Levels of social connectedness among adult Australian males

Sonia Terhaag, Jennifer Prattley, Neha Swami, Bosco Rowland and Brendan Quinn

Overview

Social connectedness is comprised of the relationships and interactions we have with others, as well as how we subjectively perceive our levels of support. In general, Australians are well-connected and have access to supportive social networks (AIHW, 2019b; Relationships Australia, 2018); however, some groups are more likely to be socially isolated or poorly supported. These groups include men, single fathers, older Australians, and those who live alone (Arbes, Coulton, & Boekel, 2014).

This disparity in social connectedness may be temporary; for example, in response to a particular event such as a relationship breakdown, but such experiences and feelings can persist over long periods of time (Moloney, Weston, Qu, & Hayes, 2012). This can be problematic for some people - loneliness, social isolation and low social support may substantially and adversely affect physical and emotional wellbeing, as well as economic participation and functioning (Berkman et al., 2000). Because of this, it is important to consider how different demographic and social characteristics and behaviours are associated with feelings of support.

Research on social connectedness among males is limited. Little is known about their social networks and how they perceive and mobilise support. International studies have indicated that males might be more susceptible to feelings of low social support - and to being less socially connected in general - than females (e.g. Cornwell, B., Laumann, & Schumm, 2008; McKenzie et al., 2018; Shields & Wheatley Price, 2005).

This chapter aims to address these gaps in the research. It describes the social connectedness of Australian men across different indicators of attachment relationships and community involvement. Men's overall subjective sense of social support and satisfaction with their relationships is also examined.

Research objectives

This chapter uses data from Waves 1 and 2 of Ten to Men (TTM) to describe:

• the types of attachment relationships between Australian men and their families and friends
• levels of community integration among Australian men, as per employment status and participation in certain community-based activities
• the level of social support men receive from their social connections overall, including changes in perceived social support over time and satisfaction with relationships.
• socio-demographic and psychosocial factors associated with men's levels of perceived social support and satisfaction with personal relationships.

Attachment relationships

Connections with family

Table 1.1 presents different types of family connections among Australian men aged 18-60 in 2015/16. The majority (64%) were either married or in a de facto relationship. When broken down by age, the greatest proportion (80%) who were married/de facto were those aged between 45-60; in contrast, only 8% of 18-24 year old men were married or in a de facto relationship. Close to one-third (30%) of all men had never been married.

Around 60% of men had at least one child (mean = 1.5). Most fathers lived with their children (83%), whereas 9% had children who sometimes lived elsewhere, and 8% had children who always lived elsewhere.

Household composition

Table 1.1 also details the household composition of adult Australian males in 2015/16 in consideration of their relationship status. Over 90% lived with at least one other person. Around one-quarter (24%) were single but lived with others (e.g. friends, family, housemates). Nearly two-thirds (63%) were living either as a coupled household or as a couple with children. Five per cent resided in single-parent households.

Notes: N > 10,391; CI = Confidence Interval; SD = Standard deviation.

Source: TTM data, Wave 2, adult cohort, weighted

Figure 1.1 presents the variation in household composition among Australian men of different ages in 2015/16. The majority of young men (<24) were single but living with others, while those over the age of 25 were mostly living with a partner and child/ren. Only 4% of those aged over 34 were single but living with others. The proportion of men living alone was significantly greater among older males; approximately 3% of Australian men aged 18-24 lived alone, compared to 5% of those aged 25-34, 6% of those aged 35-44, and 10% of those aged 45 and above. The majority of men aged 35-44 and 45-60 lived with partners and children (73% for each).

Figure 1.1: Distribution of household composition by age group among adult males in 2015/16

Read text description.

Notes: N = 10,391

Source: TTM data, Wave 2, adult cohort, weighted

Connections with friends and relatives

Most adult males have connections with relatives and friends; however, the number of close friends and/or relatives in men's lives ranges substantially (0-600). In 2015/16, around half of Australian men (51%) had between five and 15 close friends or relatives, whereas 4% had none.

When broken down by age, similar proportions of men in each group had no close friends or relatives. Approximately, 5% of men aged 45-60 had no close friends or relatives; 4% of those aged 35-44, 4% of men 25-34, and 3% of 18-24 year olds.

Community integration

Employment status

The majority of Australian men aged 18-60 (85%) were in some form of employment in 2015/16. Approximately 60% of this group were employed on a permanent basis; in comparison, a minority were employed on a casual basis (10%), were self-employed (11%), or had fixed-term contracts (4%).

Employment was highest among adult males aged 25-34 and 35-44 (91% and 89%, respectively), and lowest among men aged 45-60 (86%) and 18-24 (71%). The highest rate of unemployment was among men aged 18-24 (29%). Only 12% of all men aged over 25 were unemployed.

Involvement in community-based activities

The majority of adult Australian males attend events that bring people together, such as fetes, festivals or other community events. In 2013/14,¹ approximately 36% of men attended these events at least sometimes. Table 1.2 outlines the proportions of men participating in certain community-based activities at this time. Less than one-quarter (24%) took part in any sort of ongoing community service activity, such as volunteering, whereas approximately one-third (37%) were active members of a sports, hobby or community-based club or association. Around 15% attended religious ceremonies on at least a monthly basis.

Notes: ^a Includes volunteering at a school, coaching a sports team, working with a church or neighbourhood association; ^b Includes going to church, temple, mosque or other religious institutions or activities; ^c Such as fetes, festivals and 'other' community events.

Source: TTM data, Wave 1, adult cohort, weighted

Chapter 5 provides more detail about participation in community-based activities among Australian males in 2013/14, including information on age differences and factors associated with community-based activity involvement among this group.

Self-perceived social support and satisfaction with personal relationships

Self-perceived social support

Self-perceived social support is one of the main variables of interest in analyses presented throughout this report. It was assessed in Waves 1 and 2 using the Emotional/Informational Support subscale from the Medical Outcomes Study (MOS) Social Support Survey (Sherbourne & Stewart, 1991). Total scores range 0-100; higher scores indicate greater levels of social support. Table 1.3 shows the distribution of self-perceived social support scores among all adult Australian males - broken down by age group - in 2015/16, as measured by the Medical Outcomes Study (MOS) Social Support Scale. The overall mean self-perceived support score across all ages was 69.5 (standard deviation [SD] = 26.9; range: 0-100). The lowest quartile of respondents had a self-perceived social support score of less than 32.

Men aged 18-34 experienced significantly higher levels of self-perceived social support (average score range: 72-76) compared to men aged 35-60 years (average score range: 66-68). Accordingly, a significantly greater proportion of men aged 18-24 and 25-34 years (29% and 25%, respectively) rated their support at the maximum value of 100, compared to 19% of those aged 35 and over.

Notes: CI = Confidence Interval; SE = Standard error.

Source: TTM data, Wave 2, adult cohort, weighted

Availability of social support among adult males in 2015/16

The availability of eight different types of emotional and informational support experienced by Australian men is shown in Figure 1.2. The majority (>60%) had each type of support available most or all of the time. The type of support most commonly available to men most of to all the time was having someone they could count on to listen when needing to talk (72%). The least available support was having someone to share private worries and fears with; this was available to around 61% of men most of to all the time.

Figure 1.2: Prevalence of availability of social support across individual MOS Social Support Scale items among adult males, 2015/16

Read text description.

Notes: N = 10,384. Each MOS Social Support Scale item is rated on a five-point scale (1-5; refer: Appendix A: Methodology). Here the highest ('most', 'all') and lowest ('none', a little') response items were collapsed to create three categories: most/all of the time; some of the time; a little/none of the time.

Source: TTM data, Wave 2, adult cohort, weighted

Change in the relative level of self-perceived social support over time

Figure 1.3 illustrates that the majority of Australian men did not change from their relative level of self-perceived social support between 2013/14 and 2015/16. That is, 55% of those who were in the highest quartile of self-perceived social support in 2013/14 were also in the highest quartile in 2015/16, while 57% of those who were in the lowest quartile in 2013/14 were also in the lowest quartile in 2015/16. Furthermore, most (58%) of those in the middle 50% in 2013/14 remained in this category in 2015/16. In Figure 1.3, the blue bars represent males who comprised the highest quartile of self-perceived social support in 2013/14, the orange bars are those in the middle 50% of self-perceived social support, and the yellow bars are those in the lowest quartile of self-perceived social support at this time point.

Importantly, a minority of males did experience a significant change in their relative position. Nine per cent were in the highest quartile for self-perceived social support in 2013/14 but the lowest quartile two years later. Ten per cent were in the lowest quartile in 2013/14 but the highest in 2015/16.

Figure 1.3: Change in self-perceived social support categories over time, 2013/14 to 2015/16

Read text description.

Notes: N = 9,744. Blue bars represent the highest quartile of self-perceived social support in 2013/14, the orange bars are those in the middle 50% of self-perceived social support, and the yellow bars are those in the lowest quartile.

Source: TTM data, Waves 1 and 2, adult cohort, unweighted

Satisfaction with personal relationships

Figure 1.4 illustrates the distribution of satisfaction with personal relationships² among adult Australian males in 2015/16 - overall and by age group - as measured using the Personal Wellbeing Index on a scale from 'completely dissatisfied' to 'completely satisfied' (International Wellbeing Group, 2013). The distribution of scores was negatively skewed; in general, satisfaction with personal relationships was high among men (mean = 7.15, SD = 2.47). A small proportion (1%-2%) were 'completely dissatisfied' with their personal relationships. In comparison, 15%-20% of men were 'completely satisfied' with their personal relationships.

The average level of satisfaction ranged between 7.0 and 7.2 and did not differ significantly between the four age groups.

Figure 1.4: Satisfaction with personal relationships and among adult men by age, 2015/16

Read text description.

Notes: N = 10,577. Bars represent different levels of satisfaction with personal relationships among Australian men (%) on an 11-point scale ranging from 0 (completely dissatisfied) to 10 (completely satisfied).

Source: TTM data, Wave 2, adult cohort, weighted

Multivariable analyses: Factors associated with low social support and relationship satisfaction

Table 1.4 details the findings of two multivariable linear regression analyses that explored associations between socio-demographic factors and the number of past-year social life events with:

1. perceived social support
2. satisfaction with relationships.

There were several consistent findings regarding both outcomes, namely that indicators of social isolation were typically associated with lower levels of social support and satisfaction with personal relationships. In particular, having no close friends or family, being single, and being unemployed or out of the labour force were all associated with lower self-perceived social support and satisfaction with relationships.

Having no close friends or relatives had the strongest negative relationship with perceived support - an average of 45 points lower on self-perceived support on the MOS Social Support Scale (range: 0-100). Having no close friends or relatives was also associated with an average of 1.8 fewer points on relationship satisfaction on the PWI subscale (range: 0-10). Living in remote or very remote locations was associated with 15 fewer points on self-perceived support compared to living in major cities.

Compared to being employed, being unemployed and looking for work and being out of the labour force were independently associated with lower levels of both self-perceived social support and satisfaction with relationships.

Older age and identifying as non-heterosexual were also negatively correlated with self-perceived social support and satisfaction with relationships.

In contrast, men with lower conformity to masculinity norms experienced higher self-perceived social support and satisfaction with relationships. Having a partner - with or without children - was significantly associated with higher relationship satisfaction (around three percentage points higher).

Except for single father households, all other household types had significantly higher self-perceived social support compared to those who were single and living alone. Single fathers had similar levels of self-perceived social support to men who lived alone.

Experiencing potentially stressful life events (e.g. job loss or moving house; see Box 2.1 in Chapter 2) was associated with incrementally lower self-perceived support and satisfaction with relationships. In particular, having experienced two or more life events in the past year was associated with an average five-point lower self-perceived support score, and half a point lower relationship satisfaction score. There was no association found between experiencing one event and levels of self-perceived social support. As the nature of life events can vary and be experienced at different times and in different combinations, further investigation of the associations between certain life events and social connectedness is covered in the next chapter.

Notes: CALD = Culturally and Linguistically Diverse; SEIFA = Socio-Economic Indexes for Areas. ***p < 0.01, **p < 0.05, *p < 0.1; ^# Only available and measured at Wave 1. For a one-unit change in the explanatory variable (age, Indigenous status, etc.), one would expect a β unit change in the outcome variable (MOS Social Support Scale or PWI score), assuming that all other variables in the model are held constant. MOS Social Support Scale scores for the TTM sample ranged from 0-100; mean = 68.8; standard deviation = 26.9. PWI scores for the sample ranged from 1-99; mean = 69.8; standard deviation = 17.3.

Source: TTM data, Waves 1 and 2, adult cohort, unweighted

Conclusion

Limited research has investigated the social connectedness of Australian men. TTM findings presented in this chapter provide an indication of attachment relationships, community integration and levels of self-perceived social support among adult Australian males in 2015/16. Unadjusted analyses in the first part of the chapter indicated that most men aged 18-60 appeared well-connected with, and felt supported by, their families, friends and communities. For example, around two-thirds were married or in a de facto relationship and over 90% lived with at least one other person. Approximately half had between five and 15 close friends or relatives. Eighty-five per cent were employed. Around three-quarters experienced at least moderate levels of self-perceived social support, and satisfaction with their personal relationships was generally high.

Importantly, a significant minority of adult males experienced limited social connectedness. For example, approximately 4% had no close friends or relatives, a proportion that was fairly consistent across age groups. One-quarter of men experienced low levels of self-perceived social support, and 1%-2% were completely dissatisfied with their personal relationships.

The findings of multivariable analyses showed that certain characteristics indicating limited social connectedness - such as unemployment, being single and living alone - were significantly associated with lower levels of self-perceived social support and relationship satisfaction after adjusting for other factors. These findings support those of previous research showing that attachment relationships and indicators of community integration - such as being employed - can be crucial for developing and maintaining social networks and feelings of support (Milner, Krnjacki, Butterworth, & LaMontagne, 2016). In turn, this can be beneficial for mental and physical health and wellbeing (Aslund, Starrin, & Nilsson, 2014).

In multivariable analyses, older age was independently associated with lower self-perceived social support and relationship satisfaction. This accords with other findings, presented in this chapter and elsewhere in the report, that show a higher prevalence of living alone among men aged 45 and above, and lower rates of certain types of community integration (e.g. membership of sport or hobby clubs/associations; see Chapter 5) in older vs younger males.

In line with other research highlighting older males as a group vulnerable to lower levels of social connectedness (e.g. Cornwell, & Waite, 2009; Sixsmith & Boneham, 2003), TTM findings indicate that ageing Australian males could possibly benefit from targeted interventions to promote community engagement and social interactions and associated health and wellbeing outcomes (see Policy and practice implications).

Lastly, greater conformity to masculine norms was also independently associated with lower levels of self-perceived social support and satisfaction with relationships in multivariable analyses. Previous research has highlighted the numerous adverse outcomes associated with aligning with norms of stoicism, self-reliance and avoidance of negative emotions, including poor mental health, harmful substance use and a reluctance to seek help when needed (Herreen, Rice, Currier, Schlichthorst, & Zajac, 2021; Struik et al., 2019).

Lastly, greater conformity to masculine norms was also independently associated with lower levels of self-perceived social support and satisfaction with relationships in multivariable analyses. These findings are consistent with international studies indicating limited social connectedness - including low rates of community involvement and attachment relationships and negative social functioning - among males with rigid adherence to masculine norms (McKenzie et al., 2018; Wong, Ho, Wang, & Miller, 2017). For example, research by Sixsmith and Boneham (2003) showed that older men who exhibited higher distrust and perceived dominance over women were also more excluded from social spaces that could grow their social connectedness and capital. Further, an evaluation of men's experiences in male support groups in Western Australia showed that a greater amenability to identity change, and more flexible masculinity, enhanced men's sense of emotional support and community (Reddin & Sonn, 2003). Recommendations to address low social connectedness in the context of greater conformity to masculine norms - and in consideration of previous research - are included in the Policy and Practice implications section of this report.

Is there a bidirectional relationship between depression and self-perceived social support among Australian men?

Neha Swami, Jennifer Prattley, Sonia Terhaag, Bosco Rowland and Brendan Quinn

AIFS recognises that each of the numbers reported here represents an individual. AIFS acknowledges the devastating effects mental ill-health and limited social support can have on people, their families, friends and communities.

This chapter discusses depression and presents material that some people may find distressing. If you or someone you know is feeling depressed or suicidal, please contact one of the following services:

Overview

Mental health is a key determinant of quality of life. Mental health disorders such as depression affect a significant portion of the global population and can result in a high burden on health and functioning (Rehm & Shield, 2019; World Health Organization [WHO], 2017). It is estimated that approximately four people in every 100 of the global population suffer from a depressive disorder at any point in time and, globally, these disorders are ranked as the single largest contributor to non-fatal health loss (7.5% of all Years Lived with Disability [YLD]) (WHO, 2017).

Depression is a highly recurrent disorder. At least 50% of people who recover from an initial episode have one or more further depressive episodes in their lifetime, and the risk of reoccurrence is close to 80% for a person with a history of two episodes (American Psychiatric Association, 2000; Burcusa & Iacono, 2007; Machmutow, Grosse Holtforth, Krieger, & Watzke, 2018).

Social support has been identified as important for maintaining good mental health. It can improve resilience to stress and ease key life transitions and personal crises (Ozbay et al., 2007). However, a corollary of experiencing mental health disorders may be difficulty accessing social support and/or appropriate professional support for mental ill-health.

Existing research has investigated the relationship between self-perceived social support and a variety of mental health outcomes. However, most studies in this area have focused on understanding the one-way relationship of how social support affects mental health outcomes in general and not looked at depression specifically (Wang, Mann, Lloyd-Evans, Ma, & Johnson, 2018).

There is some international evidence indicating that the effect of social support on mental health is not necessarily proportionate to the effect of mental health on social support (Fiori & Denckla, 2012; Hakulinen et al., 2016; Ren, Qin, Zhang, & Zhang, 2018; Santini et al., 2020). For example, a prospective cohort study using a representative sample of New Zealanders investigated the bi-directional, longitudinal relationship between social connectedness and mental health (operationalised as distress). It found that perceived social connectedness was a stronger and more consistent predictor of mental health year-on-year than the alternative association of mental health on social connectedness (Saeri, Cruwys, Barlow, Stronge, & Sibley, 2018).

In Australia, there is limited evidence on the nature and direction of the relationship between indicators of social connectedness and mental health, including depression. Available studies have focused on examining the prevalence of loneliness and lack of self-perceived social support and correlations with mental health outcomes (Australian Psychological Society [APS], 2018; Relationships Australia, 2018). For example, the APS found that, in 2018, one in four Australians felt lonely, lonely Australians were more likely to be depressed, and experiencing depression increased the likelihood of being lonely (AIHW, 2019b; APS, 2018).

Using data from a representative sample of Australian adults from the Household, Income and Labour Dynamics in Australia Survey (HILDA), Ding, Berry and O'Brien (2015) found that higher levels of community participation (a proxy for social capital) predicted better mental wellbeing, and better mental wellbeing predicted greater levels of community participation.

Further research points to a bi-directional association between depression and social support; that is, the experience of depressive symptoms may cause reduced social support (Clement et al., 2015; Heinrich & Gullone, 2006; McKenzie et al., 2018), and reduced social support may cause depression (Santini et al., 2020; Wang et al., 2018). However, evidence on this relationship among men is particularly limited, and the relative magnitude and importance of each association has not been estimated using a large, nationally representative sample.

Given the scarcity of research into the reciprocal association between social support and depression (Wang et al., 2018), particularly among men, in this chapter, this bi-directional relationship is explored. The impact of the relationship is quantified in both directions for Australian men, which is important for developing prevention and intervention initiatives..

Research objectives

This chapter used data from the first two waves of Ten to Men (TTM) to:

• describe the prevalence of depressive symptoms and social support among Australian men
• examine how the experience of depressive symptoms varies among men with different levels of social support, and the converse relationship; that is, how social support levels vary according to depressive symptoms.
• test the bi-directional relationship between social support and experiencing depressive symptoms with two waves of longitudinal data using a cross-lagged panel model that accounts for: i) initial depressive symptoms and social support; ii) multiple potentially confounding observed factors; and, iii) unobserved factors impacting both depression and social support.

Depression and self-perceived social support among Australian men in 2015/16

Prevalence of depression

Depression was assessed among TTM participants using the Patient Health Questionnaire (PHQ-9; Kroenke, Spitzer, & Williams, 2001) (refer: Appendix A: Methodology). The PHQ-9 score range is 0-27; the average in 2015/16 for Australian men aged 18-60 was 4.3.

Table 3.1 shows the mean PHQ-9 scores across men of different age group and for the whole TTM sample; similar levels of depression, regardless of age, were observed in the sample. An estimated 13% of men were classified as experiencing moderately severe to severe depressive symptoms, indicated by scores of 15 or higher. Further details of the distribution of the sample for the PHQ-9 are included in Appendix B, (Table B3.1 and Figure B3.1).

Notes: SE = Standard error; CI = Confidence Interval.

Source: TTM data, Wave 2, adult cohort, weighted

Prevalence of self-perceived social support

As detailed in Chapter 1 and Appendix A: Methodology, the MOS Social Support Scale score range is 0-100, with higher scores indicating greater levels of self-perceived social support. A score of 100 indicates a respondent reported having social support 'all of the time' for each of the eight scale items; a score of 0 would suggest the respondent reported having support 'none of the time' for every item. Scores were calculated for those with valid responses to all eight items; those with missing data were removed from the analysis. Respondents were categorised into one of three groups according to their score. At Wave 2, 10% of respondents scored on the lowest quarter of the scale (0-25). Fifty-one per cent scored between 26-75, the middle 50% of the scale. Thirty-nine per cent had scores between 76-100, the highest quarter of the scale.

The mean MOS Social Support Scale score among men aged 18-60 years in 2015/16 was 69.5, indicating that adult Australian males generally reported moderate levels of self-perceived social support, and fall within the middle 50% of the scale. The mean values and 95% CIs of MOS Social Support Scale scores for all age groups, shown in Table 1.3 in Chapter 1, indicate that younger men aged 18-34 experienced higher levels of self-perceived social support (average score range: 72-76) compared to men aged 35-60 years (average score range: 66-68).

In terms of specific types of support, 'having someone to count on to listen when needing to talk' was the most common form of support; 72% of men said this was available to them most to all of the time. Relatively, the least available form of support was 'having someone to share private worries and fears with'; 61% of men said this was available to them most to all of the time. The distribution of MOS Social Support Scale scores among the TTM sample at Wave 2, overall and by age group, is depicted in Appendix B, Figure B1.1.

Correlation between depression and self-perceived social support
(Wave 2)

Overall, there was a negative relationship between depressive symptoms and levels of self-perceived social support among adult TTM participants. Greater levels of depressive symptoms were associated with lower levels of self-perceived social support. Conversely, when the level of social support increased, depression tended to decrease in severity.

Kendall's Tau-b correlation coefficient values at Wave 2 are presented in Table 3.2 Kendall's Tau-b is a measure of the strength and direction of any association between two variables (Arndt, Turvey, & Andreasen, 1999); here, it is the two separate scores for depressive symptoms and self-perceived social support. As shown in Table 3.2, there were minimal differences across age groups in that all showed a negative association between depressive symptoms and self-perceived social support. The strength of the association - the value of Kendall's Tau - also varied little by age.

Notes: N = 10,258 participants aged 18-60 years; those without valid responses to PHQ-9 and MOS items were excluded from analyses; ^a Kendall's Tau-b values range between -1 (indicating a strong negative relationship) and 1 (a strong positive relationship); ***p < 0.01, **p < 0.05, *p < 0.1

Source: TTM data, Wave 2, adult cohort, unweighted

Figure 3.1 presents the mean PHQ-9 depression scores according to levels of self-perceived social support for men across each age group at Wave 2. Men with lower levels of self-perceived social support (i.e. scores <20) demonstrated considerable variability in their depressive symptom scores. This variability was particularly marked in young men. Among those aged 18-24, men with self-perceived social support scores of less than 20 had mean depressive symptom scores ranging from 6 to 15. In the oldest age group (45-60 years), men with social support scores of less than 20 had mean depressive symptom scores ranging from around 4.5 to 9.0.

Figure 3.1: Mean PHQ-9 score by MOS total score among adult males, by age group, 2015/16

Read text description.

Notes: N = 10,258 participants aged 18-60 years; those without valid responses to PHQ-9 and MOS items were excluded from analyses.

Source: TTM data, Wave 2, adult cohort, unweighted

Figure 3.2 shows the converse relationship of the above; that is, the association of the mean self-perceived social support score according to the PHQ-9 depression score. Again, the negative association between the two measures can be seen across all age groups: as depressive scores increased, mean self-perceived social support scores decreased. However, similar to before, men with higher levels of depressive symptoms (scores >20) had greater variability in their self-perceived social support levels than those with lower depression scores. Again, this variability was more pronounced among younger men. Adult males aged 18-24 with depressive symptom scores higher than 20 had mean self-perceived social support scores ranging from around 10 to 75. Those aged 25-34 with depression scores higher than 20 had mean perceived support scores ranging from around 25 to 100. However, men aged 45-60 with high depression scores had mean support scores that varied between 25 and 60.

Figure 3.2: Mean MOS score by total PHQ-9 score among adult males by age group, 2015/16

Read text description.

Notes: N = 10,258 participants aged 18-60 years; those without valid responses to PHQ-9 and MOS items were excluded from analyses.

Source: TTM data, Wave 2, adult cohort, unweighted

Modelling bi-directional relationships over time

The correlation results presented above showed that, at one time point, overall, levels of depression and social support were negatively correlated. That is, as depressive symptoms increased, self-perceived social support tended to decrease, and vice versa. However, further investigation was required to determine if changes in one variable was temporally related; that is, caused changes in the other. To this end, statistical models - specifically, cross-lagged panel models (CLPMs) - were developed to provide insight into whether changes in depression among Australian men may have caused levels of self-perceived social support levels to change, and vice versa.

Using data from Waves 1 and 2 of TTM (2013/14 and 2015/16), an unadjusted³ CLPM first accounted for how levels of depression and self-perceived social support interacted with each other over time. Subsequently, a multivariate CLPM was developed that controlled for differences between men in terms of certain socio-demographic factors such as education level, employment status and age. Additional details are below.

Figure 3.3 is a schematic diagram of the unadjusted CLPM (Zyphur et al., 2020). It shows the different relationships that were tested. The models evaluated the extent to which depressive symptoms measured at Wave 1 predicted levels of self-perceived social support at Wave 2 (relationship a). They also tested whether the level of self-perceived social support at Wave 1 affected depressive symptoms at Wave 2 (relationship b); therefore, allowing us to describe the influence of past depression on future self-perceived social support, and vice versa (i.e. the bi-directional cross-lagged relationship). Additionally, the models were used to determine the extent to which depressive symptoms at Wave 1 influenced symptoms two years later (relationship c), and whether social support levels at Wave 1 predicted those at Wave 2 (relationship d).

Figure 3.3: A two-wave, two-variable cross-lagged panel model

Read text description.

Unadjusted cross-lagged analysis: Depression and self-perceived social support

On an unadjusted level, men with relatively higher levels of depressive symptoms at Wave 1 (2013/14) were found to have lower levels of self-perceived social support compared to others at Wave 2 (2015/16) (relationship a in Figure 3.3). Conversely, a lower level of self-perceived social support among men at Wave 1 was associated with a higher level of depressive symptoms at Wave 2 (relationship b). Notably, the negative effect of depressive symptoms on self-perceived social support (-0.688) was substantially larger than the positive association of self-perceived social support on depression (-0.007).

Both self-perceived social support and depressive symptoms were strongly predicted by their past levels. Men with relatively high depressive symptoms at Wave 1 were also likely to have a high level of symptoms at Wave 2, compared to others (relationship c in Figure 3.3). Men with low self-perceived social support compared to others at Wave 1 also reported low self-perceived social support at Wave 2⁴ (relationship d).

Multivariate cross-lagged analysis: Depression and self-perceived social support

Table 3.3 and Figure 3.4 present the findings from the multivariate cross-lagged panel analysis. This model tested the bi-directional relationship between depression and self-perceived social support while accounting for the influences of key socio-demographic factors measured at Wave 1: age, Aboriginal and Torres Strait Islander status, sexual orientation, language background, marital status, remoteness of residential location, neighbourhood disadvantage, education level, employment status, conformity to masculine norms and number of close friends and relatives, all measured at Wave 1.

Overall, the multivariate findings indicated that among Australian men, there was a bi-directional relationship between self-perceived social support and depressive symptoms. Those with relatively low levels of self-perceived social support were likely to have higher levels of depressive symptoms compared to others 2-3 years later, while higher levels of depressive symptoms led to poor self-perceived social support.

The influence of depressive symptoms on self-perceived social support appeared to be stronger than the effect of social support on depression. After adjusting for socio-demographic characteristics, the effect of self-perceived social support at Wave 1 on depressive symptoms at Wave 2 remained unaltered compared to the unadjusted model (-0.007, relationship b in Figure 3.4); however, the effect of depressive symptoms at Wave 1 on self-perceived social support at Wave 2 increased slightly (-0.700) (Figure 3.4, relationship a).

In the multivariable model (after accounting for key socio-demographics), both depressive symptoms and self-perceived social support continued to be significantly predicted by their own past levels. Men with high levels of self-perceived social support at Wave 1, relative to others, could expect to have high levels at Wave 2. Likewise, those with relatively low levels at Wave 1 would likely have low levels 2-3 years later (relationship d in Figure 3.4). Similarly, men with relatively high depressive symptoms at Wave 1 could expect high depressive symptoms at Wave 2 (relationship c).

Notes: N = 8,646. ***p < 0.01, **p < 0.05, *p < 0.1; CALD = culturally and linguistically diverse; MOS = Medical Outcomes Study social support scale; PHQ-9 = Patient Health Questionnaire (PHQ) depressive symptoms score; Identifying as non-heterosexual included men who identified as bisexual, homosexual, not sure or other. MOS Social Support Scale scores for the TTM sample ranged from 0-100; mean = 68.8; standard deviation = 26.9. PWI scores for the sample ranged from 1-99; mean = 69.8; standard deviation = 17.3.

Source: TTM data, Waves 1 and 2, adult cohort, unweighted

Figure 3.4: Two-wave cross-lagged panel model: Results with socio-demographic controls

Read text description.

Notes: N = 8,646 participants aged 18-60 years; those without valid responses to depression PHQ-9 items, social support MOS items and socio-demographic factors were excluded from the analyses; ***p < 0.01, **p < 0.05, *p < 0.1.

Source: TTM data, Waves 1 and 2, adult cohort, balanced sample, unweighted

The association of key socio-demographic characteristics with depressive symptoms and self-perceived social support 

When examining the influence of various socio-demographic factors on experience of depressive symptoms and levels of self-perceived social support among adult TTM participants (Table 3.3), the findings showed that:

1. Getting older was associated with both a decline in self-perceived social support and a decrease in depressive symptoms. Every one-year increase in age was associated with a 0.18 unit decrease in the social support score, and a 0.02 unit decrease in the depression score.
2. Aboriginal and Torres Strait Islander men, men from culturally and linguistically diverse backgrounds (CALD), those living in higher disadvantaged areas, and men with no close friends or relatives were more likely to experience lower self-perceived social support. For instance, being from a CALD background was associated with a 3.28 unit decrease in the social support score. Similarly, having no close friends or relatives was associated with a 2.81 unit decrease in the social support score.
3. Being unemployed or out of the labour force and identifying as non-heterosexual were associated with experiencing more severe depressive symptoms. Men with lower levels of education and those living in more disadvantaged areas also experienced an increased likelihood of more severe depressive symptoms. For instance, unemployment was associated with a 0.73 unit increase in the depression score compared to being employed. Identifying as non-heterosexual was associated with a 0.51 unit increase in the depression score.
4. Men from CALD backgrounds were more likely to report fewer depressive symptoms (a decrease of 0.38 units in the depression score).
5. Conformity to traditional masculine norms was not significantly associated with either depression or self-perceived social support.

Conclusion

Depression, a form of mental illness, is associated with a large global health burden. Previous analyses using TTM data have shown that, on average, one in five Australian men will experience depression at some stage of their lives, and around one in eight will experience symptoms in a 12-month period (Terhaag, Quinn, Swami, & Daraganova, 2020). The extent that mental health impacts a person's life can vary. This often depends on factors such as age and life stage or in response to particular life experiences, events and stressors (Nurius, Uehara, & Zatzick, 2013).

Social support plays an important role in creating and maintaining good mental health (Ozbay et al., 2007). This chapter examined the bi-directional relationship between self-perceived social support and depression among Australian men. Findings demonstrated that depressive symptoms and self-perceived social support were contemporaneously correlated with each other; on average, men with higher depressive symptoms experienced lower self-perceived social support over the same time period, and vice versa.

The findings of this chapter are consistent with interpersonal theories that emphasise that depression can adversely affect levels of social support and interpersonal relationships (Hames, Hagan, & Joiner, 2013; Segrin, 2001). These theories posit that the experience of depression leads to social disconnectedness, interpersonal stress and social withdrawal, which can exacerbate or maintain the experience of depressive symptoms, thus creating a cycle of depression and poor social support.

However, the findings do conflict somewhat with those of Saeri and colleagues (2018), who found that perceived social connectedness was a stronger predictor of mental health than mental health was of social connectedness among a representative sample of New Zealanders. TTM findings indicated the opposite - the effect of depression on social support was substantially greater than the effect of social support on depression. This discrepancy may be explained by different measures of social connectedness/support and mental ill-health, in addition to distinctly different target populations.

Future waves of TTM data collection will enable further examination of the relationship between social support and mental ill-health among Australian men over time.

Community engagement and participation among adult Australian males

Brendan Quinn, Jennifer Prattley, Neha Swami, Sonia Terhaag and Bosco Rowland

Overview

While limited social connectedness is associated with adverse health outcomes, certain types of social or community integration or involvement - such as volunteering, club membership or participation in regular sporting activities - have been shown to be associated with a range of positive health and social outcomes, including the expansion of social networks and social capital. For example, Yeung and colleagues (2018) recently showed that, among around 1,500 Texan adults, volunteering¹³ was predictive of better mental and physical health, in addition to enhanced life satisfaction and social wellbeing. Specifically, it was associated with increased feelings of social integration and acceptance and satisfaction with relationships with others.

Similarly, there is substantive, consistent evidence that sports participation is associated with health and social benefits (Hoye, Nicholson, & Brown, 2015), particularly club- or team-based sports due to the social nature of participation (Eime et al., 2013). A systematic review of the social and psychological health outcomes of team sports participation in adults found that the most reported benefits were improved social support, sense of belonging, social interaction, higher self-esteem and enhanced social networks (Andersen, Ottesen, & Thing, 2019). Prolonged involvement in such activities has been associated with a greater experience of positive health and social outcomes vs more temporary exposure (Oosterhoff, Kaplow, Wray-Lake, & Gallagher, 2017).

Only a small proportion of Australians are involved with volunteering and sport. The most recent national data indicate approximately three in 10 Australians (31%) aged 15 and over have undertaken some type of voluntary work (AIHW, 2019a). Rates of volunteering are lower among males (46%) compared to females (54%).

An analysis of involvement in Victorian club-based sports during 2017 indicated that participation rates fell with age: around 32% of 15-19 year olds were involved with 12 major club-based sports,¹⁴ compared to 15% of those aged 20-24, and less than 10% of Victorians aged 30-85+ (Sport Participation Research Project, 2019). However, overall, participation rates were higher among males (20%) compared to females (11%).

Despite the established associations between community engagement activities and positive health outcomes, few studies have investigated possible pathways that provide further context for relationships and mechanisms between certain types of community participation and health and wellbeing (Malinauskas & Malinauskiene, 2018; Mellor et al., 2008; Yeung et al., 2018). One exception is Brown, Hoye, and Nicholson (2012), who found that, among a random sample of adult Victorians (N = 2,990), engaging in volunteering was positively associated with social connectedness, which then improved personal wellbeing.

Identifying the characteristics of adult males who are less likely to get involved in community-based activities such as volunteering and sport could inform targeted initiatives to improve participation rates, with flow-on effects including improved social connectedness, health and wellbeing among Australian men.

Research objectives

This chapter uses data from Wave 1 of Ten to Men (TTM). ¹⁵ It focuses on exploring the community integration domain of social connectedness among adult Australian males; specifically, involvement in community-based activities such as volunteering and sport in 2013/14. It sought to:

• outline the extent of community engagement, involvement or participation among adult Australian males, with a focus on community service activities (e.g. volunteering) and membership of sporting or hobby clubs or associations
• explore whether involvement in community service activities and membership of sporting or hobby clubs/associations is associated with greater levels of self-perceived social support
• investigate the socio-demographic and psychosocial characteristics of males who were more likely to participate in such activities
• examine pathways to wellbeing: specifically, whether community engagement affects self-perceived social support and whether this, in turn, enhances personal wellbeing.

Community integration: Participation in community-based activities among adult males in 2013/14

Types of community-based activities

Rates of involvement in different community-based activities among Australian men aged 18-55 in 2013/14 are detailed in Chapter 1 (Table 1.2); however, Figure 5.1 shows that participation rates vary by age. Overall, less than one-quarter of all adult males took part in any sort of ongoing community service activity, such as volunteering, whereas around 37% were active members of a sports, hobby or community-based club or association.

Involvement in ongoing community service was significantly higher among males aged 35 and above (around 26%-28%) compared to those aged 18-34 (19%-20%). In contrast, younger men aged 18-24 were significantly more likely to be active members of sporting, hobby or community-based clubs or associations (43%) compared to those aged 25 and above (35%-36%).

Around one-third (35%) of adult males aged 18-55 did not participate in any of the community-based activities assessed here.

Figure 5.1: Estimated prevalence of participation in three types of community-based activities among adult males in 2013/14, by age

Read text description.

Notes: Ongoing community service activity - n = 13,399; Active member of sport/hobby club/assoc. - n = 13,394; Ever attends religious ceremonies - n = 13,450; None - n = 13,428. 95% Confidence Intervals

Source: TTM data, Wave 1, adult cohort, weighted

Social support by community engagement

As outlined in Chapter 1 and this report's methodology (Appendix A), subjective perceptions of levels of social support among TTM participants were measured using the MOS Social Support Scale (Sherbourne & Stewart, 1991). Figures 5.2a and 5.2b show that, on average, adult males who participated in certain community-based activities had significantly higher levels of self-perceived social support compared to those who did not. Men who participated in ongoing community service activities had an average self-perceived social support score of 71, compared to 69 among those not engaged in community service activities (Figure 5.2a). Likewise, Australian men who held active memberships with community-based sports or hobby clubs or associations had significantly greater levels of self-perceived social support compared to those without such memberships (mean = 73 vs 68, respectively) (Figure 5.2b).

Figure 5.2: Average MOS Social Support Scale score by level and type of community engagement among adult males aged 18-55 in 2013/14

Read text description.

Notes: a) n = 13,220; b) n = 13,216. 95% Confidence Intervals

Source: TTM data, Wave 1, adult cohort, weighted

Characteristics of adult males who participated in community-based activities in 2013/14

The TTM surveys collect information on a variety of socio-demographic, health and psychosocial measures. Two multivariable analyses examining associations between some of these factors and the following two outcomes were undertaken:

1. ongoing community service activity
2. active membership of a community-based sports or hobby club or association.

These analyses enable examination of whether particular groups or characteristics were associated with increased or decreased likelihoods of participating in community activities, and identification of who may benefit from additional or different avenues for participation. Findings are presented in Table 5.2.

Ongoing community service activity

Older men were more likely to be involved in ongoing community service activities compared to younger men. The odds of participating in community service also increased with greater levels of education; university-educated participants were 90% more likely to engage in community service compared to those with an education level of Year 12 or lower.

Other factors significantly associated with increased odds of participating in community service activities included living in inner or outer regional areas of Australia (compared to major cities; aOR = 1.38 and 1.42, respectively), having high personal wellbeing (aOR = 1.02), and living in areas classified as being of moderate socio-economic disadvantage (vs high disadvantage; aOR = 1.17). Men living with their partner and children were 30% more likely to participate in an ongoing community service activity compared to those living alone.

Men who spoke a language other than English at home had around a 29% lower likelihood of being involved in an ongoing community service activity. Greater conformity to masculine norms was also associated with a significantly reduced likelihood of engaging in community service (aOR = 0.99).

Employment status, having Aboriginal and Torres Strait Islander status, and experiencing depression were not associated with involvement in any ongoing community service activity when adjusting for other socio-demographic and psychosocial characteristics.

Active membership of community sports or hobby club/association

Table 5.1 also presents findings from the multivariable analysis investigating factors associated with community sports or hobby club participation. As opposed to ongoing community service activity, older age was consistently associated with a significantly reduced likelihood of having an active membership of a community-based sports or hobby club or association among Australian men. Compared to those aged 18-24, adult males aged 25 and above had around a 25% reduced likelihood of being an active member of a community sports or hobby club/association.

Unemployed men and those out of the labour force were 20%-30% less likely to have an active membership of a community-based sports or hobby club/association, while those with moderate to high depression were around 15% less likely to be members of sports or hobby clubs/associations. Adult males who spoke a language other than English at home were around 51% less likely to be an active member of a community sports or hobby club/association than those who only spoke English at home.

In contrast, higher levels of education, less socio-economic disadvantage, greater personal wellbeing, and living in inner or outer regional areas (vs major cities) were associated with an increased likelihood of being an active member of a community sports or hobby club/association. For example, adult males living in areas of middle or low disadvantage both had around a 22% greater likelihood of having an active membership with a community-based sports or hobby club/association compared to those living in areas classified as being of high disadvantage.

Conformity to masculine norms, identifying as Aboriginal and Torres Strait Islander and household composition were not significantly associated with membership of community-based sports or hobby clubs/associations.

Notes: Observations excluded from analyses included 55 participants living in an 'other' type of household composition and 742 with missing data for this variable, respectively. ^a Includes volunteering at a school, coaching a sports team, working with a church or neighbourhood association. aOR = adjusted Odds Ratio; ASGS = Australian Statistical Geography Standard; CMNI = Conformity to Masculine Norms Inventory; OTE = Other than English; PHQ = Patient Health Questionnaire; PWI = Personal Wellbeing Index; SE = (robust) Standard Error; SEIFA = Socio-Economic Indexes for Areas; W1 = Wave 1 status; *p < 0.05, **p < 0.01, ***p < 0.001.

Source: TTM data, Wave 1, adult cohort, unweighted

Mediation analysis of community participation on personal wellbeing through self-perceived social support

Determining how different types of community participation affect health outcomes, including the magnitude and direction of specific pathways and other contributing factors, is important for understanding which types of activities are beneficial for improving the health and wellbeing of males in Australia. A mediation analysis aims to examine a hypothesised causal relationship or pathway between an observed independent variable/predictor/exposure (hereafter: 'independent variable') and a dependent variable/outcome via the inclusion of a third variable, known as a mediator (MacKinnon, Fairchild, & Fritz, 2007). The following mediation analyses first used unadjusted¹⁶ models to investigate whether the association between involvement in two types of community-based activities - ongoing community service, such as volunteering, and active membership with sports or hobby clubs or associations - and personal wellbeing among Australians were either partially or fully mediated¹⁷ by self-perceived social support.

Figure 5.3 details the nature of each of the three key variables (i.e. independent, mediator and dependent), in addition to the pathways to be examined between the variables (a, b, c and c'). Specifically, analyses aimed to determine the effect (direction - positive or negative - and amount) the independent variable (community-based activity involvement) had on the dependent variable (personal wellbeing), and how much of this relationship was due to the effect of the mediator (social support).

The subsequent multivariable analyses further investigated these relationships in consideration of various socio-demographic and psychosocial characteristics.

Figure 5.3: Hypothetical mediation model: Independent Variable = Community participation activity; Mediator Variable = Self-perceived social support; Dependent Variable = Personal wellbeing

Read text description.

Notes: Self-perceived social support was assessed using the MOS Social Support Scale; Personal wellbeing was measured using the Personal Wellbeing Index (refer: Methods section). The Total Effect (c) of different types of community participation [Independent Variables (exposures/predictors)] on personal wellbeing [Dependent Variable (outcome)] is determined by any direct effects between these two variables (c'), while also considering any relationships between the Independent Variable and Mediator (social support), and social support and the Dependent Variable: c = c' + (a*b)

Unadjusted mediation analysis: Community service participation (e.g. volunteering)

On an unadjusted level (i.e. not taking other factors into account, such as socio-demographic characteristics),¹⁸ men's participation in ongoing community service activity was positively associated with increased personal wellbeing. This relationship was partly mediated by social support.

As presented in Table 5.2, the overall - total - effect of participation (path c) was an average increase of 5.2 points on the PWI scale for adult males who engaged in ongoing community service activity compared to those who did not.

There were two components to this overall effect. Firstly, there was a significant direct effect (path c'; see Figure 5.4) of participation in ongoing community service on personal wellbeing, which comprised the majority - 86% - of the overall association. Secondly, 14% of the association between community service activity and personal wellbeing was mediated through social support. Stated another way, there is an indirect effect in which participation in ongoing community service increased men's perceived level of social support; this, in turn, had a positive impact on wellbeing and accounted for the remaining 14% of the overall/total effect.

Notes: n = 12,712; ***p < 0.001. CI = Confidence Intervals

Source: TTM data, Wave 1, adult cohort, unweighted

Figure 5.4: Unadjusted mediation model: Independent Variable = Community service activity; Mediator Variable = Self-perceived social support; Dependent Variable = Personal wellbeing

Read text description.

Unadjusted mediation analysis: Active membership of sports or hobby-based clubs or associations

Table 5.3 shows that active membership of sports or hobby clubs or associations was also positively associated with personal wellbeing, and this relationship was partly mediated by social support. The overall - total - effect of participation was an average increase of 5.7 points on the PWI scale for Australian men who were active members of sports or hobby-based clubs or associations compared to those who were not (Figure 5.5; path c). As above, there were two components to the overall effect. Firstly, there was a significant direct effect between sports or hobby club/association membership and personal wellbeing, which comprised 82% of the overall association. Secondly, there was an indirect effect in which sports or hobby club/association membership increased men's perceived level of social support which, in turn, positively affected personal wellbeing and accounted for the remaining 18% of the overall/total effect.

In consideration of these findings, and those above pertaining to ongoing involvement in community service activities, self-perceived social support appears to be an important mechanism through which community engagement improves wellbeing. Involvement in community-based activities such as volunteering and sports/hobby club membership increased men's social connections, which improved their sense of social support, and this had a flow-on positive impact on their overall wellbeing.

Notes: n = 12,710; ***p < 0.001. CI = Confidence Intervals

Source: TTM data, Wave 1, adult cohort, unweighted

Figure 5.5: Unadjusted mediation model: Independent Variable = Sports/hobby club/association membership; Mediator Variable = Self-perceived social support; Dependent Variable = Personal wellbeing

Read text description.

Multivariable mediation analyses

Table 5.4 summarises two multivariable analyses investigating whether self-perceived social support mediates the relationship between community participation and personal wellbeing. As above, community participation was defined in the two respective models as: 1) ongoing involvement in community service activities such as volunteering; and 2) active membership of a sports, hobby or community-based club or association. Figures 5.6 and 5.7 depict the pathways between the independent, dependent and mediator variables, in addition to the socio-demographic and psychosocial characteristics taken into consideration in both multivariable models.

Overall, the findings of both models indicated that, after controlling for various socio-demographic and psychosocial characteristics, community participation is positively associated with self-perceived social support ('a' path), and self-perceived social support is positively related to personal wellbeing ('b' path). The total effect ('c' path) of both independent variables on personal wellbeing was also positive and significant (p < 0.001) in each instance (β = 4.03 and 4.01 in Models 1 and 2, respectively), meaning that involvement in ongoing community service activities and active membership with a sports/hobby club have indirect and positive associations with personal wellbeing through higher levels of self-perceived social support.

When examining the partial effects of the covariates, in both Models 1 and 2, all socio-demographic and psychosocial characteristics were significantly associated with personal wellbeing, except Aboriginal and Torres Strait Islander status. In both models, being employed, being married or in a de facto relationship, having a higher level of education, having less disadvantage, and speaking a language other than English at home had positive partial effects on personal wellbeing; employment status had the largest positive effect size in both models. In contrast, older age, living in a metropolitan area and having no close friends or relatives were negatively correlated with personal wellbeing; being alone had the largest negative effect size in both models.

Notes: ASGS = Australian Statistical Geography Standard; CMNI = Conformity to Masculine Norms Inventory; CV = Covariate; DV = Dependent variable; IV = Independent variable; M = Mediator; SE = Standard error; SEIFA = Socio-Economic Indexes for Areas *p < 0.05, **p < 0.01, ***p < 0.001. ^a Includes volunteering at a school, coaching a sports team, working with a church or neighbourhood association; ^b Assessed using the Personal Wellbeing Index; ^c Assessed using the MOS Social Support Scale

Source: TTM data, Wave 1, adult cohort, unweighted

Figure 5.6: Multivariable mediation analyses: Participation in community service activities on personal wellbeing through self-perceived social support

Read text description.

Figure 5.7: Multivariable mediation analyses: Active sports or hobby club or association membership on personal wellbeing through self-perceived social support

Read text description.

Conclusion

TTM findings presented in this chapter provide support for the experience of enhanced social connectedness among adult Australian males due to community integration or participation, primarily in the form of ongoing community service activities such as volunteering, and active membership of sports or hobby clubs or associations. In 2013/14, men who engaged in such community-based activities reported significantly greater average levels of self-perceived social support compared to those who did not take part. These findings align with those of other research (e.g. Andersen et al., 2019; Brown et al., 2012; Dias, Geard, Campbell, Warr, & McVernon, 2018).

This chapter investigated the cross-sectional relationship between involvement in ongoing community service activities and active membership of sports or hobby clubs/associations and personal wellbeing among Australian men, as mediated by self-perceived social support and controlling for various socio-demographic and psychosocial characteristics. Findings indicated that involvement in these community-based activities directly improved personal wellbeing, while also enhancing self-perceived social support, which further indirectly and positively affected wellbeing.

This builds on previous Australian and international findings, including those of Brown and colleagues (2012), who found that the relationship between volunteering and wellbeing was mediated by social connectedness among a random sample of around 3,000 Victorian adults surveyed in 2010. Similarly, Haugen, Säfvenbom, and Ommundsen (2013) found that the link between sports participation and loneliness was mediated by perceived social competence among around 2,000 secondary students in Norway.

Importantly, findings detailed in this chapter point to specific subgroups of Australian men with lower rates of engagement in certain social activities in 2013/14, such as men from non-English speaking or CALD backgrounds. CALD men were less likely to both participate in ongoing community service activities and be active members of sports or hobby clubs/associations. It is possible that TTM findings underestimate the extent of community service involvement among CALD men; for example, Australian research has indicated that people from CALD communities often participate in informal or unofficial volunteering activities - such as child care, helping with domestic tasks, and care of older people or those with long-term illness and disability - which might not be captured by surveys asking about specific types of volunteering (AIHW, 2017; Cultural & Indigenous Research Centre Australia, 2016). Nevertheless, research has shown that barriers exist to volunteering for CALD communities in Australia, including English language requirements, lack of time (especially to partake in volunteering activities outside local/CALD communities), and potential exploitation and racism (Cultural & Indigenous Research Centre Australia, 2016).

Regarding age differences, active memberships with sports and hobby clubs/associations were significantly higher among Australian men aged 18-24 compared to older cohorts. This reflects study findings relating to involvement in organised Victorian club-based sports at the same time of Wave 2 TTM data collection (and in the years following), which consistently indicate that participation rates among both males and females in Victoria fall with age (Eime, Harvey, & Charity, 2016; Victorian Health Promotion Foundation, 2021). Research points to numerous barriers to sports participation for Australian adults, with the main one being a lack of time or too many other commitments (37%) (AIHW, 2018). Having a job was also a key barrier to sports participation for 13% of 18-64 year olds. For around one-quarter of older adults aged 45-54 and 55-64 (25% and 28%, respectively), poor health or injury prevented participation in sporting activities.

Appendix A: Methodology

Study design and sample

TTM is a national prospective cohort study of over 16,000 males commissioned by the Commonwealth Department of Health. It is the first large-scale, nationally representative, longitudinal study to focus exclusively on investigating and improving the health and wellbeing of males in Australia.

The Human Research Ethics Committee at the University of Melbourne approved TTM pilot studies and the first wave of data collection. At recruitment, the sample comprised three distinct cohorts: 'boys' (aged 10-14 years), 'young men' (15-17 years), and 'adults' (18-57 years). Each cohort was administered distinct tailored surveys at Waves 1 and 2; that is, participants in the boys and young men cohorts were sometimes asked different questions to adult participants.

The sample was recruited using a stratified, multi-stage, cluster random sampling approach. Further details on the methodology used for TTM are published elsewhere (e.g. Currier et al., 2016; Daraganova & Quinn B. [Eds.], 2020; Pirkis et al., 2017), including in the study's Data User Guide (Bandara et al., 2019).

Data collection for the research conducted in this report was undertaken by the University of Melbourne across two waves in 2013/14 and 2015/16 (Waves 1 and 2, respectively). In consideration of the survey content for each of the three cohorts at the first two waves, this report focuses solely on adult participants. At Wave 1 there were 13,896 adult men; at Wave 2 there were 10,729 adult participants. Around three-quarters (74%; n = 10,339) of adult participants at Wave 1 were also surveyed at Wave 2.

Measures of social connectedness

The measures detailed below were asked at both Waves 1 and 2 unless otherwise indicated. Additional information on certain measures is included in the empirical chapters if they are a key focus of analyses.

Self-perceived social support

Self-perceived social support is one of the main variables of interest in analyses presented throughout this report. It was assessed in Waves 1 and 2 using the Emotional/Informational Support subscale from the Medical Outcomes Study (MOS) Social Support Survey (Sherbourne & Stewart, 1991). The subscale comprises eight items (Box A1). Respondents indicated how often each type of emotional or informational support was available to them if needed, using the options: (1) None of the time; (2) A little of the time; (3) Some of the time; (4) Most of the time and (5) All of the time. The minimum possible score for a respondent over all eight items was 8, and the maximum possible was 40. As is standard practice, MOS Emotional/Informational Support scores were only calculated if valid responses were provided for all eight items.

To create an overall self-perceived social support score ranging 0-100, where higher scores indicate greater levels of support, the following formula was used:

Attachment relationships

Marital status

Adult TTM participants were asked to classify their marital status as: (1) Never married; (2) Widowed; (3) Divorced; (4) Separated but not divorced or (5) Married/de facto.

Number of children

TTM participants were asked to report how many children (including adult children) they had. Responses were then divided into four categories: (1) No children; (2) One child; (3) Two to four children and (4) Five or more children.

Participants with children were asked how many were aged under 18. A binary variable was generated indicating that they had (1) No children under 18 or (2) One or more child under 18.

Respondents with children under 18 were asked if any of those children sometimes or always lived elsewhere. Response options were: (1) No - always live with me; (2) Yes - sometimes live elsewhere and (3) Yes - always live elsewhere.

Household composition

Household composition was determined using data collected on a) Marital status, b) Number of children, and c) Number of people respondents lived with. A variable was generated comprising the following categories: (1) Single, lives alone; (2) Couple, no children; (3) Couple with children; (4) Single parent; (5) Single, living with others (e.g. other family, friends, share house) and (6) Other (e.g. couple lives apart).

Satisfaction with personal relationships

As part of the Personal Wellbeing Index (PWI; see Subjective wellbeing under the health-related measures) (International Wellbeing Group, 2013), adult participants were asked: 'On a scale of 0 to 10 where 0 means you feel completely dissatisfied, 10 means you feel completely satisfied and 5 means you feel neutral, neither satisfied or dissatisfied, how satisfied are you with your personal relationships?'

Number of close friends or family ('Aloneness')

Participants were asked to estimate how many close friends and relatives they had, defined as 'people you feel at ease with and can talk to about what is on your mind'. For analytic purposes, participants were categorised as having (1) No close friends or relatives (i.e. experiencing 'aloneness') or (2) At least one close family member or friend. In accordance with previous literature (Andersson, Denhov, Bülow, & Topor, 2015; Heinrich & Gullone, 2006), the concept of aloneness used here indicates the absence of social contacts. Importantly, it is not necessarily a problematic, unpleasant or undesired experience; in contrast, loneliness is a lack of desired social relationships and an unwanted situation.

Community integration

Employment status

Current employment status was indicated using the following options: (1) Employed/working for profit or pay; (2) Unemployed and looking for work; or (3) Neither working nor looking for work. Employed participants indicated if their status was: (1) Permanent; (2) Casual work; (3) Self-employment or (4) Fixed-term contract work.

Community engagement

TTM participants were asked, at Wave 1 only, about their involvement in different types of social, community-based activities:

• Ongoing community service activity; for example, volunteering at a school, coaching a sports team or working with a church or neighbourhood (Yes/No).
• Active membership of a sporting, hobby or community-based club or association (Yes/No).
• Frequency of attendance at religious services; for example, going to church, temple, mosque, or other religious institutions or activities ('Never', 'About once or twice a year', 'About once every few months', 'About once or twice a month', 'About once a week or more often').
• Frequency of attendance at events that bring people together such as fetes, festivals or other community events ('Never', 'Rarely', 'Occasionally', 'Sometimes', 'Often', 'Very often').

Health-related measures

Subjective wellbeing

General life satisfaction was measured using the Personal Wellbeing Index (PWI; International Wellbeing Group, 2013), a tool intended to capture satisfaction across various life domains (e.g. health, relationships, safety, future security, life achievements). PWI scores provide an indicator of subjective wellbeing (higher scores indicate greater overall life satisfaction). The score distribution for the TTM sample (range: 1-99) was divided into quartiles indicating classifications of low (bottom 25%), middle 50%, and high (top 25%) satisfaction. Analyses for this report used a binary variable indicating low vs middle-high satisfaction.

Depression

Experience of depression was assessed using the Patient Health Questionnaire (PHQ-9) (Kroenke et al., 2001), which measures frequency of experiencing the following depressive symptoms over the past two weeks:

• Little interest/pleasure in doing things
• Feeling depressed/hopeless
• Trouble with sleep
• Tired/little energy
• Poor appetite/overeating
• Feeling of being failure
• Trouble with concentration
• Moving slower or being more restless
• Thinks better off dead/of hurting self.

Responses include: (0) Not at all; (1) Several days; (2) More than half of days and (3) Nearly every day. Individual item scores were added to create a total PHQ score that ranged 0-27; higher scores indicate greater depression severity.

Disability

Functional difficulty among Australian men was assessed in TTM using the Washington Group Short Set of questions on functioning (WG-SS). The WG-SS is a widely used instrument for assessing difficulty and disability and has been validated in numerous countries (Madans et al., 2011). The tool consists of six questions relating to core functional domains (vision, hearing, walking/movement, cognition, self-care, and comprehension and communication), with four response options on a severity scale:

• No difficulty
• Some difficulty
• A lot of difficulty
• Cannot do at all.

TTM participants were classified as having serious disability if they reported having more than some difficulty (i.e. either a lot of difficulty or cannot do at all) across at least one of the core domains. In other research (Currier et al., 2016; Madans et al., 2011) this would correspond with a 'disability' classification.

Socio-demographic and other personal characteristics

Area level of disadvantage

The area level of socio-economic disadvantage of TTM respondents' residential locations was assessed using the Socio-Economic Indexes for Areas (SEIFA),²¹ developed by the ABS. Scores were divided into three categories: (1) Lowest quartile (25%); (2) Middle 50% and (3) Highest 25%.

Residential region

Participants' residential locations were classified according to the Australian Statistical Geography Standard (ASGS):²² (1) Major cities; (2) Inner regional areas; (3) Outer regional areas and (4) Remote/very remote areas.

Indigenous status

Participants were asked if they identified as Aboriginal, Torres Strait Islander, both, or neither. A binary variable was generated indicating Indigenous status: (1) Aboriginal and Torres Strait Islander or (2) Non-Indigenous.

Language background

TTM participants who reported mainly speaking a language other than English at home were classified as culturally and linguistically diverse (CALD).

Education

TTM participants were asked about the highest qualification completed 'after leaving school'. A variable was derived comprising four categories:

1. Year 12 or less (includes those who have not completed any additional qualification)
2. Certificate/diploma (includes trade certificate, non-trade certificate, associate diploma, undergraduate diploma)
3. University degree (includes bachelor's degree, masters or postgraduate degree, doctorate
4. Other.

Sexual orientation

At Wave 1 of TTM, participants aged 15 and older were asked if they identified as: heterosexual; bisexual; homosexual; 'not sure'; or 'other'. A binary variable was generated to indicate sexual orientation; participants were classified as either (1) Non-heterosexual (i.e. identified as bisexual, homosexual, not sure, or other) or (2) Heterosexual.

Conformity to masculine norms

The extent of conformity to traditional masculine roles was measured using the Conformity to Masculine Norms Inventory (CMNI) (Mahalik et al., 2003). The CMNI is comprised of 22 items that capture notions of winning, emotional control, risk taking, violence, dominance, sexual behaviours and attitudes, self-reliance, primacy of work, power over women, attitudes towards homosexuality, and pursuit of status. Each item is answered on a four-point scale: (0) Strongly disagree; (1) Disagree; (2) Agree and (3) Strongly agree. Total scores ranged 0-66, with higher scores indicating greater conformity with masculine norms. Men were categorised into one of three groups depending on their total score: (1) those in the lowest quartile (25%), (2) the middle 50% and (3) the highest 25%.

Statistical methods and terms

This section provides an overview of the statistical methods and terms used throughout this report. The definitions here draw on the Statistical Language Glossary produced by the ABS.²³ More detailed explanations for some techniques, such as the cross-lagged panel and mediation models, are provided in the respective chapters where they have been used.

Measures of central tendency and spread

The most common measures of central tendency are the arithmetic mean, the median and the mode. Measures of spread include the range, quartiles and the interquartile range, variance and standard deviation (SD). Each of these are described in detail below.

Mean, median and mode

The mean, median and mode are three measures that represent the middle or typical value of a dataset of values. The mean is the statistical term used for what is more commonly known as the average - the sum of the values of a data series divided by the number of data points. The median is the middle value obtained when the values are arranged in order from smallest to largest. The mode is the most commonly occurring value in the dataset.

Range and quartiles

The range is the difference between the largest and smallest values in a set of data.

Quartiles divide the data (when ordered from smallest to largest) into four equal parts. The lowest or first quartile is the value below which 25% of the data values lie, and the upper or third quartile is the value above which 25% of the data lie. Half of the data points (50%) fall between the first and third quartile.

Variance and standard deviation

The variance and standard deviation (SD) are measures that summarise how close data values are to the mean. The SD is the square root of the variance. If all data values are close to the mean, then the variance and standard deviation will be small. Both measures will be higher if data values are more spread out and dispersed further away from the mean. The smaller the variance and standard deviation, the more representative the mean is of all values.

Skewness

Skewness is the tendency for data values to be clustered around either the high or low end of a dataset, rather than the centre. It is a measure of symmetry. Positive skewness indicates there is a longer right 'tail' to the dataset; that is, fewer higher than lower values. Negative skewness indicates a longer left 'tail'; that is, fewer lower than higher values. A rule of thumb for interpreting the measure of skewness:

• If skewness = 0, the data are perfectly symmetrical around the centre.
• If skewness is between −½ and +½, the distribution is approximately symmetric.
• If skewness is between −1 and −½ or between +½ and +1, the distribution is moderately skewed.
• If skewness is less than −1 or greater than +1, the distribution is highly skewed.

Modelling techniques

Linear regression

Linear regression models are used to estimate the effects of one or more factors (also called independent or predictor variables or exposures) on a continuous dependent (outcome) variable. If an effect is found, then we say there is a relationship or association between the relevant factor and the dependent variable. Estimates (also referred to as regression coefficients, denoted by the symbol β) from linear regression models are interpreted as the impact of a one-unit change in the predictor on the dependent variable, if all other variables are held constant.

For example, to investigate the effect of age on total MOS Social Support Scale score we could fit a linear regression model to the sample data and obtain an estimated age effect of β = -0.31. This means that there is an association between age and the social support score, and for every additional year in age, an individual's score decreases by an estimated average 0.31 points (assuming values of all other variables are unchanged).

Linear regression could also be used to estimate the effects of employment status, for example, which is a categorical exposure, on the social support score. In this case, one category of the employment variable would be designated as the reference category. If the variable had two categories of 'employed' and 'unemployed', and 'employed' was the reference, then a one-unit change would mean a switch from that to the unemployed category. The results of the linear regression model would include an estimate of the effect on the social support score of being unemployed vs employed.

Multivariable linear regression models are more complex models that 'adjust' (i.e. take into account) for the effects of one or more other variables, termed 'covariates', on the dependent variable. Adding to the above example, investigating the effect of employment on total MOS Social Support Scale score, examples of covariates that could be added to generate a multivariable model include residential location, education level and gender. This allows the effect of the main predictor of interest (in this example, employment) to be separated from any effect that those other factors might have on the outcome (example, social support score). Results from a multivariate model include estimates of the effect that the main predictor and each of the other covariates have on the dependent variable.

Logistic regression and odds ratios (OR)

Logistic regression models are used to estimate the effects of one or more factors on a binary dependent (outcome) variable (a variable with only two distinct values such as 0 and 1). For example, a logistic regression model could be used to estimate the effect of age and highest level of education on a binary indicator (yes/no) of involvement in a community-based activity like volunteering, or the probability that an individual regularly attends religious services (vs not attending religious services).

Estimated coefficients from logistic regression models are referred to as odds ratios (ORs). ORs are a measure of association between an exposure and an outcome; they represent the odds that an outcome (e.g. experience of depressive symptoms) will occur given a particular exposure (e.g. having at least one close family member or friend) compared to the odds of the outcome occurring in the absence of that exposure (no close family members or friends). The OR can be used to determine whether a particular exposure is a risk factor for a particular outcome, and to compare the magnitude of various risk factors for that outcome:

• OR = 1 Exposure does not affect odds of outcome.
• OR > 1 Exposure is associated with higher odds of outcome.
• OR < 1 Exposure is associated with lower odds of outcome.

As described above, in relation to linear regression models, multivariable logistic regression models adjust for the effects of one or more other covariates on the dependent variable.

Cross-lagged panel models

Cross-lagged panel models allow us to identify and describe reciprocal and causal relationships (Zyphur et al., 2020). Using longitudinal data, results from a cross-lagged panel model can tell us how the levels of a variable measured at one point in time are influenced by past levels of that same variable (i.e. an autoregressive relationship). We can also examine whether and how two variables are influenced by past and concurrent levels of each other. A general version of the model is used in this report (Chapter 3) to examine the autoregressive and bi-directional relationships between self-perceived social support and experience of depression between Waves 1 and 2. As with the linear regression models described above, the cross-lagged panel model can be adjusted for the effects of other covariates on the dependent variable.

Mediation models

A simple unadjusted²⁴ mediation model is depicted in Figure A1. A mediation analysis aims to examine a hypothesised causal relationship or pathway between an observed independent variable/exposure/predictor and a dependent variable/outcome via the inclusion of a third variable, known as a mediator (MacKinnon et al., 2007).

Figure A1.1: Simple unadjusted mediation model

Read text description.

In this report (Chapter 5), mediation models are used to explore the role that social support (mediator) plays in the relationship between community participation (independent variable/exposure) and personal wellbeing (dependent variable/outcome). An adjusted mediation model included additional covariates to control for other factors (such as gender and employment status), as described for the regression models above.

Statistical inference

In this report we draw on different methods and approaches to inference, which allow conclusions to be drawn about the wider population of Australian men by testing hypotheses and deriving estimates.

Use of weights

Sample weights were produced for the study dataset to reduce the effect of bias in sample selection and participant non-response. When these weights are used in analysis, greater emphasis is given to population groups that are under-represented in the sample, and less to groups that are over-represented. Weighting therefore ensures that the study sample more accurately represents the population of interest, enabling the generation of prevalence estimates and the inference of results to Australian males in general.

Weights were used in analyses presented in the five empirical chapters of this report, unless stated otherwise. Cross-sectional weights were applied when using data from only one wave, whereas longitudinal weights were used in analyses involving data from two waves. Analyses were conducted using Stata® svy (survey) commands, which take the clusters and strata used in the study design into account when producing measures of the reliability of estimates. Unweighted analyses were conducted when the primary interest was in examining and describing aspects of the selected analytic sample, rather than estimating or inferring results and findings to the population level.

Statistical significance

In the context of modelling produced for this report, a finding is considered 'statistically significant' if it is unlikely that the relationship or association found in the sample data between the independent/predictor/exposure variable of interest and the dependent/outcome variable is caused by chance. In other words, it is likely that the relationship is present in the wider population of Australian men and is not just unique to the sample that has been analysed.

A test statistic known as the p-value is used to determine whether or not a finding is statistically significant. If the p-value falls below a chosen level (typically either 0.01, 0.05, or 0.1, i.e. 1%, 5% or 10%), then the finding (e.g. relationship between two variables) is regarded as statistically significant. In this report, a significance level of 0.05/5% is used unless otherwise stated. Note that if a significant result is found from modelling, this does not necessarily indicate evidence of a large effect or difference, but it does imply, fairly confidently, that there is a difference.

Confidence interval

A confidence interval (CI) is a range in which it is estimated the true population value lies. CIs of different sizes represent different levels of confidence that the true population value will lie between a set of values. The most common CIs used are 95% and 99%. Unless otherwise specified, the analyses in this report used a 95% CI.

CIs can be calculated for different types of statistics, including means, proportions and regression coefficients or estimated modelling effects. These intervals can be used to guide decisions on statistical significance in a similar way to p-values (see above); if the confidence interval includes 0, this provides some evidence that the value or relationship in question has been found by chance in the analysed sample and may not be inferred to the wider population.

CIs are used to compare statistics and characteristics of two different groups; for example, the mean income of men aged 35 or more compared to those aged less than 35. A CI would be calculated for the income measure of each group; if the intervals overlap, there is evidence that the difference in mean income of men aged 35+ vs <35 is not statistically significant. That is, there is no evidence that the mean income differs between the two groups.

Chi-square test

A chi-square (X²) test is used to investigate whether distributions of categorical variables differ from one another. For example, a chi-square test could be used to investigate whether married men are more likely to play sport than unmarried men.

Data visualisations

Data visualisation elements such as charts, graphs, figures and maps are used throughout the report to provide an accessible way to show and understand data trends and relationships. Most of the data visualisations throughout this report were generated using Stata software (StataCorp, 2019); however, some in Chapters 1 and 2 were created using Power BI (Power BI, 2021) and R package (R Core Team, 2013).

References

• Aassve, A., Billari, F. C., & Ongaro, F. (2001). The impact of income and employment status on leaving home: Evidence from the Italian ECHP sample. Labour, 15(3), 501-529.
• Abbott, J., Lim, M., Eres, R., & Long, K. (2018). The impact of loneliness on the health and wellbeing of Australians: Highlights. InPsych, 40(6).
• American Psychiatric Association (APA). (2000). Diagnostic and statistical manual of mental disorders. Text Revision - Fourth. Washington, D.C.: APA.
• Andersen, M. H., Ottesen, L., & Thing, L. F. (2019). The social and psychological health outcomes of team sport participation in adults: An integrative review of research. Scandinavian Journal of Public Health, 47(8), 832-850. doi:10.1177/1403494818791405
• Andersson, G., Denhov, A., Bülow, P., & Topor, A. (2015). Aloneness and loneliness: Persons with severe mental illness and experiences of being alone. Scandinavian Journal of Disability Research, 17(4), 353-365.
• Arbes, V., Coulton, C., & Boekel, C. (2014). Men's Social Connectedness. Melbourne, Australia: Hall & Partners Open Mind, Beyond Blue. Retrieved from www.beyondblue.org.au/docs/default-source/research-project-files/bw0276-mens-social-connectedness-final.pdf?sfvrsn=4
• Arndt, S., Turvey, C., & Andreasen, N. C. (1999). Correlating and predicting psychiatric symptom ratings: Spearman's r versus Kendall's tau correlation. Journal of Psychiatric Research, 33(2), 97-104. doi:10.1016/s0022-3956(98)90046-2
• Aslund, C., Starrin, B., & Nilsson, K. W. (2014). Psychosomatic symptoms and low psychological well-being in relation to employment status: The influence of social capital in a large cross-sectional study in Sweden. International Journal of Equity Health, 13, 22. doi:10.1186/1475-9276-13-22
• Australian Bureau of Statistics (ABS). (2019). Disability, ageing and carers, Australia: Summary of findings, 2018. ABS cat no. 4430.0. Canberra: ABS. Retrieved from www.abs.gov.au/ausstats/[email protected]/mf/4430.0
• Australian Institute of Health and Welfare (AIHW). (2017). Australia's welfare 2017: In brief. Cat. no. AUS 215. Canberra: AIHW. Retrieved from www.aihw.gov.au/getmedia/5c7b48ba-f5a2-46a6-96bd-2bbae02a5139/AIHW-AUS215-AW17_inbrief.pdf.aspx?inline=true
• Australian Institute of Health and Welfare. (2018). Physical activity across the life stages. Cat. no. PHE 225. Canberra: AIHW. Retrieved from www.aihw.gov.au/getmedia/c249ef97-e219-44df-a8bd-f5e50d04064c/aihw-phe-225.pdf.aspx?inline=true
• Australian Institute of Health and Welfare. (2019a). Australia's welfare 2019 in brief. Cat. no. AUS 227. Canberra: AIHW. Retrieved from www.aihw.gov.au/getmedia/795385cc-6493-45c9-b341-7ddf6006d518/aihw-aus-227.pdf.aspx?inline=true
• Australian Institute of Health and Welfare. (2019b). Social isolation and loneliness. Canberra: AIHW. Retrieved from www.aihw.gov.au/reports/australiaswelfare/social-isolation-and-loneliness
• Australian Institute of Health and Welfare. (2020). People with disability in Australia. Canberra: AIHW. Retrieved from www.aihw.gov.au/reports/disability/people-with-disability-in-australia/contents/summary
• Australian Psychological Society (APS). (2018). Australian loneliness report: A survey exploring the loneliness levels of Australians and the impact on their health and wellbeing. Melbourne: APS. Retrieved from psychweek.org.au/wp/wp-content/uploads/2018/11/Psychology-Week-2018-Australian-Loneliness-Report-1.pdf
• Bailey, M., Cao, R., Kuchler, T., Stroebel, J., & Wong, A. (2018). Social connectedness: Measurement, determinants, and effects. Journal of Economic Perspectives, 32(3), 259-280. Retrieved from www.ncbi.nlm.nih.gov/pubmed/30362698
• Baker, D. (2012). All the lonely people: Loneliness in Australia, 2001-2009 (Institute Paper No. 9). Canberra: The Australia Institute. Retrieved from australiainstitute.org.au/wp-content/uploads/2020/12/IP9-All-the-lonely-people_4.pdf
• Ballinger, M. L., Talbot, L. A., & Verrinder, G. K. (2009). More than a place to do woodwork: A case study of a community-based Men's Shed. Journal of Men's Health, 6(1), 20-27.
• Bandara, D., Howell, L., Silbert, M., Mohal, J., Garrard, B., & Daraganova, G. (2019). Ten to Men: The Australian Longitudinal Study on Male Health - Data User Guide, Version 3.0, September 2019. Melbourne: Australian Institute of Family Studies.
• Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health: Durkheim in the new millennium. Social Science & Medicine, 51(6), 843-857. doi:10.1016/s0277-9536(00)00065-4
• Berry, H. L., & Welsh, J. A. (2010). Social capital and health in Australia: An overview from the household, income and labour dynamics in Australia survey. Social Science & Medicine, 70(4), 588-596. doi:10.1016/j.socscimed.2009.10.012
• Beyond Blue. (2014). Men's social connectedness. Melbourne, Australia: Beyond Blue. Retrieved from www.beyondblue.org.au/docs/default-source/research-project-files/bw0276-mens-social-connectedness-final.pdf?sfvrsn=4
• Blom, H., Marschark, M., Vervloed, M. P., & Knoors, H. (2014). Finding friends online: Online activities by deaf students and their well-being. PLoS One, 9(2), e88351. doi:10.1371/journal.pone.0088351
• Brinkley, A., McDermott, H., & Munir, F. (2017). What benefits does team sport hold for the workplace? A systematic review. Journal of Sports Science, 35(2), 136-148. doi:10.1080/02640414.2016.1158852
• Brown, K. M., Hoye, R., & Nicholson, M. (2012). Self-esteem, Self-efficacy, and social connectedness as mediators of the relationship between volunteering and well-being. Journal of Social Service Research, 138, 468-463.
• Brucker, D. L. (2015). Social capital, employment and labor force participation among persons with disabilities. Journal of Vocational Rehabilitation, 43(1), 17-31.
• Bryant-Bedell, K., & Waite, R. (2010). Understanding major depressive disorder among middle-aged African American men. Journal of Advanced Nursing, 66(9), 2050-2060. doi:10.1111/j.1365-2648.2010.05345.x
• Budinski, M., Baxter, J., Carroll, M., & Hand, K. (2020). Help and support. Families in Australia Survey: Life during COVID-19, Report no. 3. Melbourne: Australian Institute of Family Studies. Retrieved from aifs.gov.au/sites/default/files/publication-documents/covid-19-survey-report_3_help_and_support_final.pdf
• Burcusa, S. L., & Iacono, W. G. (2007). Risk for recurrence in depression. Clinical Psychology Review, 27(8), 959-985.
• Carroll, M., Hand, K., Budinski, M., & Baxter, J. (2020). Staying connected when we're apart. Families in Australia Survey: Life during COVID-19, Report no. 2. Melbourne: Australian Institute of Family Studies. Retrieved from aifs.gov.au/sites/default/files/publication-documents/covid-19-survey-report_2_staying_connected.pdf
• Caton, S., & Chapman, M. (2016). The use of social media and people with intellectual disability: A systematic review and thematic analysis. Journal of Intellectual & Developmental Disability, 41, 125-139.
• Choi, N. G., Pepin, R., Marti, C. N., Stevens, C. J., & Bruce, M. L. (2020). Improving social connectedness for homebound older adults: Randomized controlled trial of tele-delivered behavioral activation versus tele-delivered friendly visits. American Journal of Geriatric Psychiatry, 28(7), 698-708. doi:10.1016/j.jagp.2020.02.008
• Clement, S., Schauman, O., Graham, T., Maggioni, F., Evans-Lacko, S., Bezborodovs, N. et al. (2015). What is the impact of mental health-related stigma on help-seeking? A systematic review of quantitative and qualitative studies. Psychological Medicine, 45(1), 11-27.
• Cohen, S. (2004). Social relationships and health. American Psychologist, 59(8), 676-684. doi:10.1037/0003-066X.59.8.676
• Commonwealth of Australia. (2011). 2010-2020 National Disability Strategy: An initiative of the Council of Australian Governments. Canberra: Commonwealth of Australia. Retrieved from www.dss.gov.au/sites/default/files/documents/05_2012/national_disability_strategy_2010_2020.pdf
• Cornwell, B., Laumann, E. O., & Schumm, L. P. (2008). The social connectedness of older adults: A national profile*. American Sociological Review, 73(2), 185-203. doi:10.1177/000312240807300201
• Cornwell, E. Y., & Waite, L. J. (2009). Social disconnectedness, perceived isolation, and health among older adults. Journal of Health and Social Behavior, 50(1), 31-48. doi:10.1177/002214650905000103
• Craig, A., Tran, Y., & Middleton, J. (2009). Psychological morbidity and spinal cord injury: A systematic review. Spinal Cord, 47(2), 108-114. doi:10.1038/sc.2008.115
• Cultural & Indigenous Research Centre Australia. (2016). Giving and volunteering in culturally and linguistically diverse and Indigenous communities: Final report. Redfern, NSW: CIRCA. Retrieved from www.communitybusinesspartnership.gov.au/wp-content/uploads/2016/09/giving_and_volunteering_in_cald_and_indigenous_communities.pdf
• Currier, D., Sahabandu, S., Davidson, W., Pirkis, J., & English, D. R. (2016). Young Australian men: Health, lifestyle and socio-environmental influences. Report prepared for the Australian Government Department of Health, June 2016. tentomen.org.au/sites/default/files/ten_to_men_major_report_2016.docx
• Daraganova, G., & Quinn B. (Eds.). (2020). Insights #1: Findings from Ten to Men - The Australian Longitudinal Study on Male Health 2013-16. Melbourne: Australian Institute of Family Studies.
• Dias, A., Geard, N., Campbell, P. T., Warr, D., & McVernon, J. (2018). Quantity or quality? Assessing relationships between perceived social connectedness and recorded encounters. PLoS One, 13(11), e0208083. doi:10.1371/journal.pone.0208083
• Ding, N., Berry, H. L., & O'Brien, L. V. (2015). One-year reciprocal relationship between community participation and mental wellbeing in Australia: A panel analysis. Social Science & Medicine, 128, 246-254.
• Dobransky, K., & Hargittai, E. (2006). The disability divide in internet access and use. Information, Communication & Society, 9, 313-334.
• Eime, R., Harvey, J., & Charity, M. (2016). Age profiles of sport participants - Victoria 2015. Federation University, Victoria University.
• Eime, R. M., Young, J. A., Harvey, J. T., Charity, M. J., & Payne, W. R. (2013). A systematic review of the psychological and social benefits of participation in sport for adults: informing development of a conceptual model of health through sport. International Journal of Behavorial Nutrition and Physical Activity, 10, 135. doi:10.1186/1479-5868-10-135
• Emerson, E., Fortune, N., Llewellyn, G., & Stancliffe, R. (2021). Loneliness, social support, social isolation and wellbeing among working age adults with and without disability: Cross-sectional study. Disability and Health Journal, 14(1), 100965. doi:10.1016/j.dhjo.2020.100965
• Emerson, E., & Llewellyn, G. (2021). Identifying children at risk of intellectual disability in UNICEF's multiple indicator cluster surveys: Cross-sectional survey. Disability and Health Journal, 14(1), 100986. doi:10.1016/j.dhjo.2020.100986
• Fiori, K. L., & Denckla, C. A. (2012). Social support and mental health in middle-aged men and women: a multidimensional approach. Journal of Aging Health, 24(3), 407-438. doi:10.1177/0898264311425087
• Fong, P., Cruwys, T., Robinson, S. L., Haslam, S. A., Haslam, C., Mance, P. L., & Fisher, C. L. (2021). Evidence that loneliness can be reduced by a whole-of-community intervention to increase neighbourhood identification. Social Science & Medicine, 277, 113909. doi:10.1016/j.socscimed.2021.113909
• Friedman, C., & Rizzolo, M. (2018). Friendship, quality of life, and people with intellectual and developmental disabilities. Journal of Developmental and Physical Disabilities, 30(39-54).
• Fuhrer, R., & Stansfeld, S. A. (2002). How gender affects patterns of social relations and their impact on health: a comparison of one or multiple sources of support from "close persons". Social Science & Medicine, 54(5), 811-825. doi.org/10.1016/S0277-9536(01)00111-3
• Fulloon, S. (2020, 11 March). Anthony has Down syndrome and a job, but others with disability are often overlooked. SBS News (online). Retrieved from www.sbs.com.au/news/anthony-has-down-syndrome-and-a-job-but-others-with-disability-are-often-overlooked?cid=newsapp:socialshare:other
• Gard, M., Dionigi, R. A., Horton, S., Baker, J., Weir, P., & Dionigi, C. (2017). The normalization of sport for older people? Annals of Leisure Research, 20(3), 253-272.
• Gayman, A. M., Fraser-Thomas, J., Dionigi, R. A., Horton, S., & Baker, J. (2017). Is sport good for older adults? A systematic review of psychosocial outcomes of older adults' sport participation. International Review of Sport and Exercise Psychology, 10(1), 164-185.
• Grav, S., Hellzen, O., Romild, U., & Stordal, E. (2012). Association between social support and depression in the general population: the HUNT study, a cross-sectional survey. Journal of Clinical Nursing, 21(1-2), 111-120. doi:10.1111/j.1365-2702.2011.03868.x
• Gray, A. (2009). The social capital of older people. Ageing and Society, 29(1), 5-31.
• Groce, N. E. (2018). Global disability: An emerging issue. The Lancet Global Health, 6(7), e724-e725. doi:10.1016/S2214-109X(18)30265-1
• Hakulinen, C., Pulkki-Råback, L., Jokela, M., Ferrie, J. E., Aalto, A.-M., Virtanen, M. et al. (2016). Structural and functional aspects of social support as predictors of mental and physical health trajectories: Whitehall II cohort study. Journal of Epidemiology and Community Health, 70(7), 710-715.
• Hames, J. L., Hagan, C. R., & Joiner, T. E. (2013). Interpersonal processes in depression. Annual Review of Clinical Psychology, 9, 355-377.
• Hastings, O. P. (2016). Not a lonely crowd? Social connectedness, religious service attendance, and the spiritual but not religious. Social Science Research, 57, 63-79. doi:10.1016/j.ssresearch.2016.01.006
• Hatch, S. L., & Dohrenwend, B. P. (2007). Distribution of traumatic and other stressful life events by race/ethnicity, gender, SES and age: A review of the research. American Journal of Community Psychology, 40(3-4), 313-332.
• Haugen, T., Säfvenbom, R., & Ommundsen, Y. (2013). Sport participation and loneliness in adolescents: The mediating role of perceived social competence. Current Psychology, 32(2), 203-216.
• Hawkley, L. C., & Capitanio, J. P. (2015). Perceived social isolation, evolutionary fitness and health outcomes: A lifespan approach. Philosophical Transactions of the Royal Society of London B Biological Sciences, 370(1669). doi:10.1098/rstb.2014.0114
• Heinrich, L. M., & Gullone, E. (2006). The clinical significance of loneliness: a literature review. Clinical Psychology Review, 26(6), 695-718. doi:10.1016/j.cpr.2006.04.002
• Herreen, D., Rice, S., Currier, D., Schlichthorst, M., & Zajac, I. (2021). Associations between conformity to masculine norms and depression: age effects from a population study of Australian men. BMC Psychology, 9(1), 32. doi:10.1186/s40359-021-00533-6
• Holt-Lunstad, J. (2018). Why social relationships are important for physical health: A systems approach to understanding and modifying risk and protection. Annual Review of Psychology, 69, 437-458. doi:10.1146/annurev-psych-122216-011902
• Hoye, R., Nicholson, M., & Brown, K. (2015). Involvement in sport and social connectedness. International Review for the Sociology of Sport, 50(1), 3-21.
• Ingram, R. E., & Luxton, D. D. (2005). Vulnerability-stress models. In B. L. Hankin & J. R. Z. Abela (Eds.), Development of psychopathology: A vulnerability-stress perspective (pp. 32-46). doi:10.4135/9781452231655
• International Wellbeing Group. (2013). Personal wellbeing index, 5th edition. Melbourne: Australian Centre on Quality of Life, Deakin University. www.acqol.com.au/instruments#measures
• Kelly, D., Steiner, A., Mason, H., & Teasdale, S. (2021). Men's Sheds as an alternative healthcare route? A qualitative study of the impact of Men's Sheds on user's health improvement behaviours. BMC Public Health, 21(1), 553. doi:10.1186/s12889-021-10585-3
• Kim, H. C., & Zhu, Z. Y. (2020). Improving social inclusion for people with physical disabilities: The roles of mobile social networking applications (MSNA) by disability support organizations in China. International Journal of Environmental Research and Public Health, 17(7). doi:10.3390/ijerph17072333
• Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606-613.
• Leopold, T. (2012). The legacy of leaving home: Long-term effects of coresidence on parent-child relationships. Journal of Marriage and Family, 74(3), 399-412. doi: doi.org/10.1111/j.1741-3737.2012.00964.x
• Liebler, C. A., & Sandefur, G. D. (2002). Gender differences in the exchange of social support with friends, neighbors, and co-workers at midlife. Social Science Research, 31(3), 364-391. doi:10.1016/S0049-089X(02)00006-6
• Lim, M. H., Rodebaugh, T. L., Zyphur, M. J., & Gleeson, J. F. (2016). Loneliness over time: The crucial role of social anxiety. Journal of Abnormal Psychology, 125(5), 620-630. doi:10.1037/abn0000162
• Llewellyn, G., Emerson, E., & Honey, A. (2013). Left behind: 2013 Monitoring the social inclusion of young Australians with self-reported long term health conditions, impairments or disabilities 2001-2011. Sydney: Centre for Disability Research and Policy Faculty of Health Sciences, University of Sydney. Retrieved from ses.library.usyd.edu.au/bitstream/handle/2123/9026/Left_Behind_2011.pdf?sequence=2&isAllowed=y
• Machmutow, K., Grosse Holtforth, M., Krieger, T., & Watzke, B. (2018). Identifying relapse prevention elements during psychological treatment of depression: Development of an observer-based rating instrument. Journal of Affective Disorders, 227, 358-365.
• MacKinnon, D. P., Fairchild, A. J., & Fritz, M. S. (2007). Mediation analysis. Annual Review of Psychology, 58, 593-614.
• Madans, J. H., Loeb, M. E., & Altman, B. M. (2011). Measuring disability and monitoring the UN Convention on the Rights of Persons with Disabilities: The work of the Washington Group on disability statistics. BMC Public Health, 11 Suppl 4, S4. doi:10.1186/1471-2458-11-S4-S4
• Mahalik, J. R., Locke, B. D., Ludlow, L. H., Diemer, M. A., Scott, R. P., Gottfried, M., & Freitas, G. (2003). Development of the conformity to masculine norms inventory. Psychology of Men & Masculinity, 4(1), 3.
• Malinauskas, R., & Malinauskiene, V. (2018). The mediation effect of perceived social support and perceived stress on the relationship between emotional intelligence and psychological wellbeing in male athletes. Journal of Human Kinetics, 65, 291-303. doi:10.2478/hukin-2018-0017
• Maulik, P. K., Eaton, W. W., & Bradshaw, C. P. (2010). The effect of social networks and social support on common mental disorders following specific life events. Acta Psychiatrica Scandinavica, 122(2), 118-128. doi.org/10.1111/j.1600-0447.2009.1511.x
• McKenzie, S. K., Collings, S., Jenkin, G., & River, J. (2018). Masculinity, social connectedness, and mental health: Men's diverse patterns of practice. American Journal of Men's Health, 12(5), 1247-1261. doi:10.1177/1557988318772732
• McPhedran, S. (2010). Regional living and community participation: Are people with disability at a disadvantage? Australian Social Policy, 9, 111-135.
• Mellor, D., Hayashi, Y., Firth, L., Stokes, M., Chambers, S., & Cummins, R. (2008). Volunteering and well-being: Do self-esteem, optimism, and perceived control mediate the relationship? Journal of Social Service Research, 34(4), 61-70.
• Milner, A., Krnjacki, L., Butterworth, P., & LaMontagne, A. D. (2016). The role of social support in protecting mental health when employed and unemployed: A longitudinal fixed-effects analysis using 12 annual waves of the HILDA cohort. Social Science & Medicine, 153, 20-26. doi:10.1016/j.socscimed.2016.01.050
• Mithen, J., Aitken, Z., Ziersch, A., & Kavanagh, A. M. (2015). Inequalities in social capital and health between people with and without disabilities. Social Science & Medicine, 126, 26-35. doi:10.1016/j.socscimed.2014.12.009
• Moloney, L., Weston, R., Qu, L., & Hayes, A. (2012). Families, life events and family service delivery: A literature review (Research Report No. 20). Melbourne: Australian Institute of Family Studies. Retrieved from aifs.gov.au/publications/families-life-events-and-family-service-delivery
• Moylan, M. M., Carey, L. B., Blackburn, R., Hayes, R., & Robinson, P. (2015). The Men's Shed: Providing biopsychosocial and spiritual support. Journal of Religion & Health, 54(1), 221-234. doi:10.1007/s10943-013-9804-0
• Mushtaq, R., Shoib, S., Shah, T., & Mushtaq, S. (2014). Relationship between loneliness, psychiatric disorders and physical health? A review on the psychological aspects of loneliness. Journal of Clinical Diagnostic Research, 8(9), WE01-04. doi:10.7860/JCDR/2014/10077.4828
• National Disabilities and Carer Council. (2009). Shut out: The experience of people with disabilities and their families in Australia. National Disability Strategy Consultation Report. Canberra: Australian Government. Retrieved from www.dss.gov.au/sites/default/files/documents/05_2012/nds_report.pdf
• Nurius, P. S., Uehara, E., & Zatzick, D. F. (2013). Intersection of stress, social disadvantage, and life course processes: Reframing trauma and mental health. American Journal of Psychiatric Rehabilitation, 16(2), 91-114.
• Oosterhoff, B., Kaplow, J. B., Wray-Lake, L., & Gallagher, K. (2017). Activity-specific pathways among duration of organized activity involvement, social support, and adolescent well-being: Findings from a nationally representative sample. Journal of Adolescence, 60, 83-93. doi:10.1016/j.adolescence.2017.07.012
• Ozbay, F., Johnson, D. ioC., Dimoulas, E., Morgan, C. A., Charney, D., & Southwick, S. (2007). Social support and resilience to stress: From neurobiology to clinical practice. Psychiatry (Edgmont (Pa.: Township)), 4(5), 35-40. Retrieved from pubmed.ncbi.nlm.nih.gov/20806028
• Palmer, M., & Harley, D. (2012). Models and measurement in disability: An international review. Health Policy Plan, 27(5), 357-364. doi:10.1093/heapol/czr047
• Pirkis, J., Currier, D., Carlin, J., Degenhardt, L., Dharmage, S. C., Giles-Corti, B. et al. (2017). Cohort Profile: Ten to Men (the Australian Longitudinal Study on Male Health). International Journal of Epidemiology, 46(3), 793-794i. doi:10.1093/ije/dyw055
• Power BI. (2021). Overview and Learning. Support.office.com [online] Retrieved frompowerbi.microsoft.com/en-us/learning/.
• R Core Team. (2013). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Retrieved from www.R-project.org
• Ravitz, P., Maunder, R., Hunter, J., Sthankiya, B., & Lancee, W. (2010). Adult attachment measures: A 25-year review. Journal of Psychosomatic Research, 69(4), 419-432. doi:10.1016/j.jpsychores.2009.08.006
• Reddin, J. A., & Sonn, C. C. (2003). Masculinity, social support and sense of community: The Men's Group experience in Western Australia. The Journal of Men's Studies, 11(2), 207-223. doi.org/10.3149/jms.1102.207
• Rehm, J., & Shield, K. D. (2019). Global burden of disease and the impact of mental and addictive disorders. Current Psychiatry Reports, 21(2), 10.
• Relationships Australia. (2017). January 2017: Loneliness. Retrieved from www.relationships.org.au/what-we-do/research/online-survey/january-2017-loneliness
• Relationships Australia. (2018). Is Australia experiencing an epidemic of loneliness? Findings from 16 waves of the Household Income and Labour Dynamics of Australia Survey. Canberra: Relationships Australia. Retrieved from www.relationships.org.au/what-we-do/research/an-epidemic-of-loneliness-2001-2017
• Ren, P., Qin, X., Zhang, Y., & Zhang, R. (2018). Is social support a cause or consequence of depression? A longitudinal study of adolescents. Frontiers in Psychology, 9, 1634.
• Riches, V. C., Parmenter, T. R., Wiese, M., & Stancliffe, R. J. (2006). Intellectual disability and mental illness in the NSW criminal justice system. International Journal of Law and Psychiatry, 29(5), 386-396. doi:10.1016/j.ijlp.2005.10.003
• Rimmer, J. H., & Marques, A. C. (2012). Physical activity for people with disabilities. Lancet, 380(9838), 193-195. doi:10.1016/S0140-6736(12)61028-9
• Ruhindwa, A., Randall, C., & Cartmel, J. (2016). Exploring the challenges experienced by people with disabilities in the employment sector in Australia: Advocating for inclusive practice. A review of literature. Journal of Social Inclusion, 7(1), 4-19.
• Saeri, A. K., Cruwys, T., Barlow, F. K., Stronge, S., & Sibley, C. G. (2018). Social connectedness improves public mental health: Investigating bidirectional relationships in the New Zealand attitudes and values survey. Australian and New Zealand Journal of Psychiatry, 52(4), 365-374. doi:10.1177/0004867417723990
• Santini, Z. I., Jose, P. E., Cornwell, E. Y., Koyanagi, A., Nielsen, L., Hinrichsen, C. et al. (2020). Social disconnectedness, perceived isolation, and symptoms of depression and anxiety among older Americans (NSHAP): a longitudinal mediation analysis. The Lancet Public Health, 5(1), e62-e70.
• Schulz, R., & Tompkins, C. A. (1990). Life events and changes in social relationships: Examples, mechanisms, and measurement. Journal of Social and Clinical Psychology, 9(1), 69-77. doi:10.1521/jscp.1990.9.1.69
• Segrin, C. (2001). Interpersonal processes in psychological problems. Guilford Press.
• Sherbourne, C. D., & Stewart, A. L. (1991). The MOS social support survey. Social Science & Medicine, 32(6), 705-714. doi:10.1016/0277-9536(91)90150-b
• Shields, M. A., & Wheatley Price, S. (2005). Exploring the economic and social determinants of psychological well-being and perceived social support in England. Journal of the Royal Statistical Society (Series A [Statistics in Society]), 168(3), 513-537.
• Sixsmith, J., & Boneham, M. (2003). Older men's participation in community life: Notions of social capital, health, and empowerment. Ageing International, 28(4), 372-388.
• Smith, J. A., Watkins, D. C., & Griffit, D. M. (2020). Equity, gender and health: New directions for global men's health promotion. Health Promotion Journal of Australia, 31(2), 161-165.
• Sport Participation Research Project. (2019). Sport Participation Rates: Aggregation of 12 sports, Victoria 2017. Melbourne: VicHealth, Sport and Recreation Victoria, and Victoria University and Federation University Australia. Retrieved from www.sportandrecreationspatial.com.au/resources/SPRP%20integrated%20de-identified%202017%20300519%20RE.pdf
• Stansfeld, S. (2006). Social support and social cohesion. In M. Marmot & R. Wilkinson (Eds.), Social determinants of health (pp. 148-171). Oxford, UK: Oxford University Press.
• StataCorp. (2019). Stata Statistical Software: Release 16. College Station, TX: StataCorp LLC.
• Stensrud, S., Risberg, M. A., & Roos, E. M. (2015). Effect of exercise therapy compared with arthroscopic surgery on knee muscle strength and functional performance in middle-aged patients with degenerative meniscus tears: A 3-mo follow-up of a randomized controlled trial. American Journal of Physical Medicine & Rehabilitation, 94(6), 460-473. doi:10.1097/PHM.0000000000000209
• Stokes, H., Turnbull, M., & Wyn, J. (2013). Young people with a disability: Independence and opportunity: A literature review (Research Report 39). Melbourne: Youth Research Centre, Melbourne Graduate School of Education, University of Melbourne. Retrieved from education.unimelb.edu.au/yrc/assets/docs/yrc-publication-docs/research_report39.pdf
• Struik, L. L., Abramowicz, A., Riley, B., Oliffe, J. L., Bottorff, J. L., & Stockton, L. D. (2019). Evaluating a tool to support the integration of gender in programs to promote men's health. American Journal of Men's Health, 13(6), 1557988319883775. doi:10.1177/1557988319883775
• Taylor, J., Cole, R., Kynn, M., & Lowe, J. (2018). Home away from home: Health and wellbeing benefits of men's sheds. Health Promotion Journal of Australia, 29(3), 236-242. doi:10.1002/hpja.15
• Terhaag, S., Quinn, B., Swami, N., & Daraganova, G. (2020). Mental Health of Australian Males. In G. Daraganova & B. Quinn (Eds.), Insights #1: Findings from Ten to Men - The Australian Longitudinal Study on Male Health 2013-16. Melbourne: Australian Institute of Family Studies.
• Tough, H., Siegrist, J., & Fekete, C. (2017). Social relationships, mental health and wellbeing in physical disability: A systematic review. BMC Public Health, 17(1), 414. doi:10.1186/s12889-017-4308-6
• Trimble, L. B., & Kmec, J. A. (2011). The Role of Social Networks in Getting a Job. Sociology Compass, 5(2), 165-178.
• Umberson, D., & Montez, J. K. (2010). Social relationships and health: A flashpoint for health policy. Journal of Health and Social Behavior, 51 Suppl., S54-66. doi:10.1177/0022146510383501
• Verdonschot, M. M., de Witte, L. P., Reichrath, E., Buntinx, W. H., & Curfs, L. M. (2009a). Community participation of people with an intellectual disability: A review of empirical findings. Journal of Intellectual Disability Research, 53(4), 303-318. doi:10.1111/j.1365-2788.2008.01144.x
• Verdonschot, M. M., de Witte, L. P., Reichrath, E., Buntinx, W. H., & Curfs, L. M. (2009b). Impact of environmental factors on community participation of persons with an intellectual disability: A systematic review. Journal of Intellectual Disability Research, 53(1), 54-64. doi:10.1111/j.1365-2788.2008.01128.x
• Victorian Health Promotion Foundation. (2021). Sport participation in Victoria, 2019: Research summary. Melbourne: VicHealth. Retrieved from www.vichealth.vic.gov.au/-/media/ResourceCentre/PublicationsandResources/Sport-Participation-Research-Summary-2019.pdf?la=en&hash=21D72C0EA15ABFB4F444B23EB95F374442F7EA31
• Vornholt, K., Villotti, P., Muschalla, B., Bauer, J., Colella, A., Zijlstra, F. et al. (2018). Disability and employment: Overview and highlights. European Journal of Work and Organizational Psychology, 27(1), 40-55.
• Wang, J., Mann, F., Lloyd-Evans, B., Ma, R., & Johnson, S. (2018). Associations between loneliness and perceived social support and outcomes of mental health problems: A systematic review. BMC Psychiatry, 18(1), 156.
• Whiteman, S. D., McHale, S. M., & Crouter, A. C. (2011). Family relationships from adolescence to early adulthood: Changes in the family system following firstborns' leaving home. Journal of Research on Adolescence: The Official Journal of the Society for Research on Adolescence, 21(2), 461-474. doi:10.1111/j.1532-7795.2010.00683.x
• Wong, E., Backholer, K., Gearon, E., Harding, J., Freak-Poli, R., Stevenson, C., & Peeters, A. (2013). Diabetes and risk of physical disability in adults: a systematic review and meta-analysis. Lancet Diabetes Endocrinology, 1(2), 106-114. doi:10.1016/S2213-8587(13)70046-9
• Wong, Y. J., Ho, M. R., Wang, S. Y., & Miller, I. S. (2017). Meta-analyses of the relationship between conformity to masculine norms and mental health-related outcomes. Journal of Counseling Psychology, 64(1), 80-93. doi:10.1037/cou0000176
• World Health Organization (WHO). (2017). Depression and other common mental disorders: Global health estimates. Geneva: WHO. Retrieved from apps.who.int/iris/bitstream/handle/10665/254610/WHO-MSD-MER-2017.2-eng.pdf
• World Health Organization, & World Bank. (2011). World report on disability. Geneva, Switzerland: WHO. Retrieved from www.who.int/disabilities/world_report/2011/report.pdf
• Xia, N., & Li, H. (2018). Loneliness, Social isolation, and cardiovascular health. Antioxidiants and Redox Signaling, 28(9), 837-851. doi:10.1089/ars.2017.7312
• Yen, I. H., Shim, J. K., Martinez, A. D., & Barker, J. C. (2012). Older people and social connectedness: How place and activities keep people engaged. Journal of Aging Research, 2012, 139523. doi:10.1155/2012/139523
• Yeung, J. W. K., Zhang, Z., & Yeun Kim, T. (2018). Volunteering and health benefits in general adults: Cumulative effects and forms. BMC Public Health, 18(8). Retrieved from bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-017-4561-8
• Zyphur, M. J., Allison, P. D., Tay, L., Voelkle, M. C., Preacher, K. J., Zhang, Z. et al. (2020). From data to causes I: Building a general cross-lagged panel model (GCLM). Organizational Research Methods, 23(4), 651-687.