position. While middle-aged women and men are physically active to the same extent, at least within a Swedish context, they seem to be devoted to different keep-fit activities (Engström 2008). Regarding social position, put simply, the more privileged the position, the more likely a person is participating in movement culture. For example, according to Wilson (2002, 5) ‘[s]tudies have repeatedly shown that indicators of social class are positive predictors of sport involvement in general and that members of the upper classes are more likely to be both sports participants and sports spectators’ (see also Shaw et al. 2010). No doubt, some type of connection between physical activity, gender and social position is evident in modern societies. While there is currently a connection between social position and having a fit body, Stearns (2002) has shown that this situation is very different from previous times. Where a century ago being overweight and pale used to connote a privileged social position, today being fit and tanned connotes the same position. A result of this social stratification of the fit body is that a lot of people do not develop a sufficiently physically active lifestyle to support their health. However, we do not yet adequately understand what produces the social stratification.

In this article, we develop further a cultural sociological perspective of social stratification of physical activity patterns (Bourdieu 1978, 1988; Engström 2008). A basic premise of this perspective is that physical activity’s function regarding energy expenditure does not necessarily affect how people experience and relate to physical activity. While different physical activities may have the same physiological effect on the body, they may at the same time be experienced differently. Consequently, they could be regarded as different phenomena. Put simply, physical activity is experienced differently depending on who is physically active, when and where the activity takes place, and what logics are guiding the practice. However, knowledge is still limited regarding the relationship between the ‘who’, ‘when’, ‘where’ and ‘what’.

In order to more clearly signal the diversity of physical activity, from now on, we will use the concept movement cultures, and rather than talking about ‘being physically active’, we will refer to participation in movement cultures. Movement cultures are the ways in which social groups deal ‘with the need and desire for movement beyond labour or maintaining life’ (Crum 1993, 341). Different movement cultures are characterised by different norms, values and practices. As human beings interact in different movement cultures, they (learn to) make use of different collections of cultural ‘tools’ that serve as the means for the performance of action (Sewell 2005, 46). These tools are not coherent systems of symbols and meanings. Rather, they are more or less multi-dimensional, and as such, they enable actors ‘to play on the multiple meanings of symbols – thereby redefining situations in ways that they believe will favour their purposes’ (Sewell 2005, 51). This is why ‘to be physically active’ should be viewed as a sign of to what extent humans desire and manage to master the cultural tools that serve as means for participation in movement cultures. Of course, humans can also move on their own, but moving on one’s own is still participation in movement culture. Solitude does not mean that you are unaffected by the social conventions that are linked to social practices.

Highlighting the need and desire for movement beyond labour or maintaining life means that certain practices that include physical activity, such as occupational physical activity, some household activities (cleaning, gardening, home repair, etc.) and self-powered transport (walking to and from work and other places), are not included (cf van Tuyckom and Scheerder 2010) in the concept of participation in movement culture. Obviously, the distinction between participation in movement culture in this sense, and the physical activity associated with employment and daily chores, is not clear-cut. It contradicts the messiness of social life. Rather, it is brought forward by methodological necessity.

In the study, we pay particular attention to gender and social position. Overall, we regard both gender and social position as social constructions (Bourdieu 1978, 2001; Lorber 2018). By the term gender, we refer to ‘the constellation of characteristics and behaviours which come to be differentially associated with and expected of men and women in a particular society’ (Burr 1998, 11). This means that it is through certain characteristics and behaviours, certain ways of moving and relating to movements, for example, that people become recognised as ‘real’ women and men, respectively. Such characteristics may, however, vary in relation to, for example, social position, which is in turn typically defined in terms of occupation (and occupational income), educational status and similar variables (see, e.g. Scheerder and Vos 2011; van Tuyckom and Scheerder 2010; Wilson 2002). Both gender and social position may imply a static condition, i.e. that a person ‘has’, once and for all, a certain gender and social position. Conversely, we emphasise that gender and social position are resulting from continuously ongoing processes, which contribute to produce gendered and classed bodies. Moreover, gender and social position must be defined relationally. Consequently, what is considered to be feminine is what is at the same time considered not to be masculine, and vice versa, and what is considered to be high-class is what is at the same time considered not to be low-class, and vice versa. Both temporality and the relational character of the categories are key aspects also in Bourdieu’s conceptualisation of habitus, that is,

necessity internalized and converted into a disposition that generates meaningful practices and meaning-given perceptions; it is a general, transposable disposition which carries out a systematic, universal application – beyond the limits of what has been directly learnt – of the necessity inherent in the learning conditions. (Bourdieu 1984, 170)

In short, habitus designates embodied dispositions of social tastes, mores, and actions that are developed over time (see also Engström, Redelius, and Larsson 2018), and it conveys the dimensions of gender and social position seen as a life trajectory.

One important feature of Bourdieu’s work with social class is that classes are not only organised vertically in terms of upper, middle and lower classes. Bourdieu (1984) suggests that also a horizontal axis must be acknowledged. This horizontal axis is framed in terms of access to cultural or economic capital, where some social groups have access to cultural capital but not so much to economic capital (as is the case in the cultural sector), while other groups have access to economic capital but not necessarily to cultural capital (as is the case in the business sector). Some time ago, Palme (1990), used this notion to analyse the taste for different sports among what he called the upper middle class. Palme found that upper middle class families with access to cultural capital had developed ‘a suspicion against, for example, sports that honours competition and toughness (e.g. ice hockey), or which require large financial investments in their own person (golf, alpine skiing)’ (277; our translation). Upper middle class families with access to economic capital, on the other hand, ‘highlight ice hockey as an exemplary sport that fosters go-ahead-spirit and the ability to cope with hard times’ (280; our translation; see also Karp 2000). Palme’s study indicates that the relationship to participation in movement culture may differ not only on the vertical axis (upper, middle and lower class), but also on the horizontal axis (high cultural and economic capital, respectively). Accordingly, in this study, we have endeavoured to gather information that will allow analysis of the vertical as well as the horizontal dimension of social position.

Finally, the cultural sociological perspective may contribute to problematise the focus on ‘behaviour change’ in research about physical activity patterns. In everyday thinking, economic inequities sometimes explain social stratification. Participation in (some) movement activities is simply taken to be too expensive. If this were true, there is obviously a range of activities, quick walks and jogging, for example, that would seem accessible to most people. Within research, however, other reasons are also suggested, such as dearth of ‘cultural capital’, that is, a lack of ‘appropriate preferences and tastes as well as skills and knowledge’ (Wilson 2002, 6; see also Bourdieu 1978, 1988). This perspective offers one way of how what Bourdieu (1984, 2001) called symbolic violence can be avoided when it comes to understanding social stratification of physical activity patterns. Symbolic violence is ‘a gentle violence, imperceptible and invisible even to its victims, exerted for the most part through the purely symbolic channels of communication and cognition (more precisely misrecognition), recognition, or even feeling’ (Bourdieu 2001, 1). It means that certain characteristics and behaviours, which are incorporated into the taste of privileged social groups, and as such work as markers of social distinction from other social groups, are nevertheless considered desirable for all people. Symbolic violence may then result in a stigmatisation of people who fail to live up to what is deemed ‘healthy’ (cf. Bourdieu 1984). If we do not increase our understanding of the connections between social position and participation in movement cultures, ambitions to support physically active lifestyles will fall short, and, what is worse, they may increase stigmatisation of non-participation (Stearns 2002). Consequently, there is a great need for deeper and more nuanced understanding of the social stratification of movement cultures.

A 50-year follow-up study

The article is part of a longitudinal research project that commenced in 1968 with a first data collection among a cohort consisting of 2144 Swedes who were born in 1953. Although Engström at that time used the concept ‘physical activity’, we believe that he was dealing rather with participation in movement culture (cf. Engström, Redelius, and Larsson 2018). His concern was not about the functional aspects of physical activity, but about the social aspects (which may still have functional consequences), that is, ‘who’ – in terms of what gender and social position – develop a taste for participation in those movements cultures that are available in contemporary society. Engström found that, after the school years, participation in movement culture was highly gendered among people who were born in 1953. Gradually, however, the women in the study returned to movement cultures with increasing age and in their 50s they were as active as the men of the same age, but in somewhat different activities (Engström 2008). Concerning social position, the taste for participation was much related to growing up in a privileged home and remaining for the rest of their lives in the same social stratum (Engström 2008; see also Wilson 2002). For instance, the relationship between participating in exercise activities in midlife had a much stronger relationship with academic education than with participation in club sports during adolescence (Engström 2008). Engström’s understanding of this was that although both could be regarded as ‘physical activity’, in the lived experiences of individuals, keep-fit exercises in midlife have little to do with competitive sports during adolescence.

In his 2008 publication, Engström outlined a theoretical framework that could facilitate an understanding of the social dynamics that underpin social stratification of movement cultures. This framework was based on Bourdieu’s cultural sociological perspective, specifically his concept logic of practice (Bourdieu 1977, 1990). The framework was subsequently systematised in a new publication, where Engström and colleagues (2018) suggested that ‘[m]ovement activities, like golf, tennis, skiing, and so on, may be perceived as being the same regardless of context simply because they are nominally the same’ (894). Conversely, one activity, such as skiing, could be quite a different sort of practice depending on in what context it occurs:

One can participate in an organised competition (competing), practice skiing for the purpose of fitness training (training), attend a ski school in order to learn the skill (practising), glide along a ski track for recreation and fun (playing) or make an adventurous attempt to climb a mountain (conquering). These different forms of doing cross-country skiing have various meaning; hence, they appeal to different individuals and also require the individual to make different forms of investments. (Engström, Redelius, and Larsson 2018, 901)

What, then, guides the practice of, for example, skiing in different contexts? According to Engström (Engström 2008; Engström, Redelius, and Larsson 2018; see also Bourdieu 1988) the answer to this question is: the logic of the practice. The logic of the practice is similar to the logic of the logician, but it differs in terms of temporality and directionality (Bourdieu 1990, 86). Where the logic of the logician takes no notice to such worldly matters, temporality and directionality are inevitably inherent in social life. Thus, the logics of practice guide how practice is played out as well as the attitude of its participants towards themselves, the practice and each other. Simultaneously, they create distinctions between those who have acquired a taste for the practice, and those who have not, which means that some are attracted by certain forms of activity while others may be appalled. It should be noted that the logics of practice are not to be confused with reasons, at least not in the reflexive and rational sense. Logics of practice operate on a pre-reflexive level; they are experienced on the level of the unconscious. Hence, most people will have difficulties with describing how they experience logics of practice in other ways than ‘I (don’t) like it’. This is why logics of practice must be explored with indirect methods.

Engström and colleagues (2018) theorised the occurrence of three overarching logics, performing, improving and experiencing. Designating one of the logics ‘experiencing’ does not mean that participants do not experience things within the frames of the other logics, merely that while the other logics support and engender performances and bodily improvements, the third logic is foregrounding experience as such, as in the cases of playing, experiencing nature and moving to music. Performing can be about competing with others, performing individual feats (conquering) or expressing a certain aesthetic content. Improving can be about subjecting to hard training or developing mastery of certain abilities through practising. In real activities, several logics are present, but one or two can be the dominating – and most valued – ones. The more one-dimensional the activity, the more excluding it may be; and conversely, the more multi-dimensional the activity, the more inclusive it may be.

Engström saw a connection between social position and a propensity to develop taste or dis-taste for participation in movement activities that are guided to varying degrees by different logics of practice. Up until now, data from the longitudinal project has not allowed for any distinct analyses of such relationships. However, the latest data collection included questions that enabled us to explore the said relationship empirically.

Method

As was stated earlier, the follow-up study originated in 1968 when the individuals included in the study were 15 years old. Engström then conducted a survey with 2144 randomly selected youths in different regional parts across Sweden who answered a questionnaire regarding their attitude towards school physical education and their leisure time sport activities. In subsequent years, data collections about participation in movement culture and lifestyle issues took place with the same group on five occasions: 1973, 1978, 1983, 1994 and 2006/2007.

During the winter of 2016 and 2017, the research participants, now 63 years old, were asked to complete a sixth questionnaire about their participation in movement culture and their tastes and dis-tastes for different aspects of movement cultures. The questionnaire was based on a number of questions that Engström used in previous surveys, and a number of questions that had sprung out of Engström and colleagues’ (2018) work on logics of practice in movement culture. It was sent out by post, but it was also possible to answer online. The original mailing was complemented twice by a reminder in the form of a postcard to those who had not responded. In addition, the questionnaire was sent out a second time to those who had not responded. After a period of two months, the researchers made phone calls to the approximately 300 research participants who had still not responded. Overall, our efforts resulted in 846 individuals submitting complete questionnaires, i.e. a response rate of 56 per cent. This means that the dropout rate was larger than in any of Engström’s previous data collections. We discuss this issue further under the subsequent section (Table 1).

Table 1. Number of participants in the survey, from 1968 to 2016.

Year	Population	Participants	Dropout
1968	2464	2144	13%
1973	2105	1749	17%
1978	2067	1609	22%
1983	2072	1675	19%
1994	2007	1657	17%
2006/2007	1979	1518	23%
2016	1533	1081/846	29/45%

Dropout analysis

The general trend is that over time more and more of the research participants have chosen to disengage from the study. This trend can be explained in several ways. Some research participants have died or moved abroad, while others for various reasons have opted not to participate. When phoned by us, some indicated that in the early years they saw participation as ‘mandatory’, while later they realised that participation in research is actually voluntary. This development is in tandem with the fact that the present-day information to research participants more clearly emphasises that participation is voluntary and that the participants may at any time choose to disengage from the research. A number of research participants refrained from participating because they were ‘flooded by market surveys of different sorts’, thus indicating a difficulty to distinguish between market research and ‘proper’ research. Finally, some told us that they were ‘too old for this’ (i.e. participate either in the research or in movement cultures).

The relatively high dropouts among the 63-year-olds may have affected the results of the study. The drop-outs include a larger proportion of men than women, as well as a larger proportion of people who do not have higher education. This indicates that the dropout is weakly selective, that is, the data in the 2016/2017 survey is somewhat less representative compared with previous data collections. In order to make an estimate of how this relationship may have impacted on the results, a comparison was made between data from participants and non-participants in 2006/2007 and 2016/2017 respectively regarding a number of selected variables (membership of a sports club and degree of participation in movement culture). The analysis showed differences between participants and non-participants in the 2016/2017 study, indicating that respondents were slightly more physically active in comparison to non-respondents. This means that some caution should be observed in cases where comparisons are made regarding changes over time.

Measurements

In the study, we have used a number of social categories that we want to elaborate on. The category gender is self-reported. While gender is not necessarily a straightforward category, due to space limitations we will not elaborate on this issue here. We will merely re-emphasise that we see gender as a social construction, meaning that the characteristics and behaviours that are in this study found to statistically relate to women and men are not to be seen as essentially feminine or masculine. Rather, they designate what values are typically ascribed to the genders in this particular historical, cultural and social context.

Regarding social position, Bourdieu (1984) showed that education can be an important marker of this category, an insight used also in this study. Since previous data collections, we have information about the participants’ educational level in terms of: elementary school education (school years 1–9, ages 7/8–15/16), secondary school education (school years 10–12, ages 16/17–18/19), and higher education (university studies). Like Bourdieu (1984), we assume that level of education is a measure of cultural capital.

While cultural capital presumably focuses more on the horizontal dimension of social class (Bourdieu 1984), we also wanted information about the vertical dimension of social class. Thus, in line with Svallfors (2004), we asked the research participants to assess their own social position. To investigate the participants’ self-reported class affiliation, we posed the following question in the questionnaire: ‘What social class do you think you belong to now?’ The following answers were available:

• Working class

• Between working and middle class

• Middle class

• Between middle and upper class

• Upper middle class

• Between upper middle and upper class

• Upper class

• Do not know

For the benefit of statistical analysis, three groups were constructed based on the seven alternatives. The two categories Working class and Between working and middle class were designated Working class; the three categories Middle class, Between middle and upper middle class, and Upper middle class were designated Middle class; and finally, the two categories Between upper middle class and upper class, and Upper class were designated Upper class. We assume that self-reported class affiliation mirrors economic capital (the vertical dimension of social class).

Participation in movement culture. Since the follow-up started in 1968, Engström has used the following question to assess participation in movement culture in the studied cohort: ‘To what extent have you participated in physical activity in your free time during the last year? Tick the option that suits your description best:

• I have moved very little.

• I have moved very little but have taken occasional walks or similar.

• I have had daily exercise while doing everyday chores such as cleaning, gardening, dancing, recreational walks, cycling to and from work, dog walking, etc.

• I have participated in lighter forms of exercise like walking (or other activities with equivalent effort) at least once a week.

• I have participated in more exhausting exercise such as fast walking, jogging, swimming, gymnastics or equivalent at least once a week.

• I have regularly participated in hard training or competition where the physical effort has been extensive, for example running and different ball games’.

In the analysis, we merged the first two options, rendering five levels of participation in movement culture.

Logics of practice. Above, based on Engström, Redelius, and Larsson’s (2018) elaboration, we suggested that there are three overarching logics of practice: Performing, Improving and Experiencing. Each of these logics can be divided into two or three sub-categories: (a) Competing, Conquering and Expressing, (b) Training and Practising, and (c) Playing, Experiencing nature and Moving to music. Since Engström (2008) divided between Physical training (as in fitness/gym training) and Movement and concentration training (as in tai chi, yoga, etc.), we decided to test dividing the category training in the same way in this study. This is not because of any rational considerations that fitness/gym training and tai chi, yoga and similar activities are different, but rather that Engström (2008) pointed out that these activities may ‘taste’ differently and thus appeal to different participants.

Each of the different logics were broken down into three statements that highlight key features of the logics. For example, regarding Competing, the respondents were asked to assess their spontaneous relation to the following statements: ‘I like when physical achievements can be measured’. ‘I like when my achievements can be compared with the achievements of others’. ‘I like competition’. The respondents were asked to assess their relation to the statements on a five-degree Likert scale which went from ‘I agree completely’ (5 points) to ‘I do not agree at all’ (1 point). The analysis of each logic was based on the sum (3-15 points) of the answers of each logic.

Statistical analysis

Throughout the article, we use descriptive statistical analysis, i.e. cross tabulations. In tables and figures, the unit is given in percentages. In cases where material has been subject to significance testing, this has been calculated using chi² and the result is presented in three levels (see also tables and figures): p < 0.05; p < 0.01 and p < 0.001. These are marked with one, two and three asterisk characters respectively (*).

Results

Overall, the results from this study differ only marginally from Engström’s (2008) study of the same cohort some ten years earlier.

The results in Figure 1 indicate that the majority participate in some form of light to moderate exercise activity. Few move ‘very little or not at all’, or are involved in strenuous exercise. At the age of 63 of all the 846 participants, 41 per cent of the women and 39 per cent of the men correspond to what Engström (2008) designated as ‘exercisers’ (the last two of the five alternatives). This is a marginal decrease compared to ten years earlier (Engström 2008; see Figure 2).

Figure 1. Proportion of exercisers at the age of 63. Women and men. Percent. N = 846.

Figure 2. Proportion of exercisers (at least once a week corresponding to fast walks, jogging, swimming, keep-fit gymnastics or equivalent training). 15–63 year of age.

Overall, the most common activity reported by the participants at age 63 is walking, but among the exercisers gym/weight training, keep-fit gymnastics, cycling, running/jogging and swimming are common activities.

The relationship between education and participation in movement culture is strong within the studied cohort. The highest proportion of participants is to be found among those with the highest education (university education), almost 50 per cent, while in the group with the lowest education (elementary school education) just over 20 per cent participate. A strong relationship is evident also between participation and class affiliation. The highest proportion of participants (55 per cent) is found in the group who define themselves as belonging to the Upper class and the lowest proportion (just over 20 per cent) among those who define themselves as belonging to the Working class.

Let us now move on to what we believe is the unique contribution of this particular data collection, namely the exploration of how gender and social position relate to the participants’ taste for movements guided by different logics of practice.

Taste for movements

The questionnaire included questions concerning the extent to which the research participants, now 63 years of age, could agree with different statements about the logics of practice (i.e. competing, conquering, expressing, physical training, movement and concentration training, practising, playing, experiencing nature and moving to music). In this analysis, to increase the reliability of the statistical analysis, we have included the rather large proportion of the research participants (32% of the women and 33% of the men) who reported that they had participated in lighter forms of exercise like walking (or other activities with equivalent effort) at least once a week (Figure 3).

Figure 3. Proportion of women and men who are attracted by training corresponding to logics of practice in movement culture. N = 613.

As far as this group of 613 participants of movement culture is concerned, the results show that the logic of practice that, by far, most people related positively to, almost 85 per cent of the respondents, was experiencing nature. This would indicate that among Swedish 63 year-olds, regardless of background and present social position, to stay in nature for its own value, or to participate in movement activities in nature without expectations other than that it is pleasant, is an important source of physical activity. Competing, conquering, physical training, playing and moving to music appear tasteful to quite a few (between approximately 30 and 40 per cent), while expressing, movement and concentration training and practising were not valued by most people. Now, we will take a closer look at the relationships between taste for different logics of practice in movement culture and respectively gender, educational level (cultural capital) and self-reported class affiliation (economic capital).

Figure 4 indicates that many of the logics of practice (conquering, expressing, physical training and experiencing nature) do not indicate gender. Women and men indicate to about the same extent that they feel a taste for and participate in movement cultures where these logics of practice prevail. Some logics seem, however, to indicate gender to a greater degree, specifically competing, movement and concentration training, practising, playing and moving to music. Men to a greater extent indicated a taste for competing and practising, while women to a greater extent indicated a taste for movement and concentration training, playing and moving to music. This may suggest that, among these 63-year-olds, competing and practising bodily skills to a greater extent indicates masculinity, while movement and concentration training, playing and moving to music to a greater extent indicates femininity.

Figure 4. Proportion of exercisers corresponding to logics of practice in movement culture in relation to gender. N = 613. Chi² test: *p < 0.05; **p < 0.01; ***p < 0.001.

As Figure 5 indicates, differences referring to level of education are smaller than differences referring to gender. Several logics of practice have no relationship to level of education. Both high- and low-educated people engage to the same extent in movement cultures where competing, conquering, expressing, experiencing nature and moving to music prevail. Some logics stand out, however, as they relate more strongly to level of education. These are physical training, movement and concentration training, practising and playing, which all appeal more to the tastes of high-educated people than low-educated people. Hence, physical training, movement and concentration training, practising and playing can all be said to indicate cultural capital.

Figure 5. Proportion of exercisers corresponding to logics of practice in movement culture in relation to level of education. N = 613. Chi² test: *p < 0.05; **p < 0.01; ***p < 0.001.

Finally, as Figure 6 indicates, as with level of education, many of the logics seem to have no relationship to class affiliation. People from working, middle and upper classes seem all to enjoy competing, physical training, movement and concentration training, playing, experiencing nature and moving to music to the same extent. Again some logics stand out, however, as they relate more strongly to class affiliation. These are conquering, expressing and practising, which all appeal to the taste of upper class people. Hence, conquering, expressing and practising can all be said to indicate economic capital.

Figure 6. Proportion of exercisers corresponding to logics of practice in movement culture in relation to social class. N = 613. Chi² test: *p < 0.05; **p < 0.01; ***p < 0.001.

In summary

The analysis shows that the previous patterns of participation in movement cultures reported by Engström (2008) have changed only marginally over the last ten years. The new knowledge that can be added based on the results presented here is how logics of practice of movement culture sometimes contribute to social stratification of participation.

As is summarised in Table 2, the logics of competing and practising movement skills to a greater extent than other logics indicate masculinity, while the logics of movement concentration training, playing and moving to music rather indicate femininity. Moreover, while the logics of physical training, movement and concentration training, and playing to a greater extent indicate cultural capital, conquering and expressing to a greater extent indicate economic capital, thus mirroring the horizontal dimension of social position. In the case of practising, it seems to indicate both cultural and economic capital.

Table 2. Summary of the results concerning the relationships between gender and social position, and an expressed taste for different logics of practice in movement culture.

	Performing	Training	Experiencing
Men	Competing	Practising
Women		Movement and conc. training	Playing Moving to music
University		Physical training Movement and conc. training Practising	Playing
Upperclass	Conquering Expressing	Practising

The summary shows that the overarching logic of performing among Swedish 63 year-olds indicates masculine and upper-class values (both cultural and economic capital), and that the overarching logic of experiencing indicates feminine values and cultural capital. The overarching logic of improving seems to hold a middle-ground position. Importantly one of the logics of practice, experiencing nature, which is actually also the one that most of the 63 year-olds feel a taste for, seems not to be linked to either gender or social position. We believe that this may have important implications for any attempt to promote physical activity among the population.

Discussion and conclusions

One overall ambition of our work was to test the applicability of Engström’s idea of logics of practice as a tool to empirically, yet more indirectly in relation to other more mechanistic approaches (see, e.g. Bauman et al. 2012; Biddle, Mutrie, and Gorely 2007), analyse social stratification of participation in movement culture. More precisely the purpose of the article was to investigate the relationship between the social stratification of movement cultures (regarding gender and social position) and the different logics that guide practice within those cultures among a cohort of Swedish 63 year-olds. Let us first discuss briefly how we value the applicability of Engström’s idea of logics of practice in empirical analyses, and how gender and social position relate to participation in different movement cultures. Subsequently, we will discuss some societal and pedagogical implications based on the results.

In our view, the empirical study justifies the use of logics of practice as a tool for analysing social stratification of participation in movement culture. Perhaps the analysis presented here does not offer one exhaustive explanation of human participation in movement culture that is often sought after in studies of physical activity patterns (Bauman et al. 2012). However, this was not expected either. Since human action hardly follows universal laws, there is probably not one analytical approach that can offer ‘the whole picture’. We believe, though, that Engström’s perspective can offer new insights about how to understand participation in movement culture.

As was suggested in the introduction, few people could have missed the message that physical activity is crucial to their health. Still, not all human beings develop a physically active lifestyle that is sufficient to support health. Clearly, the results of our study mirror results from previous research (for an overview, see Wilson 2002; see also Scheerder, Vanreusel, and Taks 2005; Scheerder and Vos 2011), that is, the more privileged the social position, the more likely that people participate in movement culture. Based on our analysis we can add important information about how this relationship can be interpreted. We suggest that, at least partly, the social stratification of movement cultures is related to how the logics of practice that govern activities within different movement cultures express social norms and cultural values that are associated with gender as well as both the vertical and the horizontal dimension of social position (see Table 2). Because of this, people of different gender and with diverse social positions may relate differently to the same activities.

In order to understand an individual’s or a group’s lifestyle choices, including participation in movement culture, Bourdieu (1984) suggested that we start from the concept of habitus; a term combining our living conditions with our embodied dispositions. Habitus is reflected in our practices and lifestyles. In this context, Bourdieu talks about taste, which manifests habitus in practice. Through our engagement with certain practices, we signal our sense of belonging while simultaneously distancing ourselves from the taste that is not ours. In this sense, practices become taste markers for social affiliation. Understanding participation in movement culture in this cultural sociological way, rather than in behaviouristic ways where focus is to a great extent on the need for ‘behaviour change’, or in rational terms where focus is on the reasons that people provide to why they are – or are not – physically active, means that physically inactive lifestyles are not to the same extent seen as indicating moral, psychological and physical disorder (Stearns 2002). Instead, we believe that Engström’s perspective can offer critical understanding of how people’s taste – or distaste – for participation in certain movement activities relates to symbolic violence.

Symbolic violence (Bourdieu 1984) is manifested as an imposition of the norms of a group possessing greater social power on those of a subordinate group, for instance, when highly educated middle- and upper-class people patronise people from ‘lower’ social groups because they are not physically active, lazy even (cf. Kreuser et al. 2013). What people from privileged social groups tend to forget is that being fit and physically active are some of the ways par excellence through which middle- and upper-class identities are formed. This means that people in non-privileged groups may not develop a taste for what is deemed ‘healthy for them’. Moreover, should all people participate to the same extent in similar movement activities, these would not ‘work’ as distinction markers.

That people experience a distaste for participation in movement culture may signal that they lack a sense of meaning in the activities on offer. Put in more everyday terms, these activities are then perceived as ‘not for me’. The propensity that more privileged people to a greater extent participate in movement culture can thus be interpreted as that the higher the social position, the more likely it is that the movement activities on offer appeal to someone’s taste and are experienced as ‘for me’. Pedagogically, this is a very important insight.

Understanding participation in movement culture in a socio-critical way may enable active measures when it comes to supporting participation in movement culture. Such measures are not about how to ‘get’ people physically active in certain ways, e.g. through ‘programming’ (e.g. Cohen et al. 2009; Tavares and Plotnikoff 2008), but about considering the breadth of the activities on offer as well as the meaningfulness of these activities. Regarding the breadth of activities, one overall conclusion from this study is that narrow investments in promoting people's participation in movement culture are more likely to be misaligned compared to broad investments simply if activities that are based one-sidedly on a single logic will only attract a limited number of people, that is, individuals from certain social groups (see also Engström 2008; Engström, Redelius, and Larsson 2018). Although efforts to change behaviour among certain groups of people may to some extent be successful, if the goal is to reach many people with diverse backgrounds, and to reduce the risk of stigmatising certain social groups, then the success potential is more uncertain. Judging from the results of our study, broad investments, with a diversified range of activities, have a greater chance to reach a great many people with different genders and diverse backgrounds.

Regarding the meaningfulness of activities, we offer the following elaboration based on the logic of practice of experiencing nature put in relation to other logics. Experiencing nature appeals, at least among upper middle-aged Swedes regardless of gender and social position. Enabling more people of this age-group to have the opportunity to participate in movement activities while experiencing nature could thus be one promising way to promote participation in movement culture among this group. Experiencing nature is not to any great extent linked to so-called distinction value, that is, that 63 year-old Swedes only to a limited extent signal their particular group affiliation by participating in such activities. Further, when offering activities in nature, it may be important not to stress too much, for example, competition or skill development. This could indeed be attractive to men, but maybe not so much to women. Similarly, a relatively greater emphasis of conquering (such as including challenging hardships while moving in nature) could be attractive to upper class people, but it would possibly contribute to working and middle class people staying away. However, since we are not dealing here with mechanistic behaviour patterns, it is important that any pedagogical intervention does not rely too heavily on taken for granted properties among certain social groups.

To conclude, we believe there is still much to learn from the now more than fifty-year follow-up study that Lars-Magnus Engström commenced in 1968, for example about participation in movement culture, the social stratification of that same culture, and how it is possible to think pedagogically inclusive based on insights from the project.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Centrum for Idrottsforskning [grant number P2016-0015].