The concept of ‘well-being’

Concern with young people’s well-being has been the focus of significant research and policy activism for at least three decades in Australia (Hamilton & Redmond, Reference Hall, Smith and Toumbourou2010). Despite this interest, there remains considerable diversity in definitions and application of the concept among researchers (VicHealth, 2015; White & Wyn, Reference White and Wyn2013; Wyn, Cuervo, & Landstedt, Reference Wyn, Cuervo, Landstedt, Wright and McLeod2015). Well-being suffers from being an ‘idea whose time is come’ (White Reference White2010, p. 159), appealing in its simplicity and common-sense applications. Adding to this complexity, activists have used the concept polemically to counter deficit-oriented policy arguing well-being offers a more inclusive aspiration (Bessant, Reference Bessant2007; McNamara, Reference Marmot and Wilkinson2013). White argues that the concept of well-being is holistic and ‘rejects the compartmentalisation of people’s lives’ (White, Reference White2010, p. 159).

Operationalising a holistic notion of well-being is challenging for researchers (Hamilton & Redmond, Reference Hall, Smith and Toumbourou2010). This has led some ‘well-being’ researchers to focus on the internal psychological world, others to emphasise the interpersonal (family) world and still others to focus on the external (community or environmental) world (De Plater, Reference Davie2008; National Institute for Health and Clinical Excellence, Reference Murphey, Bandy, Schimitz and Moore2009). A study of the relevant literature also reveals the different positions taken as to whether well-being is a product (an outcome of policy) or a process (subjective experience) (Atkinson, Reference Atkinson2013). It is also common for writers to identify ‘risk’ and ‘protective’ factors (Viner et al., Reference Viner, Ozer, Denny, Marmot, Resnick, Fatusi and Currie2012). Particularly in the health literature, this led to a focus on individual behaviours such as drinking, smoking and exercise (Atkinson, Reference Atkinson2013; Currie et al., Reference Cuervo, Worth, Dwyer and Skelton2012). While there is a large literature on individual risk factors that may affect young peoples’ well-being, much less has been published from quantitative studies on community-level factors (or what the social determinants literature refers to as ‘proximal’ factors).

This paper adopts a multi-dimensional understanding of well-being (see Figure 1), exploring material, subjective and relational factors (Bourke & Geldens, Reference Bourke and Geldens2007) with the authors taking the position that well-being is a process, rather than an outcome. Well-being is reflected in young people’s sense of self in the world and is central to ontological security (Hulse & Saugeres, Reference Hamilton and Redmond2008) or identity work (Cote, 2007). There is a range of ways in which well-being can be operationalised in research; however, in this article, it is measured by a modified version of Cowen’s Perceived Self-Efficacy Scale (PSE) used in the Raine Study, a Western Australian Pregnancy Cohort (Straker et al., Reference Steinbeck, Hazell, Cumming, Skinner, Ivers, Booy and Hawke2017a). Perceived self-efficacy is the belief that one can master difficult tasks or cope with certain life stressors (Schwarzer & Warner, Reference Ruming2013). High levels of perceived self-efficacy reflect a sense of control over one’s environment and an optimistic belief in being able to cope with, or change, challenging situations (Schwarzer & Warner, Reference Ruming2013). Previous research has identified self-efficacy as influential in health behaviours such as smoking, physical exercise and sexual health practices (Conner & Norman, Reference Conner and Norman2005), educational outcomes (Bandura, Reference Bandura1997) and persistence (Luszczynska & Schwarzer, Reference Luscombe, Cheng, Balzer, Chow, Paxton, Steinbeck and Hawke2005). These characteristics are similar to those identified in the resilience literature, however, in this article we also seek to understand the external enablers that support self-efficacy among young people (Ungar, Reference Ungar, Theron, Liebenberg and Ungar2015).

There is a strong consensus in the literature that material factors have profound health impacts, described as the social determinants of health (Goodman et al., Reference Gilman, Meyers and Perez2003; Marmot & Wilkinson, Reference Luszczynska, Schwarzer, Conner and Norman2006; Viner et al., Reference Viner, Ozer, Denny, Marmot, Resnick, Fatusi and Currie2012). These social determinants include income, education, employment, ethnicity or race, age and gender. Despite this, there remains a policy, practice and research focus on the internal world of young people. Ungar (Reference Ungar, Theron, Liebenberg and Ungar2015), a leading social work researcher on resilience in young people, argues that much greater attention needs to be paid to the factors that support well-being which are beyond the control of the individual young person. In a similar vein, White (Reference White2010) argues that well-being ‘inheres within the community as a collectivity’ (p. 168). The availability of support services and appropriate opportunities for engagement are important determinants of young people’s well-being (Bernard, Stephanou, & Urbach, Reference Bernard, Stephanou and Urbach2007; Davie, Reference Currie, Zanotti, Morgan, Currie, de Looze, Roberts, Samdal, Smith and Barnekow2015; Wyn et al., Reference Wyn, Cuervo, Landstedt, Wright and McLeod2015). Infrastructure that supports young people’s participation in community life such as the availability of facilities, public transport, quality public amenities and safety are all material factors likely to impact on young people’s well-being (White & Wyn, Reference White and Wyn2013). If a ‘place-based approach’ is scaffolded over Ungar’s ideas, opportunities exist now to refresh the approach used to address disadvantage and promote well-being. Infrastructure that supports young people in the long term could slow the widening rural and urban divide that has seen almost half of all new jobs created in the last decade based in Australian city centres (Department of Infrastructure and Regional Development, 2015). While it is evident that rural communities experience greater disadvantage than urban communities (Vinson, Rawsthorne, Beavis, & Ericson, Reference Vinson, Rawsthorne, Beavis and Ericson2015), less is known about how young people experience this disadvantage or how this might change as they age.

Well-being is at its core subjective. A young person’s well-being will be shaped by their perceptions and experiences of belonging, of safety and connectedness. This sense of belonging is not only an individual experience but also a collective experience, with young people developing an understanding of how much they are valued or welcomed within a community through their peer group (White & Wyn, Reference White and Wyn2013). Subjective well-being is enhanced when one’s values, experiences and sense of self concur with the broader communal meanings or understandings (Mission Australia, Reference Meltzer, Muir and Craig2017; White, Reference White2010). In white Australia, there is widespread acceptance of an idyllic narrative that represents rural as consisting of ‘simple, harmonious, cohesive and homogeneous communities, ordered and free of social conflict’ (Scott, Lyons, & MacPhail, Reference Schwarzer, Warner, Prince-Embury and Saklofske2015, p. 221). This representation creates a discursive meaning of ‘rural’, most often in juxtaposition to the ‘urban’ other (Winterton, Chambers, Farmer, & Munoz, Reference Winterton, Chambers, Farmer and Munoz2014). Within the sociological literature, there are varying interpretations of how rural young people’s well-being is shaped by these discourses. Some researchers suggest young people are ‘forced’ to leave rural communities to increase their life opportunities (Davie, Reference Currie, Zanotti, Morgan, Currie, de Looze, Roberts, Samdal, Smith and Barnekow2015, p. 17) while others highlight the oppressive nature of the rural idyllic discourse for young people (Rawsthorne, 2003). Little longitudinal research on the experiences of rural young people in Australia exists, thus limiting our understanding of how young people’s sense of belonging changes over time, if at all.

Well-being is concerned with social relations and is seen as the product of the interaction of an individual with their broader social context (VicHealth, 2015). Relational factors are essential to understand well-being, as ‘people become who and what they are in and through their relatedness to others’ (White, Reference White2010, p. 164). Traditionally, research has focused on young people’s relationships with peers, school and family to understand well-being. There is some research (Mission Australia, Reference Meltzer, Muir and Craig2017; Wyn et al., Reference Wyn, Cuervo, Landstedt, Wright and McLeod2015) that suggests broader community supports, beyond family and school, are also likely to be influential. Previous research has found that participation in non-school activities and voluntary organisations has positive impacts on young people’s well-being, particularly when supervised by supportive adults (Erbstein, Hartzog, & Geraghty, Reference De Plater2013). Gilman, Meyers, and Perez (Reference Fabiansson2004) found young people’s involvement in what they call extracurricular activities enhances life satisfaction. Recent Australian qualitative research with young people aged 12–20 years has also highlighted the importance of ‘trusted adults’ in young people’s transition from childhood to adulthood (Meltzer, Muir, & Craig, Reference McNamara2016, Reference Meltzer, Muir and Craig2018). This research concluded that these relationships were most valuable when chosen by the young person and marked by low-key, direct and equitable conversations (Meltzer et al., Reference Meltzer, Muir and Craig2018, pp. 582–583). Within the Australian policy context, young people’s participation in physical activities, particularly sport, is strongly advocated as contributing to health and well-being (Mission Australia, Reference Meltzer, Muir and Craig2017). Research undertaken with Norwegian young people identified a strong relationship between participation in competitive sports and mental health (Breistøl Clench-Aas, Van Roy, & Kjærsti Raanaas, Reference Breistøl, Clench-Aas, Van Roy and Kjærsti Raanaas2017).

With approximately one-third of 10–13-year-olds living in regional or remote areas (32.8%; ABS, 2011), the well-being of Australian rural and regional adolescents still remains poorly documented and understood. Additionally, the use of a single ‘rural’ category obscures differences within young people living outside major urban areas (Davie, Reference Currie, Zanotti, Morgan, Currie, de Looze, Roberts, Samdal, Smith and Barnekow2015).

This article seeks to contribute to greater understanding of the impact of material, subjective and relational factors on rural and regional young people’s well-being. It aims to inform youth policy, which currently assumes a normative developmental process which is often presented as context free (Fabiansson, Reference Erbstein, Hartzog and Geraghty2015). Accordingly, this article seeks to increase our understanding of what factors impact on rural young people’s well-being. This may have important future implications in terms of support strategies, community action and urban migration.

Research design and data collection

The data reported in this article were collected as part of the Adolescent Rural Cohort study of Hormones, Health, Education, Environment and Relationships (ARCHER) study which explores the relationship between universal measures of adolescent behaviours, mental and physical health and hormone changes through puberty (Steinbeck et al., Reference Scott, Lyons and MacPhail2012). The ARCHER study collected both physiological and social data; however, this article focuses on specific social data only. The study was conducted in regional and rural Central Western New South Wales (NSW), Australia. Adolescents in school years 5–7 were recruited primarily through schools and flyers placed in community organisations and print media. Both adolescents and parents/guardians were asked to complete annual ARCHER surveys. The local community, including young people, has been involved since the conception of the study through ARCHER Community Consultative Committees and focus groups (Robbins et al., Reference Robbins, Rawsthorne, Paxton, Hawke, Skinner and Steinbeck2012). A young person also contributed as an author to the writing of this manuscript.

Participant profile

Baseline participant demographic profiles are provided in Table 1. At baseline, the mean parent/carer age was 42.7 years and median adolescent age was 11 years, with slightly fewer female adolescent participants (45%) than males. Comparisons with local government area data indicated that the adolescent participants represented the region in terms of gender distribution but had slightly fewer youth of Aboriginal background (11% vs 17% in the region). While the young participants came from diverse socioeconomic backgrounds, comparisons with households in the region indicated that participating families had higher rates of completing higher education and were less likely to be in rental housing. For two-thirds of the households, the highest household occupation was managerial (64%) and for 17% was sales, machinery operation, labouring or home duties (Luscombe et al., Reference Luscombe, Cheng, Balzer, Chow, Paxton, Steinbeck and Hawke2017). Caution should be exercised in generalising the findings from this study due to the nature of the sample.

Table 1.

Relationship between baseline demographics and PSE at baseline and Wave 2

* Indicates a statistically significant association with level of PSE at p < 0.05.

^a PSE data were not collected for 15–16-year-olds in the follow-up, who were 13–14 years at baseline.

^b Cultural background of the young person was reported by parent/guardian.

Key concepts and data analysis

This article reports findings from a number of specific variables within the larger ARCHER study. These variables are reported in two waves, at Baseline when participants first joined the study and at Wave 2, some 2 years later. Hence, the paper is not reporting longitudinal change over time but cross-sectional data at two time points.

Limitations

Caution needs to be exercised in interpreting the data presented in this article. While we present data at two points in time in this article, we have not explored change over time in specific individuals. Due to the survey design, at the time of writing, longitudinal data were only available for a subset of the sample (those aged 9–12 at baseline). And, as noted above, PSE was not collected for 15–16-year-olds in the follow-up, who were 13–14 years at baseline. Accordingly, the data are analysed as two cross-sectional data sets. This enables us to explore whether associations identified in the data increase, remain or decrease over time. Tests of association (between say PSE and parental occupation) were undertaken within each data set, not between the two data sets. Due to changes in survey design, recruitment and attrition, the number of participants who provided data on PSE declined from 342 at baseline to 217 approximately 2 years later. Full longitudinal analysis on the wide range of health issues studied by the ARCHER team will be undertaken in the future once all data have been collected, cleaned and the impact of attrition identified.

Further, the size and specific nature of the sample also limit the generalisability of the findings. Some data analyses (for example by location) are based on small cell sizes.

Additionally, the findings in this article are solely based on quantitative survey data. It is likely that qualitative research would provide additional, contextualised insights.

Findings

We analysed data on young people’s Perceived Self-Efficacy in relation to each of the dimensions of well-being. Most young people (81%, n = 275/338) had high PSE at baseline and at Wave 2 (87%, 188/217). In this section, we focus on those associations that appeared to be most influential for PSE.

Material factors

There is a large body of research on the relationship between SES and health (Marmot & Wilkinson, Reference Luszczynska, Schwarzer, Conner and Norman2006). Among the young people in our study, the impact of SES was as anticipated, with those reporting lower socioeconomic factors also reporting lower levels of PSE. A statistically significant relationship was evident in data on Highest Household Occupation, in which adolescents with caregivers employed in professional or managerial roles reported higher PSE, particularly at Wave 2. Interestingly, other material factors, such as age, gender and cultural background, had less impact. This finding contrasts with previous research and requires further exploration. Differences in PSE between Aboriginal and non-Aboriginal young people decreased over time. This would suggest that efforts to build PSE among young people need to be tailored to ensure the inclusion of those with less financial resources.

Previous research suggests the existence and quality of these environmental factors/facilities will affect young people’s perceived self-efficacy and their capacity to participate as active citizens (Mission Australia, Reference Meltzer, Muir and Craig2017; White & Wyn, Reference White and Wyn2013). The survey asked participants to rate a range of environmental factors/facilities. These questions aimed to identify the material factors beyond the control of individual young people that are likely to impact on young people’s well-being (VicHealth, 2015).

Table 2 shows tests of association between material (amenities) factors and PSE over time and reveals a statistically significant association with responses to questions about ‘places for young people to meet’ and ‘overall tidiness of the area’. This suggests an association between perceived disorder and lower levels of PSE. Possibly reflecting the scarcity of public space for mid-teens in these communities, we also found a statistically significant association between PSE and places for young people to meet at Wave 2. Generally, there was a decline in young people’s positive rating of each of these environmental factors/facilities over time. This was reflected in open-text responses such as

Table 2.

Relationship between young person’s baseline ratings of environmental factors and facilities and PSE at baseline and Wave 2

* Indicates a statistically significant association with level of PSE at p < 0.05.

Over time, young people’s rating of all public amenities declined, with the availability of public transport consistently poor. Young people living outside of the regional cities (Orange and Dubbo) were at least twice as likely to rate public transport as ‘poor’. Drilling further into the data, we found differences by location, with those living in villages or on farms having the worst access to public amenities. For example, young people living in the regional cities were more positive about places to meet than those living in towns (93.1% compared to 84.6%). These differences point to the spatial impact of material factors on young people’s well-being.

These findings support other research with young people. Data collected as part of the Communities That Care research in regional Victoria reported just over half of their adolescent participants (50.6%; made up of 44% Year 6; 56% Year 8 and 52% Year 10) felt there were sufficient prosocial opportunities within the community Hall et al. (Reference Rosier and McDonald2015). It is possible that poorer neighbourhoods are marked by greater disorder or, conversely, in neighbourhoods subject to greater levels of surveillance, the surveillance itself contributes to fear of crime. Previous research has established that environmental and social disorder (captured in the survey through ‘overall tidiness of the area’) can have adverse mental health impacts (Vinson & Rawsthorne, Reference Vinson and Rawsthorne2013). Research by Ross and Mirowsky (Reference Rosier and McDonald2001) concluded that ‘people who report that there is a lot of crime, graffiti, vandalism, trouble, drug use, dirt and danger in their neighbourhood have more chronic health problems, worse self-reported health and worse physical functioning than people in neighbourhoods typified by order and safety’ (p. 266). To some extent, these findings run counter to the idyllic discourse of rural communities as inherently safer and ordered than urban environments (Valentine, Reference Valentine1997).

Subjective factors

Research findings suggest that a sense of belonging is likely to impact on young people’s well-being (Fabiansson, Reference Erbstein, Hartzog and Geraghty2015; Mission Australia, Reference Meltzer, Muir and Craig2017; Scott et al., Reference Schwarzer, Warner, Prince-Embury and Saklofske2015; White, Reference White2010). Our analysis found strong statistical associations between a sense of community belonging and PSE at Wave 2 (see Table 3). Those young people with a stronger sense of community belonging reported higher levels of PSE. The reverse also held – that is, that those young people reporting low sense of community belonging reported lower levels of self-efficacy.

Table 3.

Relationship between young person’s baseline ratings of community belonging and PSE at baseline and Wave 2

N.B. ‘agree’ combines agree and strongly agree responses; ‘disagree/neutral’ combines neither agree nor disagree, disagree and strongly disagree responses.

* Indicates a statistically significant association with level of PSE at p < 0.05.

At baseline, the relationship was not as marked, apart from the statement ‘I like this area’ which was statistically significant. Our analysis found that perceptions of safety were important as was feeling ‘part’ of the area in relation to young people’s PSE. For those young people who felt a strong sense of belonging, the community was associated with safety and nurture as demonstrated in this young person’s quote:

Recent research undertaken by Mission Australia also identified safety concerns among young people, particularly those from lower socioeconomic background. Forty-one and a half percent (41.5%) of low SES young people in the 2016 Youth Survey identified their area as unsafe, compared to 27.7% of high SES young people (Mission Australia, Reference Meltzer, Muir and Craig2017, pp. 19–20).

Relational

Young people’s participation in activities beyond school and family environments has significant implications for their social and emotional well-being. There are obvious benefits for physical and mental health but also benefits for social relations (Mission Australia, Reference Meltzer, Muir and Craig2017). Social benefits of non-school activities include improved communication and social skills, building resilience and coping mechanisms, establishing new social networks and gaining early leadership experiences (Bailey, Reference Bailey2005). Relational factors at both the personal and community level are seen to contribute to young people’s overall well-being (Fabiansson, Reference Erbstein, Hartzog and Geraghty2015; Mission Australia, Reference Meltzer, Muir and Craig2017).

Participation, particularly in sports, has strong policy and public support within Australia. The ABS General Social Survey found two-thirds of young people (aged 15–17) had participated in a social group and 9 out of 10 young people had participated in sport or recreational physical activity in the past 12 months (ABS, 2014). Internationally, well-being was found to be associated with competitive sports (Breistøl et al., Reference Breistøl, Clench-Aas, Van Roy and Kjærsti Raanaas2017). We found no statistically significant association between the amount of sports participation and PSE. Those young people reporting less than average participation in sports were found to have similar PSE as those more involved. Likewise, competency (how well) in sport appears to have minimal impact (either positive or negative). This general trend also holds for non-sports-related activities such as computer gaming, music or arts-based hobbies. This is an important finding for those wishing to build young people’s well-being (parents, policy makers and health professionals) as it casts some doubt on sports participation as a key strategy. Interestingly, our analysis suggests that active participation in organisations, clubs, teams or groups did make a difference with those young people involved in more organisations, clubs, teams or groups reporting higher PSE. This suggests that encouraging young people to become active in decision-making and organisational matters rather than simply participating in a pre (adult) organised activity may build Self-Efficacy.

In the following section, data in relation to ‘significant other adults’ were reported, recognising that ‘positive adolescent adaptation and development depends in part on the presence of supportive … caring adults’ (Erbstein et al., Reference De Plater2013). Having one or more caring adults in a child’s life increased their potential to ‘flourish’ and complete the transition to independent adult successfully (Meltzer et al., Reference McNamara2016; Murphey, Bandy, Schimitz, & Moore, 2013). These caring adults may include other relatives, neighbours, friends of parents, teachers, coaches and religious leaders (Murphey et al., 2013).

A test of association between responses to questions from the California Health Kids Survey (caring other adult) and PSE suggests identified several significant associations (see Table 4). A statistically significant association was found at both baseline and Wave 2 in relation to all these statement except ‘who always wants me to do my best’. In general, high PSE was associated with perceived support from other adults.

Table 4.

Association between young person’s baseline ratings of relationships with significant other adults and PSE at baseline and Wave 2

* Indicates a statistically significant association with level of PSE at p < 0.05.

Table 5.

Association between young person’s baseline ratings of relationship with peers and PSE at baseline and Wave 2

* Statistically significant p < 0.05.

This analysis suggests a sharp increase in the importance of having someone ‘who tells me when I do a good job’ as the participants move through adolescence. This contrasts with what could be interpreted as external expectations (always wants me to do my best) which appear to have little relationship with PSE. This concurs with findings from other Australian research on the importance of trusted adults talking not telling (Meltzer et al., Reference McNamara2016, pp. 60–62). The associations found between what we have called significant other adults and PSE are more marked than other relational factors such as participation in non-school activities. However, it is probable that contact with these significant other adults occurs through participation in organised activities like music and sport.

Analysis between the baseline and 2-year follow-up suggests very little change over time in terms of the respondent’s perceptions of significant other adults in their lives. In general, shifts tended to be positive, with the ‘very much true’ percentage increasing approximately 5% across all statements except ‘whom I trust’. Data on perceptions of ‘trust’ ran somewhat counter, with an increase in respondents believing it was ‘not at all true’ that they had someone whom they trusted over time (1.7%–3.8%). While this increase is concerning, other Australian research found that 39.8% of young people ‘rarely’ trusted people (Fabiansson, Reference Erbstein, Hartzog and Geraghty2015, p. 98).

Peers play an increasingly important role in young people’s well-being as they move through adolescence. Our analysis found no statistical association between how many friends participants had and PSE (see Table 5). There was, however, an association at Wave 2 between the amount of contact with peers and PSE. Those participants with only occasional contact with peers (less than once per week) were more likely to report lower PSE compared to those having three or more contacts per week with peers.