Ageing is generally celebrated because of its indication of success at dealing with health and life challenges that previously resulted in early deaths. However, this advancement has been met with other significant challenges including reduced physical functioning (Bowling and Dieppe 2005; Tomás et al. 2016) and the experience of many losses including friends passing away or children moving away. Despite the difficulties associated with growing old, the elderly are regarded as a resource to families and communities (WHO 2015). Just like any other stage of life, relationships are at the heart of managing challenges presenting in old age (Siedlecki et al. 2014) and are considered important for maintaining or enhancing psychological well-being. Research on relationships has highlighted the essence of quality social interactions for well-being (Chipps and Jarvis 2016; Jarvis et al. 2015; Webster et al. 2015), the need for a sense of belonging (Baumeister and Leary 1995) and interconnectedness (Jarvis et al. 2015; Provencher et al. 2014), as well as the provision and experience of support (Siedlecki et al. 2014) across a wide variety of networks ranging from family to external relational ties. Current evidence among the elderly has revealed similar patterns in the importance of various kinds of relationships (c.f., Nguyen and Seal 2014; Taylor et al. 2013; Taylor et al. 2017).

Context may influence the patterns and dynamics of important relationships. For example, urban and rural settings may be linked to differences in psychosocial and relational well-being and these differences may be evident in people from areas with various degrees of urbanisation in the North West Province of South Africa (Vorster et al. 2000). Although there were recent studies on relational patterns among the elderly in South Africa (e.g., Roos 2016), previous research has not adequately accounted for possible existing differences that could emerge as a result of urbanisation and the diffusion and shrinkage of social ties in urban and rural social contexts, and whether the kinds of important relationships and reasons for the importance thereof in the various contexts dovetail or differ. The findings of this study will add to our knowledge of relational functioning among older individuals in rural and urban contexts and provide necessary information on how relational interactions can be effectively transformed into interventions targeted at enhancing well-being of older adults. Policies affecting the well-being of the elderly could be improved with knowledge gleaned from exploring why certain relationships are important.

Relational Well-Being

At the heart of relationship models is the need to understand why positive social relationships are so crucial to the individual and the mechanisms underlying the initiation and maintenance of such relationships. According to Aron and Aron (1996), the desire for relationships arises from the need to extend oneself by including another in the self. Relationships are also deemed to be at the heart of the experience of psychological well-being. For instance, Ryff and Singer (1998) indicated that the need for deep meaningful relationships underscored the experience of positive mental health. In addition, Ryff and Keyes (1995) in their conceptualisation of psychological well-being noted that positive relations with others was a key indicator of well-being. There are also culturally motivated and indigenous perspectives on relational well-being that have been identified. Indigenous world views among Hawaiians regard relational well-being as encompassing all dimensions of human ecology including family, ancestors, neighbourhood, and the community at large with these dimensions being described as interdependent (McGregor et al. 2003). Furthermore, McCubbin et al. (2013) defined relational well-being as the experience of financial stability, the ability to care for others, respect for others, valuing ancestral language, health promoting practices, the appreciation of culture and trust and predictability. Although providing general information on the importance of relationships, these studies are limited in explaining age-related unique experiences of relational interactions amidst social changes and urbanisation.

Relational Well-Being among the Aged

Following theoretical propositions on the importance of relational well-being, empirical research has shown that the ageing process could be associated with a loss of cognitive and physical functions and a reduction in social networks (Chen and Feeley 2014) at times due to poor social connections and infrequent social activities (Holtzman et al. 2004; Ramlagan et al. 2013). Relational well-being is crucial at this stage as older individuals require constant positive interactions as a form of support and a means to satisfy their emotional needs (Shiovitz-Ezra and Leitsch 2010).

Different relational networks have been identified among the aged and these include spouses, children, friends (Routasalo et al. 2006) and the broader community (Melchiorre et al. 2013). Moreover, previous research has shown that several types of relationships have implications for the experience of relational well-being among older adults. For instance, relational experiences among the elderly that occur as a result of interactions with family and friends have generally been found to reduce loneliness (Routasalo et al. 2006) and increase positive affect (Li et al. 2014). Such interactions have been described as intimate and regular because of their value in providing support. Also, while support from spouses improved relational well-being through the reduction of loneliness, those from children and other family members did not (Chen and Feeley 2014). Research among the elderly in China demonstrated that, in comparison to support from friends that only enhanced the experience of positive affect, social interactions with spouses and children simultaneously increased positive affect while also decreasing negative affect (Li et al. 2014). This was ascribed to the fact that the exchange and frequency of support from family were higher as compared to friends. In line with the tax specificity model (Penning 1990), interactions with friends seemed to foster more positive affect because these are orientated towards leisure activities and mutual interests. Melchiorre et al. (2013) corroborated the finding on interactions in the family, indicating that older adults across Europe who lived in large households benefited and experienced a greater amount of support. Although there exists evidence on the utility of various kinds of relationships, there is still a lacuna in research addressing the question of reasons underlying the importance of relationships among older people as viewed from a bottom-up perspective.

Social Contexts and Relational Patterns

The social context of relationships, whether urban or rural, could result in differential relational experiences with peers and other generations (Nelson and Prilleltensky 2010). Moreover, while some research has supported the notion of ageing being associated with loneliness (Chen and Feeley 2014), others have found that there are a wide variety of networks at the disposal of the aged (WHO 2015). Provencher et al. (2014) also argued that it is important to understand how older individuals form a sense of belonging in different social contexts amidst changes that facilitate exclusion. To fill this gap, the present study seeks to unearth the motivations that underlie important relationships among older individuals living in rural and urban contexts. The contradictory findings regarding older people’s access to social networks and social changes across rural and urban contexts require a revisit to existing relational patterns and experiences across these two settings in order to further our understanding of relational well-being in different contexts.

In South Africa, rapid social change is transforming family relations, increasing intergenerational tensions and adaptations of traditional family structures (Campbell 1994). Older adults in Africa living in rural settings tend to experience that they receive very little attention from policy makers (Beales 2000) and rather tend to feel excluded and invisible (Falconer et al. 2002; UNISDR 2003). This is alarming as it was shown that people in rural areas had lower levels of physical, physiological and psychosocial health than those in urban areas (Vorster et al. 2000). Moreover, in most rural settings in South Africa, individuals depend on agriculture as a source of livelihood (Roos et al. 2013). This form of livelihood is at times challenged by agricultural drought (Roos et al. 2013), changing family structures due to the consequences of HIV/AIDS and urban migration (Schatz 2007). One of the means of dealing with these challenges is relational coping (Roos et al. 2013), which includes pro-active collective planning, innovative attempts by women to secure jobs, sharing of resources and interactions with divine realities. In congruence with the theme of relational coping, Chilisa (2011) noted that in rural African contexts, older people interacted with the environment, divinities, the living and non-living individuals (Chilisa 2011) for several reasons. For instance, while relationships with the environment, divinities and non-living individuals determined appropriate behaviour and a purpose for life, intergenerational interactions with family and the community took the form of reciprocal support (Bohman et al. 2009; Chilisa 2011).

Older adults living in urban communities are faced with other forms of challenges and perhaps opportunities that might have implications for their relational experiences. Due to urbanisation, older people are increasingly losing their place in rural settings and there is immense social exclusion being experienced in urban settings (Walsh et al. 2014). The experience of social exclusion was described as being manifested in little concern for another, fewer opportunities for interaction, financial insecurity and changing compositions of localities with the moving away of friends and family. Also noteworthy is the “loss of togetherness” and feelings of insecurity as a result of crime (Buffel et al. 2013). Other challenges faced by older adults in urban areas include barriers to health care as a result of the absence of an escort or high cost of transport to healthcare providers and private sector fees for medicine and treatment (Aboderin and Beard 2015). In urban settings, intergenerational support, which was clearly guided by norms of reciprocal care and interaction across generations, seemed to be replaced by family values which support individuality (Roos 2016). In the South African context, challenges of urbanisation and poverty (Aboderin and Hoffman 2015) seemed to have influenced the relational ties and experiences across and within generations. Roos (2016) argued that these challenges could weaken support and fragment the tissue of the community, but there is still a need for further research to understand relationships among the aged in South Africa.

Research Aim

The primary aim of the present study was to describe the reasons for important relationships in rural (Ganyesa and Tlakgameng) and urban (Ikageng) South African contexts as described by older individuals. Against the backdrop of possible changes in relational patterns across both rural and urban contexts, a secondary aim of the study was to compare the reasons for important relationships across the urban and rural samples.

Method

A basic descriptive qualitative research design was chosen because the aim of the study was to use semi-structured interviews to explore important relationships and the reasons thereof among older individuals living in rural and urban communities in South Africa (Creswell and Plano Clark 2011; Creswell and Poth 2013). The need to understand a phenomenon from laypersons perspectives underlies most qualitative research designs.

Context of the Study

The rural context for this study involved two proximate villages/areas in the North West Province of South Africa, namely Ganyesa, falling under the Bophirima district municipality, and the deep rural Tlakgameng, which lies about 35 km North-East from Ganyesa. These mostly Setswana-speaking communities face challenges such as unemployment and marginal growth in trade and agriculture (Bophirima District Municipality, South Africa 2006; Holborn 2011), an increase in infection rate of HIV/AIDS (Ndebele 2007) and a high dependency rate (Francis 2002). Based on the latest statistics available for Statistics South Africa (2011), 19.1% of the population of adults above the age of 20 had some secondary education, 7.9% had tertiary education and 20% had no education at all. Only 20% of the population had an average total household income ranging from 9000 to 19,000 ZAR, with the rest of the population earning a lower income (Statistics South Africa 2011), which implies severe hardship. Research by Roos et al. (2013) in this community showed that older individuals capitalised on their social relations in order to aid in coping with the challenges faced. In 2011, 5.4% of the Ganyesa population were 65 years old or older (Statistics South Africa 2011).

The Ikageng community (urban context of the study), is a town in the Dr. Kenneth Kaunda District of the North West Province with Setswana as the most commonly spoken language. 23% of the population were reported to have a total household income ranging from 19,000 to 38,000 ZAR in 2011 (Statistics South Africa 2011), with the rest of the population earning a lower income. Of the population of adults that were above 20 years of age, 39% of the population were reported to have some secondary school education, 3.9% had a higher education and 8.7% were reported to have no formal education. The elderly (65 and above) comprised 4.5% of the population (Statistics South Africa 2011).

Sample and Participants

Data for this study were gathered in 2015 as part of the 10-year follow-up of the overlap between the South African leg of the Prospective Urban Rural Epidemiology study (PURE-SA; Teo et al. 2009) and the FORT 3 project (Wissing 2008/2012). Participant recruitment started by the administration of a household survey in systematically identified households from each community. From the data obtained in this survey, potential participants were identified based on adherence to the inclusion criteria, namely that participants had to be at least 35 years old, should not have had any reported lifestyle disease, tuberculosis or be HIV positive, and should not have used any chronic medication. Using this procedure, N = 2010 (rural n = 1006; urban n = 1004) participants were included in the study at baseline in 2005. All of these participants were tracked over the course of the next 10 years and in 2015, N = 924 participants were included based on the fact that they were still available and willing to participate in the follow-up study. Out of the total sample in 2015, a subsample of 50 older people was selected in each of the rural (n = 16 males, n = 34 females) and urban (n = 14 males, n = 36 females) area for inclusion in this specific study. Mean ages for the selected sample were 67 years (SD = 5.31) and 70 years (SD = 8.35) for the rural and urban groups, respectively. For the rural sample, ages ranged from 60 years to 81 years, while the urban sample ranged from 60 years to 103 years.

Data Collection

Apart from several physiological measures, the 2015 follow-up study included a battery of questionnaires investigating various aspects of health such as medical history and lifestyle behaviours. The questionnaires included, amongst others, a demographic questionnaire and open-ended questions where participants were asked to indicate the three most important relationships in their lives and the reasons why each of these relationships were regarded as such. Three community members from the rural and urban areas, respectively, who were fluent in both English and Setswana, were trained as field workers in order to administer the translated questionnaires in interview-format in Setswana. Under the supervision of the principal investigator of the FORT 3 project, responses to the open-ended questions were translated from Setswana to English by research assistants who were fluent in both Setswana and English, and translations were checked by another research assistant.

Measures

A demographic questionnaire and a semi-structured open-ended question were relevant for the present study.

Demographic Questionnaire

This questionnaire obtained demographic characteristics of the participant, including the participant’s gender, age, and highest level of education.

Semi-Structured Open-Ended Question

As part of a battery of questionnaires evaluating and exploring various aspects of health and psychosocial well-being, the present study focused on the two questions: “Please list three kinds of important relationships in your present life” and “For each of them, please specify why it is important, and how this importance is manifested (try to be as specific as possible)”. We summarised the important relationships and analysed the reasons for these relationships. A similar methodology of semi-structured open-ended questions on ‘what’ and ‘why’ of a phenomenon was implemented by Delle Fave et al. (2016) in their qualitative study.

Below, we present an integrated view of our findings combining the relationships mentioned by participants and the reasons for the importance thereof.

Ethical Considerations

Ethics approval for both projects (FORT 3 and PURE-SA) were obtained from the Health Research Ethics Committee (HREC) of the Faculty of Health Sciences of the North-West University, South Africa, with ethics approval numbers of NWU-00002-07-A2 and NWU-00016-10-A1, respectively. On obtaining written informed consent, participants were informed that participation was voluntary and that they could withdraw at any point in time during the study. Participants who were not literate and could not provide written consent were given the opportunity to do so using their thumb prints. Confidentiality was assured by stressing the importance of confidentiality in field worker training and requiring field workers to sign a confidentiality agreement. During data capturing codes instead of identifying information were used to identify participants and this anonymised dataset was used for data analysis. Any researcher who gained access to the data also signed a confidentiality agreement form.

Data Analysis

The qualitative responses from urban and rural samples were analysed using the method of thematic analysis. A computer assisted analysis software (CAQDAS) known as Atlas-ti was used in organising and analysing the data. We followed the steps for carrying out thematic analysis outlined in Braun and Clarke (2006) and Clarke and Braun (2017) (see Fig. 1 for a schematic presentation of the data analysis process). We used an inductive approach of data coding, which was aimed at understanding what relational experiences across urban and rural samples looked like and how it differed or were similar. The first step of the analysis involved familiarising ourselves with the important relationships that were indicated and the reasons thereof by reading and re-reading the data. This was then followed by generating features of the data that provide a succinct description of ideas that had been conveyed. We generated several data-driven codes representing the kinds of relationships and others on the reasons for these relationships. We further analysed these codes with the aim to search for themes describing both the kinds of relationships and the reason why these relationships are important. The focus of this stage of the analysis was to identify overarching themes on the reasons why these different relationships are significant. After initial generation of themes by the first author, the other authors revised these for credibility and some themes were merged with others. In the last stages of the steps in the analysis we refined some of themes and presented a report of the findings accompanied by relevant excerpts. Data saturation was reached when no new themes emerged during the analysis of the data. In each of the samples (rural and urban), data saturation was reached before the 50th participant and we only continued with analysis up to 50 participants to increase the trustworthiness of the findings of our study.

Fig. 1
figure 1

Schematic presentation of the data analysis process

Full size image

Results and Discussion

An overview of our findings showed that there were more similarities in the responses from older individuals living in the rural and urban areas than differences. Themes that were similar across both groups are presented, followed by the thematic areas that were unique to the different social contexts. Findings will be accompanied by excerpts providing evidence for identified themes. Excerpts from urban participants will be identified with U/P and from rural participants with R/P (Table 1).

Table 1 Summary of themes and sub-themes on reasons for important relationships
Full size table

Relational Support Dynamics

Our findings revealed that across both rural and urban samples, relational support was a critical reason for the importance of relationships across a wide variety of networks. These networks included spouses, children, friends, neighbours, religious organisations and the community in general who provided support and were an avenue for the experience of mutual support. Participants indicated providing support, receiving enacted support, and mutual support as reasons for these relationships being important to them. These sub-themes are discussed below.

Providing Support

One of the facets of relational support that was identified in this sample of older individuals was the need to provide support to others in terms of care, emotional support and affection. Across both contexts, older individuals noted that having the opportunity to offer assistance to others and expressing affection to family members were important motivations for enacting and maintaining social networks.

“Children … ..I love my children very much” (R/P 1, 60-year old female)

“Parents … … I love my parents so much” (R/P 17, 64-year old female)

“Children … … I love them, I protect them” (U/P 19, 64-year old female)

“My husband … ..He is the head of the family and I'll give him support and love” (U/P 1, 60-year old female)

Previous research has distinguished between enacted, perceived and provided support (Siedlecki et al. 2014; Thomas 2010) and found that provided support among individuals between ages of 45 and 65 predicted positive affect in an American sample (Siedlecki et al. 2014). Similar findings were confirmed by Thomas (2010) who indicated that giving support to others was related to an increase in older adults’ well-being. Thomas (2010) argued that giving support to children and other family members bolstered identity and fostered a sense independence and utility to others. More so, providing support allowed older adults to engage in socially productive behaviour (Krause et al. 1992). This finding is in line with identity theory by Stryker (2007) who indicated that social interactions provide a platform for the formation of identity, which in turn influences behaviour (Stryker 2007).

Enacted Support

Apart from being able to provide support and express affection to significant others, there was a need to know that support would be available when required. Support in this case was in the form of practical care such as children going on errands and the supply of essential material resources.

“Neighbours … … Is the first person to respond in good or bad times, and also looks after my house when I am not there” (R/P 5, 63-year old female)

“Children … … .Because they take care of me and they take me to the doctor when I am sick” (U/P 18, 64-year old female)

Given that people in this group were at a stage in their lives characterised by high dependence, it was expected that a motivation for relational interactions grounded in experiencing various kinds of support would emerge. In line with our findings, Siedlecki et al. (2014) found that enacted support was important for the well-being of older adults because it provided an opportunity for improved life conditions. Diener et al. (2010) also espoused that as compared to perceived support, enacted support was crucial for increase in life satisfaction because life satisfaction was partly based on the living conditions of the individual. In the present study, where both contexts were socio-economically disadvantaged, it was expected that the experience of enacted support would be an important motivation for relational interactions since such interactions had the capability of improving their general life conditions.

Participants also noted that they expected obedience from children as part of the relational dynamics across generations. It was the duty of children to meet the needs of parents and they were obligated to fulfill their requests.

“Cooperating with children … to send them to run errands and they should be willing to do that” (R/P 1, 60-year old female)

“Children … .They listen, I love them” (U/P 4, 60-year old female)

“My children … .Because they take care of me, they do everything that I say they must do” (U/P 7, 60-year old female)

This finding has been corroborated by Nguyen and Seal (2014) who found that in order for older individuals in China to experience happiness from intergenerational relations, children were required to show love and respect to them. They argued that the elderly believed that it was the responsibility of children to carry out household chores and make life easy for the older individuals. In the African context, the extended family is generally thought of as a safety net for the aged who are unable to work and earn an income (Kakwani and Subbarao 2007), although this has not always been sufficient in rural poor households (Schwarz 2003). The notion of a ready safety net from extended family members could explain why relationships with children and grandchildren formed an important source of support.

Mutual Support

Reciprocal support and mutual exchange of support emerged as another aspect of relational support dynamics across both samples. This relational support dynamic points to the importance of both receiving and giving support in social interactions.

“Family … . We help each other” (U/P 8, 61-year old female)

“Sister … .She helps me and I'll always support her” (U/P 1, 60-year old female)

Although most studies have focused on the experience of social support, there is evidence indicating that mutual/reciprocal support through social engagement and informal social activities are important relational experiences among the aged (Huxhold et al. 2013). In an African-American sample, it was found that participants were very much involved in reciprocal support networks with their extended family, friends and church members. These reciprocal exchanges involved the giving and receiving of support across social networks. For example, older children in the family provided financial support to the elderly while older parents took care of the younger children (Taylor et al. 2017). Schatz (2007) in a study across 21 villages in South Africa, argued that although older women might receive financial assistance from their children or grandchildren, these older women also engage in providing support to vulnerable children in the home and fostering orphaned children. This exchange of support between older adults, children and their grandchildren point to the importance of several relational ties and demonstrates the collective orientations underscoring relationships in this context. This finding dovetails with the theoretical model of thriving through relationships which espouses that relational support enables an individual to thrive through two basic mechanisms: source of strength support and relational catalyst support (Feeney and Collins 2015). While source of strength support refers to the buffering effect of support in times of adversity, relational catalyst pertains to continuous presence of support in the absence of adversity in order to aid engagement in life opportunities. In line with the work of Feeney and Collins (2015), our study showed that relational dynamics in the form of reciprocal exchange could provide life opportunities for growth and development of different members of the family.

Spirituality and Relational Dynamics

Spiritual organisations also emerged as sources of important relationships because of positive experiences from interactions with members of these organisations. These interactions offered opportunities for religious expressions through prayers and also the experience of support from other members of these communities. From the need for spiritual guidance to the experience of pleasure during religious interactions, benefits of religious engagements emerged as crucial reasons for important relationships. Subthemes will now be presented.

Accountability, Sense of Security and Expectation of Support

Among participants in the urban context, accountability to a higher power underscored their motivation for affiliation and engagement in religious activities. In addition, a sense of security and expectation of support influenced social interactions in religious settings. The sense of security was mostly expressed to be crucial for a proper passage into the ancestral world after death. Also noteworthy in the excerpts is the appreciation of the necessary vicissitudes of life from a spiritual source.

“Church … .. Because it teaches me laws” (U/P 15, 63-year old female)

“God … .. So that He can see the works that I have done, so that when I die He would have seen my works” (U/P 15, 63-year old female)

The assurance that they would be properly buried might have been important for the older people in this sample, since they were at a stage in their life where thoughts of death and the “life after” were frequent. The experience of church-based informal social support was found among a sample of African-Americans. Support included transportation, financial assistance, support during illness and help with chores (Taylor et al. 2017). Other studies have indicated that church members play an important role in the daily lives of African-Americans (Taylor et al. 2013). In line with one of the motivations listed in our study, Taylor et al. (2017) reported that church members may support a family with funeral expenses. Moreover, a higher level of religious involvement was argued to be associated with higher received support (Krause and Hayward 2014).

Sense of Identity and Spiritual Support

In the rural sample, the importance of interactions within religious communities and with a Higher Being were motivated by the need for a sense of identity and spiritual support from God in the form of answers to prayers.

“Church … ..It builds me spiritually” (R/P 25, 74-year old female)

“God … .. Because when I pray, after a while, I see answers and that life changes for the better” (R/P 11, 61-year old male)

There was also the experience of emotional support from other congregants as described in the following excerpt:

“Church … Because it helps me when am having troubles and it also helps me when am not well emotionally” (R/P 14, 63-year old female)

Closely in line with this finding, research has shown that underscoring successful ageing among Spanish older individuals was the experience of spirituality (Tomás et al. 2016). This could be because an intimate relationship with and dependency on a transcendent force enabled coping with difficulties (Crowther et al. 2002). Moreover, Chilisa (2011) put forward that relationships with the environment, divinities and non-living individuals determined appropriate behaviours in an African context and, in line with the present study, a sense of identity within the community of older individuals.

Need for Resources and Generativity

The need for resources was found to underlie intimate relational interactions especially between children and their parents. This need for resources was linked to ensuring children’s well-being and as an expression of own generativity. Across both contexts, there was an emphasis on accessing resources in order to cater for the needs of children. Participants indicated that money and education were crucial because it either provided the opportunity for children to have access to resources in the future or were the actual resources required to meet the needs of children in the present. In addition, participants were also concerned about providing children with the necessary investment that will assist them to provide for their parents in future.

“Money … I want to be able to support myself and my children because their father passed away” (R/P 10, 61-year old female)

“Work … I want to work for my kids so they have a beautiful place to live in, I want them to have a bright future” (R/P 10, 61-year old female)

“Money … … Because I can take care of my family” (U/P 14, 64-year old male)

It is however noteworthy that the expectation of future support from children was unique to the rural sample.

“Children … ..Because in future they will look after and work for me” (R/P 23, 76-year old female)

“Children … .I wish children will help me” (R/P 19, 73- year old female)

“Children … ..I want my children to work so that we can move forward (R/P 20, 73-year old female)

Roos et al. (2013) argued that due to challenges brought about by droughts in the rural areas in South Africa, older individuals engaged in relational coping. One dimension of this relational coping was the norm of reciprocal care and interaction across generations. This norm might have resulted in older individuals feeling responsible for the well-being of their children or grandchildren with the aim of reducing the incidence of poverty in future generations. Chilisa (2011) and Amoateng et al. (2004) argued that collectivist and communal orientations in the African contexts tend to inform expectations and provision of support across relational ties because of the importance placed on in-group members. Apart from the expectation of future support from children, the need to access resources for children was grounded in genuine concern for the future of the younger generation. This finding was corroborated in an elderly Chinese sample where successful ageing was characterised by financial stability which allowed older people to take care of themselves and their families (Nguyen and Seal 2014).

Cordiality and Harmony

The motivations for relational interactions exceeded instrumental benefits to include the need for harmony and cordiality across various relational networks. Participants noted that they appreciated the ability to work well and live in peace with friends and families.

“Family … … .I get along with my family” (R/P 6, 60-year old female)

“Family … … We get along with each other” (R/P 1, 64-year old female)

A prominent reason for certain important relationships among older individuals was the experience of interpersonal harmony. Previous research revealed that maintaining harmony in family relationships was a key attribute of successful ageing among Chinese and Hmong elders living in the USA (Nguyen and Seal 2014). These authors noted that the elderly negotiated family roles and responsibilities in order to ensure harmony. Conflicts were avoided by ensuring a balance in responsibilities assigned to different family members. Similar findings have been confirmed in other studies (Liang and Luo 2012; Nosraty et al. 2015). In a study by Pflug (2009), happiness in the South African context was found to be related to harmony and a stable environment where everybody is well-off. Relatedly, a study in the Eastern Cape of South Africa showed that life satisfaction was determined by not only financial status but the ability to live well with others (Møller et al. 2015). The emphasis on harmony and stability in these studies point to a collectivist orientation (the need to protect the stability of the in-group) which often underlies relationships in the African context. This dovetails with our findings on the importance of harmony with significant others.

The following themes were unique to either the urban and rural samples.

Appreciation of Family as Source of Well-Being

In the urban sample, there was a general recognition of the role of the family in enhancing well-being and the appreciation of such familial ties. Apart from the satisfaction of needs, appreciation of mutual support, and opportunities for expressing affection, the older individuals in this group reported that the family played a central role in their lives. One noteworthy role was the provision of a sense of identity. Looking back at their lives, these participants were appreciative of the role of family bonds in shaping them to be who they are today.

“Children … … They are my life and I make sure that they fit in the community” (U/P 1, 60-year old female)

“Family … ..Because I love them” (U/P 4, 60-year old female)

In the rural sample, emphasis on the family was primarily linked to support required and not the appreciation of the existence of these significant others. Several studies have pointed to the central role of the family in successful aging (Liang and Luo 2012; Nguyen and Seal 2014). Stowe and Cooney (2014) commented on Rowe and Kahn’s (1997) theorising of successful ageing indicating that family relationships among other determinants were crucial for development with potential long-term consequences. In exploring the mental well-being and social capital of the elderly in Western Cape, South Africa, Chipps and Jarvis (2016) indicated the value of a primary network for this sample of older individuals. Particular emphasis was given to the value of relationships with children in this context.

Dependability and Trust

In the rural sample, one of the reasons for important relationships was the opportunity to experience dependable and trusting relationships. Such relationships included individuals outside the family that participants had the opportunity to build reliable relationships with. The fact that these relational ties would be available anytime it would be needed created a sense of intimacy and reassurance for the participants.

“Friend … ..Because we advise each other and in future the person will look after my children” (R/P, 60-year old female)

“Friend … .We are like a tooth and a cavity, we trust each other at all times” (R/P 4, 68-year old male)

In line with our findings, Kawachi (2006) argued for the importance of trust in interpersonal relationships. In addition, Webster et al. (2015) found that trust mediated the relationship between social networks and mental health. They argued that trust is a resource that shapes social networks. In rural communities, residents may be more involved in and dependent on their social networks when compared to urban communities, who may have more access to other resources which can lead to a lesser level of dependence on others. Due to greater reliance on others, people from rural communities may place more emphasis on trust and dependability in their relationships.

Critical Reflections from the Study

Ageing in the South African context has not received the warranted scholarly attention. Although this stage of life is characterised by challenges, there are positive experiences that serve as buffers and resources for well-being, one of which is relationships. As part of the primary aim of the study, our findings, in line with previous research on relational well-being (Siedlecki et al. 2014), elucidated the central role of relational experiences in an individual’s well-being. An important contribution to research on relationships is that apart from the apparent need for support, which is underscored in most scholarly work on relationships, there is evidence for the giving of support, mutual exchange of support, interpersonal and intrapersonal harmony as well as an opportunity to extend one’s self to future generations. Also noteworthy is further confirmation of the existence of horizontal and vertical relatedness not only for support but also for establishing purpose and identity.

Related to the secondary aim of the study, the themes highlighted in the present study reflected more similarities than differences in the relational experiences of older individuals living in urban and rural contexts in South Africa. This might be because urbanisation tends to impact more on the younger generation (Campbell 1994) with only some rippled effect on the aged. Despite the similarity, our findings revealed that dependability and trust was a key reason for important relationships only in the rural context. It is possible that the relative less exposure to urbanisation has resulted in the preservation of closely-knit relationships that were easily accessible in the times of need. It is also likely that the social change and increasing individualising tendencies occurring in urban contexts have reduced the expectation and emphasis being placed on the need for dependable relationships. An alternative interpretation is that rural participants expressed the need for dependability and trust in friends because most of the members of the nuclear or extended family had moved to the cities, reducing the availability of familial networks, while the urban participants might have had the luxury of family life.

The family appeared to be especially important in urban settings, probably because of frequent contact without the problem of migration that occur in rural settings. Other social ties such as friends and neighbours might have had a smaller impact in meeting relational needs due to urbanisation. An important policy implication is the need to widen the network of ageing individuals in urban settings in particular by creating other means of social interactions for this group. State-enacted support for the elderly is also necessary in communities where such support may be less effective.

Limitations of the Study

Some of the limitations of our study include the use of semi-structured questions that did not provide much room for in-depth exploration. An in-depth follow-up study will provide further insight. Another limitation of the study is that a unique sample that speaks only to the relational experiences of two groups of participants in the North West Province of South Africa. Also, although the two communities differ with respect to their levels of urbanisation, participants from both communities were Setswana-speaking which implies a shared cultural heritage. Further studies can explore other communities with a greater number of differences.

Conclusions

The reasons for important relationships among older individuals in the South African context reveals an interdependent dimension of relational well-being including the experience of financial stability, respect for the elderly, predictability and trust as highlighted by McCubbin et al. (2013). This suggests that ageing in this context should not be understood only from the point of view of the need for support and decline in well-being but must be considered to also encompass positive experiences such as harmony and connectedness with both the living and supernatural. Our findings add to the work of Roos et al. (2013) in South African rural communities, which emphasises relational coping by showing that apart from coping, relationships are enjoyed as an end in and of itself and as an opportunity to extend oneself. The multifaceted reasons for important relationships provided by participants from both contexts necessitate that new intervention programs in communities give some consideration to strengthening relational ties by facilitating skills and opportunities for the building, maintenance and promotion of various kinds of interpersonal and group resources and resilience. Support networks are important in times of adversity, but also for capitalising on opportunities as suggested by Feeney and Collins (2015).

As relationships had been shown in the literature to be of utmost importance for psychosocial health and even physical well-being, future research can explore the effectiveness of various kinds of interventions to strengthen relational networks and enhance relational wellbeing for older people especially in rural areas where family members might have migrated to cities. Building cultures of positivity in a community can contribute to thriving of older people living in these communities.