Kinship and friendship: The association between different sources of social contact and self-rated health in Turkey

Abstract

How does one’s connectedness in their social relationships affect their health? For a long time, this question has piqued the interest of sociologists. From Durkheim’s theory of suicide to recent empirical studies in medical sociology, the positive association between social integration and well-being has been well established. In this paper, we revisited this topic by focusing on how different sources of social contact is associated with self-rated health in Turkey. To do so, we examined how contact with a close friend, a parent, a sibling, and other family member – outside of one’s nuclear family – compares in terms of their association with self-rated health by using ISSP Social Network Data. We found that frequency of contact with a parent and other family member showed statistically significant and positive associations with self-rated health in some models, which was not the case for a sibling. In the full model accounting for all contact variables, however, only the frequency of contact with a close friend had a statistically significant and positive association with self-rated health. This study aims to start a discussion about the growing importance of non-kin ties in people’s lives in Turkey despite the increasing familialism in Turkish politics and social policymaking. Policymakers and social workers tackling issues such as social isolation, social exclusion, and loneliness in Turkey should consider the prevalence of non-kin networks on individuals’ well-being.

Keywords

Friendship kinship social contact self-rated health Turkey

Introduction

Since the beginnings of sociology as a discipline, social integration has been positively associated with well-being (Berkman et al., 2000). Durkheim’s (1951) theory of suicide stipulates that a well-balanced level of social integration would protect individuals from isolation, and thus from the risk of committing suicide. He put it simply as “the bond that unites [individuals] with the [group] attaches them to life [and] prevents their feeling personal troubles so deeply” (1951: 209–210). The more connected a person is in their social network, and thus in society, the less likely they are to feel lonely and unhappy. The preponderance of research has shown the reverse to be true as well; that the level of loneliness and social isolation have a negative effect not only on one’s mental health (Baumgartner & Susser, 2013; Richardson et al., 2017), but also on physical health (Luo et al., 2012; Løhre, 2012), and even cause greater mortality risk (for a review, see Holt-Lunstad et al., 2015). Therefore, social integration through one’s social ties is regarded as an important factor in a variety of health outcomes.

Despite the difficulty in defining and measuring the concept of social integration (for a review, see Baumgartner & Susser, 2013), extensive research has focused on its consequences. Previous studies have investigated the individual outcomes of social interaction by addressing questions such as: What kinds of benefits do people gain from their social interactions? Would more social interactions render someone happier or more satisfied? What effect would the quality and quantity of one’s social interactions have on these possible outcomes? One of the most common individual outcomes in these studies on social interactions is its impact on individual well-being. Scholars have questioned which aspects of one’s social interactions – source, quantity, quality, or diversity – are more impactful on well-being, and whether this association between social interaction and well-being is much more complicated than it appears (Ermer & Proulx, 2019; Ren et al., 2022; Yelpaze et al., 2021).

Past research in Turkey has primarily focused on the associations between social integration and outcomes such as life satisfaction, or well-being (Alıkasifoglu et al., 2012; Ergin & Mandiracioglu, 2015; Yalcin, 2015; Şener et al., 2008). Some of these studies examine these associations either for the older (Ergin & Mandiracioglu, 2015; Şener et al., 2008) or the younger population (Alıkasifoglu et al., 2012). They all demonstrated that social integration has a positive association with well-being and life satisfaction. More recent studies suggest that there are some mediators between social connectedness and well-being, such as loneliness and death anxiety (Ergün & Satici, 2024) or psychological vulnerability (Yelpaze et al., 2021). Therefore, more studies are needed to examine the association between social integration and health in Turkey by focusing on different dimensions and various mechanisms.

Turkey is considered to have a collectivistic culture with high levels of familism and stronger norms around filial piety (Ar & Karancı, 2019; Dost-Gözkan, 2022; Yavuz, 2018). According to the Life Satisfaction Survey (Turkstat, 2021) based on a nationally representative sample, family members are the primary source of life satisfaction in the lives of Turkish people. In the same survey, while 67.6% of the participants stated that their family members were the primary source of their life satisfaction, 16.8% specifically stated that it was their children, and only 4% stated that it was their parents. Indeed, ‘familialism’ is emphasized as a prominent cultural value in both state discourse and existing social policies in the country (Akkan 2018; Yılmaz, 2015). A meta-analysis study found that social support received from friends is more strongly associated with overall well-being than support received from family members (Yalcin, 2015). Yet, Turkey has also been undergoing substantial demographic and economic transformations since the early 2000s, which are putting significant burden on family relationships (Dost-Gözkan, 2022; Yavuz, 2018). Considering that Turkey has more recently suffered from economic recession makes it more critical to distinguish the effects of kin and non-kin ties on the lives of Turkish people to understand repercussions of these substantial and long-term transformations.

In this study, we address the gap in the literature by examining the different sources of social contact to understand the association of each with one’s self-rated health. We focus on frequency of contact with four different groups – parent, sibling, other family member (outside of one’s nuclear family), and close friend. We aim to understand how these different sources of social contact compare to one another in terms of their association with self-rated health. To do so, we used ISSP Social Network Data for Turkey, which was collected in 2019 using a three-stage stratified probability sampling method through face-to-face interviews and provides a sample of the adult population over the age of 18. We conducted ordered logistic regression analysis to examine the association between self-rated health and sources of social contact in Turkey.

Literature review

The association between social integration and health outcomes has been extensively examined (Berkman & Glass, 2000; Seeman, 1996), and scholars have attempted to understand this association by utilizing different but overlapping concepts such as social capital, social networks, or social engagement. Social contact (interaction or connectedness) emerges as an imperfect but fundamental indicator of social integration in most of these studies. This is especially true because social contact is the channel through which access to and accumulation of social resources occurs. Scholars have also argued that it is critical to distinguish between sources of social contact within their networks (Seeman, 1996) since social contact with different groups of people carries different sets of relational goods (Sabatini, 2009).

Frequency of contact has been demonstrated to have either non-linear effect on psychological well-being (Ren et al., 2022) or the connection between frequency of contact and mortality is much more modest than it was indicated before (Shor & Roelfs, 2015). Past studies have further discussed the differential link between quantity and quality of these social interactions and self-rated health (Fiorillo & Sabatini, 2011) as well as well-being (Sun et al., 2020). Aside from the number and frequency of contacts one has on a daily basis, the quality of interactions represents an important dimension with a significant impact on health outcomes. Therefore, even if one maintains regular contact with family or friends, whether these interactions involve support or strain may influence their life outcomes. For instance, studies on parent-child relationships have indicated negative consequences of social strain in parents’ lives, such as depressive symptoms (Wilkinson et al., 2023). Given that kin ties are more likely to entail both support and strain than non-kin ties, interactions with family may have more ambivalent consequences rather than just positive or negative outcomes (Fingerman et al., 2008; Lee & Szinovacz, 2016), compared to interactions with friends. Whether the relationships in which contact occurs are supportive, aversive, or ambivalent is an important component in shaping the impacts of social contact (Holt-Lunstad & Uchino, 2019; Uchino et al., 2012).

There are several reasons why one’s social contact affects their health outcomes. Some scholars have stated that frequency of social contact is a predictor of well-being and health because individuals have a strong motivation to contact with others and obtain their sense of belonging through these contacts (Baumeister & Leary, 1995; Ryan & Deci, 2017). Others have argued that social contact could provide much more than belonging to individuals, which would eventually also impact their health. Social contact may offer people the impression that in case need arises, social support is available through it (Antonucci, 2001). While social contact does not necessarily imply support exchanges, it may facilitate them under many circumstances. For instance, in networks with higher levels of social capital, social contact may increase access to other members’ resources and support, thereby protecting health effects under adverse conditions (Berkman & Glass, 2000).

The link between one’s social contact and their health also varies for different groups such as older people, minority groups, and those with lower socioeconomic status (SES). This line of research suggests that certain segments of the population may be more vulnerable to social exclusion, and as a result, to negative health outcomes – both mental and physical – associated with a lack of regular contact and connections. Studies have demonstrated that social isolation is associated with negative health outcomes among racial and ethnic minorities (Steel et al., 2018; Taylor et al., 2024), as well as those of lower SES (Algren et al., 2020). Research has also documented stronger and more positive association between social integration and self-rated health among older people (Kang & Michael, 2013), especially for older men (Caetano et al., 2013) and those who are unpartnered (Ermer & Proulx, 2019). Yet, the convoy model of social relationships (Kahn & Antonucci, 1980) highlights the importance of life course nature of social relations, with their impact on health and well-being influenced by both personal and situational characteristics. In this sense, changes in the quantity, quality, and composition of one’s convoy throughout the life course would have a substantial link to their well-being (David-Barrett et al., 2016).

How do the different sources of social contact compare to one another when their association with health outcomes are considered? Most studies have reported their findings on the association between social contact and health outcomes without distinguishing contact with family and contact with friends (Chon et al., 2018; Gorman & Sivaganesan, 2007) or have simply focused on contact with either family (Akaeda, 2018) or friends (Brown & Greenfield, 2021). Yet, health effects have been shown to vary depending on the source of the interaction. When it comes to physical health, some studies have found a stronger positive association between physical activity and social contact with friends than with family (Larsen et al., 2014; Nemoto et al., 2021). Communication with close friends has also been shown to be a better predictor of well-being than communication with family members during the COVID-19 pandemic (Brown & Greenfield, 2021). This could be because different sources of social contact produced distinct sets of relational resources pertaining to health. Furthermore, considering the obligatory nature of family ties and voluntary nature of friendships, distinguished associations of these sources of contact with health outcomes become more palpable.

Based on the findings of these empirical studies, we propose that the frequency of contact with family (kin) has associations with self-rated health. We further argue that contact with different groups of family members is each linked to self-rated health (H1).

H1a

Higher frequency of contact with parent is associated with self-rated health controlling for socio-demographic variables (sex, age, education, marital status, being in paid work).

H1b

Controlling for socio-demographic variables, the inclusion of frequency of contact with sibling improves the model compared to using only frequency of contact with parent as the main predictor.

H1c

Controlling for socio-demographic variables, the inclusion of frequency of contact with other family member improves the model compared to using frequency of contact with parent and sibling as the main predictors.

Controlling for socio-demographic variables, the inclusion of frequency of contact with close friend (non-kin) improves the model compared to using frequency of contact with parent, sibling, and other family member as the main predictors.

Methods

In this study, we used ISSP Social Network Data for Turkey which were collected in 2019 using three-stage stratified probability sampling (Çarkoğlu & Kalaycıoğlu, 2020). Data were collected through face-to-face interviews with individuals over 18 years old. Of those eligible to participate in the survey, 48% of respondents have completed the interview (N = 1521). The mean age of the respondents was 37.3 years (SD = 13.1). The sample comprised more female (52.4%) than male respondents (47.6%). The majority of the respondents (61%) were married. In terms of education, 27.6% of respondents had only completed primary education or less, while 12.2% had completed lower secondary, 40.6% had completed upper/post-secondary, and 19.5% had completed tertiary or higher education levels. Finally, 42% of the respondents reported that they were in paid work. Frequencies of these socio-demographic variables were presented in Table 1.

Table 1.

Descriptive statistics for socio-demographic variables and self-rated health.

		Freq./Mean	Perc./Std. Dev.	Miss. %
Age		37.3^a	13.1^a	0.46 %
Sex	Male	724	47.6	No miss.
Sex	Female	797	52.4
Education	Primary or less	420	27.6	0.13 %
	Lower secondary	186	12.2
	Upper/post-secondary (non tert.)	617	40.6
	Tertiary/upper tert.	296	19.5
Marital status	Not married	592	38.9	0.07 %
Marital status	Married	928	61.0
In paid work	No	880	57.9	0.13 %
In paid work	Yes	639	42.0
Self-rated health	Poor/fair	280	18.4	0.33 %
	Good	520	34.2
	Very good	490	32.2
	Excellent	226	14.9

^aMean and standard deviation are reported for age.

Due to missing values, the analytic sample varied between 1254 and 1283 across models. To conduct our analysis and report the model results, we used R (4.3.2) programming language and several R packages (Arel-Bundock, 2022; Gohel & Skintzos, 2022; Lüdecke et al., 2021; R Core Team, 2023; Venables & Ripley, 2013).

Study variables

Outcome

In this analysis, we used self-rated health as our outcome measure. Respondents were asked to rate their current state of health using five categories, “1 = Excellent”, “2 = Very good”, “3 = Good”, “4 = Fair”, and “5 = Poor.” We recoded self-rated health into a four-category ordinal variable by combining “fair” and “poor” categories, since the latter category has less than 5% of the respondents.

Main predictors

Our main predictors were the frequency of contact variables. The ISSP survey included a battery of questions asking respondents their frequency of contact with parent, adult child, sibling, other family member (specified as such in the dataset), and close friend. The response categories were “1 = daily,” “2 = several times a week,” “3 = once a week,” “4 = two to three times a month,” “5 = once a month,” “6 = several times a year,” “7 = less often,” and “8 = ever.” Two additional response categories were provided for the “no longer alive/lack of siblings, other family members, close friends” and “living together.” For example, in the case of frequency of contact with parent, these two response categories became “My parents are no longer alive” and “The parent I have contact with the most frequently lives with me,” respectively. We combined “daily” and “living together,” and recoded “no longer alive” as missing. We chose to drop these values because the variable did not provide information about the frequency of contact while the parent was alive or the timing of death, so the long-term impact of contact on self-rated health could not be measured. For frequency of contact with sibling, other family member, and close friend, we recoded the response category indicating absence (e.g., lack of close friend) as 0. We reverse coded the original response scales. Thus, higher values of these variables indicate a higher frequency of contact. We did not use frequency of contact with adult child in our models to avoid the age effects due to frequency of contact with parent, and because more than half of the respondents in the sample did not have an adult child.

We presented descriptive statistics of the main predictors in Table 2. The percentage of missing values was not high except for the frequency of contact with parent (14.6%). This was particularly due to the fact that we recoded parents who were “no longer alive” as missing. Hence, we did not use any method to address missingness in the main analysis. We included the percentages of missing values in Tables 1 and 2.

Table 2.

Descriptive statistics for frequency of contact variables.

Variable	Miss. %	Mean	Std.Dev.	Median	Range
Frequency of contact: parent	14.60 %	6.92	1.49	7	7 (1–8)
Frequency of contact: sibling	0.92 %	6.32	2.10	7	8 (0–8)
Frequency of contact: other family	1.58 %	5.99	2.11	7	8 (0–8)
Frequency of contact: close friend	1.12 %	6.54	1.62	7	8 (0–8)

Controls

We employed various socio-demographic controls in this analysis: sex (“0 = male”, “1 = female”), age, education (“0 = no formal equation” to “6 = upper-level tertiary”), marital status (“0 = not married”, “1 = married”), and being in paid work (“0 = no”, “1 = yes”).

Analytical strategy

We used ordered logistic (proportional odds) regression to examine the associations between our outcome and main predictors controlling for socio-demographic variables. Model 1 started with the frequency of contact with parent and control variables. Each subsequent model included a new contact variable. Hence, Model 2 had frequency of contact with parent and sibling, and Model 3 included frequency of contact with parent, sibling, and other family member as main predictors. In addition to these three contact frequency variables which involve family members, Model 4 had frequency of contact with close friend. The models had the following equation:

\begin{array}{l} \log [\frac{P (S R H \leq P o o r / f a i r)}{1 - P (S R H \leq P o o r / f a i r)}] & = α_{1} + β_{1} X_{1} + β_{2} X_{2} + \dots + β_{n} X_{n} \\ \log [\frac{P (S R H \leq G o o d)}{1 - P (S R H \leq G o o d)}] & = α_{2} + β_{1} X_{1} + β_{2} X_{2} + \dots + β_{n} X_{n} \\ \log [\frac{P (S R H \leq V e r y g o o d)}{1 - P (S R H \leq V e r y g o o d)}] & = α_{3} + β_{1} X_{1} + β_{2} X_{2} + \dots + β_{n} X_{n} \end{array}

In this equation,

α

refers to different intercepts for each threshold and

β

refers to the regression coefficients assumed to be constant for each level of the outcome. To check this last assumption, also called proportional odds or parallel regression assumption, we used Brant Test. The omnibus Brant test was statistically significant in all four models leading to the violation of proportional odds assumption. Upon closer inspection, the results showed that marital status was responsible for this violation, and after dropping this variable, proportional odds assumption holds for each model. However, the results with and without marital status (not reported) were quite similar. So, we decided to keep marital status in the models. We also checked multicollinearity for each model. The generalized variance inflation factor (Fox, 2016) values varied between 1.04 and 1.5, indicating that multicollinearity is not an issue.

Results

We presented the results of the models predicting self-rated health in Table 3. Model 1 included only frequency of contact with parent, which has a statistically significant and positive relationship with self-rated health, and controls. This result supports H1a. Model 2 includes frequency of contact with sibling in addition to frequency of contact with parent and control variables. In this model, frequency of contact with parent has a statistically significant and positive association with self-rated health, whereas frequency of contact with sibling does not. This is also the case in Model 3, which includes the frequency of contact with other family member in addition to frequency of contact with parent, frequency of contact with sibling, and controls. In this model, both frequency of contact with parent and frequency of contact with other family member have a statistically significant and positive association with self-rated health.

Table 3.

Ordered logistic regression models predicting self-rated health.

	Model 1	Model 2	Model 3	Model 4
Frequency of contact: parent	1.118*	1.120*	1.107*	1.075
Frequency of contact: parent	(0.040)	(0.046)	(0.046)	(0.045)
Frequency of contact: sibling		0.999	0.986	0.975
Frequency of contact: sibling		(0.031)	(0.031)	(0.031)
Frequency of contact: other family			1.063*	1.009
Frequency of contact: other family			(0.028)	(0.029)
Frequency of contact: close friend				1.248*
Frequency of contact: close friend				(0.050)
Age	0.975*	0.974*	0.974*	0.975*
Age	(0.005)	(0.005)	(0.005)	(0.006)
Sex	0.903	0.906	0.911	0.943
Sex	(0.108)	(0.109)	(0.110)	(0.115)
Lower secondary	1.207	1.218	1.183	1.120
Lower secondary	(0.220)	(0.222)	(0.218)	(0.207)
Upper/post-secondary (non tert.)	1.330*	1.332*	1.302	1.213
Upper/post-secondary (non tert.)	(0.193)	(0.194)	(0.191)	(0.179)
Tertiary/upper tert.	1.726*	1.755*	1.720*	1.589*
Tertiary/upper tert.	(0.291)	(0.297)	(0.293)	(0.273)
Marital Status	1.372*	1.367*	1.330*	1.371*
Marital Status	(0.162)	(0.162)	(0.158)	(0.164)
In paid work	1.254	1.258	1.245	1.223
In paid work	(0.156)	(0.158)	(0.157)	(0.155)
Poor/fair\|Good	0.262*	0.259*	0.315*	0.716
Poor/fair\|Good	(0.096)	(0.098)	(0.122)	(0.298)
Good\|Very good	1.483	1.476	1.798	4.247*
Good\|Very good	(0.542)	(0.552)	(0.697)	(1.777)
Very good\|Excellent	8.154*	8.164*	9.973*	24.065*
Very good\|Excellent	(3.017)	(3.091)	(3.911)	(10.216)
N	1283	1276	1259	1254
AIC	3342.0	3321.7	3276.6	3234.3
BIC	3398.7	3383.6	3343.3	3306.1
RMSE	2.45	2.44	2.44	2.44

* p < .05.

Finally, Model 4 includes all frequency of contact variables. Only the frequency of contact with close friend had a positive and statistically significant association with self-rated health. It is noteworthy that the frequency of contact with close friend has the highest effect size among the contact variables in the four models. Both AIC and BIC scores informed us that each model performed better than the previous one. Since the models are nested, we can compare the models for goodness of fit. Here, we fitted each model to the data used in Model 4 for comparison. The likelihood-ratio test (results not shown) indicates that the Model 4 has a better fit compared to model 1 (χ²(df_diff = 3) = 35.79, p < .001) whereas Models 2 and 3 do not improve on Model 1 with additional parameters. Model 3 improves on Model 2 (χ²(df_diff = 1) = 4.75, p < .029), but not on Model 1. Using ordered logistic regression models with controls, we tested various combinations of contact variables (results not shown). Regardless of whether we included two or three contact variables in the models, only the frequency of contact with close friend consistently showed a statistically significant association with self-rated health. Moreover, when frequency of contact with close friend was included, it was the only contact variable that showed a statistically significant association with self-rated health. These results provide support for H1a and H2, partial support for H1c, but not for H1b.

When we look at the controls, age, having tertiary/upper tertiary compared to primary or less education, and marital status have a statistically significant association with self-rated health in all models. Having upper/post-secondary (non-tertiary) compared to primary or less education has a statistically significant association with self-rated health in the first two models but not in others. Although the sign of the coefficient indicates lower self-rated health for women compared to men, sex does not have a statistically significant association with self-rated health. Similarly, having lower education compared to primary education or less, and being in paid work are not statistically significant.

Conclusion and discussion

In this paper, we investigated the association between social contact and well-being by focusing on different sources of social contact. Except for the final model, we found that contact with a parent and other family member is positively associated with self-rated health whereas contact with a sibling had no significant association with self-rated health. Yet, only contact with a close friend had a statistically significant association with self-rated health in the final and full model.

Why is frequency of contact with a close friend associated with self-rated health, while frequency of contact with parent, sibling, and other family member shows mixed results? This is an especially unexpected finding for Turkish society because the most dominant discourses about family portray it as the primary source of support for all its members (Akkan, 2018; Yılmaz, 2015). As previously stated, research and national surveys show that Turkish people associate family ties with life satisfaction (Turkstat, 2021). Yet, in addition to the central location of family in Turkish people’s lives, frequency of contact with kin might also be a source of strain and conflict. Family ties are partially obligatory, supported with strong filial and familial norms compared to the voluntary and choice-based ties with friends. This could be particularly true in Turkey due to the strong social norms, which enforce filial responsibilities, intergenerational solidarity, and respect for family members, particularly for older family members. Therefore, contact with family members, except the one with parents, may be due to social embeddedness in kin networks rather than affectual ties. Furthermore, contact with family members frequently entails a set of responsibilities, which may result in higher levels of strain as well as higher levels of support. While younger adults are more likely to receive various forms of support from their parents through frequent contact (downward flows of support), contact with much older parents may potentially involve upward flows of support, particularly caregiving by adult children.

In contrast, frequent contact with friends may be devoid of the types of obligations that exist in contact with family members and therefore serve as a space to distance oneself from the obligatory responsibilities associated with family ties. Furthermore, one may be more likely to participate in a more diverse set of social activities within their non-kin networks. Frequent contact with friends through these social activities may also improve their sense of belonging and, as a result, their health outcomes (Baumeister & Leary, 1995; Ryan & Deci, 2017). We argue that the choice-based aspect of friend ties would be a reason why this mechanism of belongingness behind this association works so well. The link between frequency of contact with close friends and self-rated health raises the question of whether kin networks are becoming less prevalent, or if non-kin networks are replacing them in Turkey. Demographic changes in family structures, household size and composition may lead to increasingly individualistic lifestyles, especially in urban areas (Çobanoğlu, 2020), with crucial consequences for parent-child relationships (Dost-Gözkan, 2022; Yavuz, 2018). Indeed, Şener et al. (2008) found mixed results regarding the association between the quality and quantity of interaction with adult children and life satisfaction among the older population in Turkey.

Turkey is still considered to have a collectivistic culture with high levels of familism and strong norms around filial piety. Yet, it is also undergoing significant transformations in attributes to family ties (Kagitcibasi & Ataca, 2015) and mixed ideals of familialism and individualism in social policy making (Kılıç, 2010). As individualism grows in Turkey, the relevance of non-kin networks formed willingly (achieved) and based on preferences may become increasingly significant for one’s social integration. In a cross-cultural comparative study conducted on college students from the U.S., Jordan and Iran, Brannan et al. (2013) found that social support from friends was not associated with college students’ well-being solely in Iran. This is particularly because collectivistic values are still prevalent in Iran compared to the other countries, and therefore, the youth values support from family more than support from friends. Despite its collectivistic culture, a meta-analysis study almost a decade ago showed that support received from friends was more influential on well-being than support received from parents in Turkey (Yalcin, 2015). Akaeda (2018) also examines the role of defamialization in the association between interaction with family and relatives and happiness with a cross-national comparison. Therefore, cross-cultural variation in the association between contact with family members and well-being is worth examining.

This study has some crucial implications for both policymaking and the larger field of research. The frequency of contact with a parent and other family member is associated with self-rated health; however, they are not the only source of contact linked to well-being. This study has shown that contact with non-kin contact such as with a close friend could have a better connection to self-rated health, which implies that alternate sources of contact should be examined by researchers and strengthened in policies aiming to improve individual well-being. Furthermore, the various implications of maintaining frequent contact with family members, such as increased number of responsibilities and the possibility of ambivalent feelings should be considered. By doing so, social policymaking might become more inclusive in terms of supporting individuals for maintaining contact with their family members without jeopardizing their well-being. This is especially crucial for women who are expected to undertake these family responsibilities. Furthermore, different sources of social contact with kin should also be investigated independently and comparatively rather than being collapsed under ‘kin ties’ since this study demonstrated that not all family ties relate to well-being. Recognizing that people might rely on their close friends just as much, if not more, than on their family members would help us better understand the association between social interaction and well-being.

There are several limitations in this study. First, we utilized cross-sectional data, and our regression results are correlational, which do not warrant causal interpretation of the findings. Inevitably, we cannot rule out the possibility of bidirectionality and reverse causation between main predictors and outcome. For instance, contact with a close friend might be a consequence of better self-rated health. Moreover, the interplay between contact variables may play a role in changes in significant associations across models. Therefore, caution is warranted in interpreting the findings, particularly for variables with mixed results. Second, we only examined the frequency of contact, therefore the quantity rather than the quality of kin and non-kin ties. Future studies should also explore how self-rated health is affected by support and strain within these ties. Third, this study used survey data on frequency of contact rather than a network analysis to examine the association between social contact and self-rated health. Therefore, further studies employing network analysis would provide a more complete understanding of this association between one’s social contact and health. Furthermore, qualitative or mixed-methods research could help more thoroughly examine the mechanisms behind the link between interactions with kin and non-kin and well-being. Future research should also concentrate on social contact with other family members such as contact with children or contact with spouse and should investigate the interactional effects of several socio-demographic variables such as age, sex, marital status, and SES. Policymakers and social workers tackling issues such as social isolation, social exclusion, and loneliness in Turkey should consider the prevalence of non-kin networks on individuals’ well-being.

Footnotes

Author contributions

Gülçin Con Wright: Conceptualization; Methodology; Writing - Original Draft; Writing - Review & Editing. Tahir Enes Gedik: Methodology; Formal analysis; Writing - Original Draft; Writing - Review & Editing.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Open research statement

As part of IARR’s encouragement of open research practices, the author(s) have provided the following information: This research was not pre-registered. The data used in the research are available. The data can be obtained at: The materials used in the research are not available. The materials can be obtained by emailing: gulcin.con@tedu.edu.tr

ORCID iDs

Gülçin Con Wright

Tahir Enes Gedik

References

Akaeda

(2018). Social contact with family and relatives and happiness: Does the association vary with defamilialization? European Sociological Review, 34(2), 157–168. https://doi.org/10.1093/esr/jcx088

Akkan

(2018). The politics of care in Turkey: Sacred familialism in a changing political context. Social Politics: International Studies in Gender, State & Society, 25(1), 72–91. https://doi.org/10.1093/sp/jxx011

Algren

M. H.

Ekholm

Nielsen

Ersbøll

A. K.

Bak

C. K.

Andersen

P. T.

(2020). Social isolation, loneliness, socioeconomic status, and health-risk behaviour in deprived neighbourhoods in Denmark: A cross-sectional study. SSM-Population Health, 10, 100546. https://doi.org/10.1016/j.ssmph.2020.100546

Alikasifoglu

Erginoz

Ercan

Uysal

Kaymak

D. A.

(2012). 90. Association between self-rated health and peer characteristics among Turkish adolescents. Journal of Adolescent Health, 50(2), S57. https://doi.org/10.1016/j.jadohealth.2011.10.154

Antonucci

T. C.

(2001). Social relations: An examination of social networks, social support, and sense of control. In Birren

J. E.

Schaie

K. W.

(Eds.), Handbook of the psychology of aging (5th ed., pp. 427–453). Academic Press.

Karanci

A. N.

(2019). Turkish adult children as caregivers of parents with alzheimer’s disease: Perceptions and caregiving experiences. Dementia, 18(3), 882–902. https://doi.org/10.1177/1471301217693400

Arel-Bundock

(2022). modelsummary: Data and model summaries in R. Journal of Statistical Software, 103(1), 1–23. https://doi.org/10.18637/jss.v103.i01

Baumeister

R. F.

Leary

M. R.

(1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Interpersonal Development, 117(3), 57–89. https://doi.org/10.4324/9781351153683-3

Baumgartner

J. N.

Susser

(2013). Social integration in global mental health: What is it and how can it be measured? Epidemiology and Psychiatric Sciences, 22(1), 29–37. https://doi.org/10.1017/S2045796012000303

10.

Berkman

L. F.

Glass

(2000). Social integration, social networks, social support, and health. In Berkman

L. F.

Kawachi

(Eds.), Social epidemiology (pp. 137–173). Oxford University Press.

11.

Berkman

L. F.

Glass

Brissette

Seeman

T. E.

(2000). From social integration to health: Durkheim in the new millennium. Social Science & Medicine, 51(6), 843–857. https://doi.org/10.1016/S0277-9536(00)00065-4

12.

Brannan

Biswas-Diener

Mohr

C. D.

Mortazavi

Stein

(2013). Friends and family: A cross-cultural investigation of social support and subjective well-being among college students. The Journal of Positive Psychology, 8(1), 65–75. https://doi.org/10.1080/17439760.2012.743573

13.

Brown

Greenfield

P. M.

(2021). Staying connected during stay‐at‐home: Communication with family and friends and its association with well‐being. Human Behavior and Emerging Technologies, 3(1), 147–156. https://doi.org/10.1002/hbe2.246

14.

Caetano

S. C.

Silva

C. M.

Vettore

M. V.

(2013). Gender differences in the association of perceived social support and social network with self-rated health status among older adults: A population-based study in Brazil. BMC Geriatrics, 13(1), 122–214. https://doi.org/10.1186/1471-2318-13-122

15.

Çarkoğlu

Kalaycıoğlu

(2020). International social survey programme: Social networks and social resources - ISSP 2017 (Turkey). (Version 1.0.0). [Dataset]. GESIS Data Archive. https://doi.org/10.4232/1.13455

16.

Chon

Lee

Kim

Lee

K. E.

(2018). The association between frequency of social contact and frailty in older people: Korean frailty and aging cohort study (KFACS). Journal of Korean Medical Science, 33(51), Article e332. https://doi.org/10.3346/jkms.2018.33.e332

17.

Çobanoğlu

(2020). Türk aile yapısında ve kültüründe yakın gelecekte önem kazanacağı öngörülen değişimler [Changes in the Turkish family structure and culture anticipated to gain importance in the near future]. Journal of Social Sciences and Humanities, 4(2), 68–80. https://dergipark.org.tr/en/pub/jssh/issue/59331/846785

18.

David-Barrett

Kertesz

Rotkirch

Ghosh

Bhattacharya

Monsivais

Kaski

(2016). Communication with family and friends across the life course. PLoS One, 11(11), Article e0165687. https://doi.org/10.1371/journal.pone.0165687

19.

Dost‐Gözkan

(2022). Adolescent–parent relationships and youth well‐being in Turkey. Child Development Perspectives, 16(3), 173–179. https://doi.org/10.1111/cdep.12459

20.

Durkheim

(1951). In Spaulding

J. A.

Simpson

(Eds.), Suicide [1897]. Translated. The Free Press.

21.

Ergin

Mandiracioglu

(2015). Demographic and socioeconomic inequalities for self-rated health and happiness in elderly: The situation for Turkey regarding world values survey between 1990 and 2013. Archives of Gerontology and Geriatrics, 61(2), 224–230. https://doi.org/10.1016/j.archger.2015.06.011

22.

Ergün

Satici

S. A.

(2024). Social connectedness to ontological well-being: Mediating role of loneliness and death anxiety. Current Psychology, 43(6), 5057–5064. https://doi.org/10.1007/s12144-023-04692-0

23.

Ermer

A. E.

Proulx

C. M.

(2019). Associations between social connectedness, emotional well-being, and self-rated health among older adults: Difference by relationship status. Research on Aging, 41(4), 336–361. https://doi.org/10.1177/0164027518815260

24.

Fingerman

K. L.

Pitzer

Lefkowitz

E. S.

Birditt

K. S.

Mroczek

(2008). Ambivalent relationship qualities between adults and their parents: Implications for the well-being of both parties. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 63(6), P362–P371. https://doi.org/10.1093/geronb/63.6.P362

25.

Fiorillo

Sabatini

(2011). Quality and quantity: The role of social interactions in self-reported individual health. Social Science & Medicine, 73(11), 1644–1652. https://doi.org/10.1016/j.socscimed.2011.09.007

26.

Fox

(2016). Applied regression analysis and generalized linear models. Sage.

27.

Gohel

Skintzos

(2022). Flextable: Functions for tabular reporting. https://CRAN.R-project.org/package=flextable

28.

Gorman

B. K.

Sivaganesan

(2007). The role of social support and integration for understanding socioeconomic disparities in self-rated health and hypertension. Social science & medicine, 65(5), 958–975. https://doi.org/10.1016/j.socscimed.2007.04.017

29.

Holt-Lunstad

Smith

T. B.

Baker

Harris

Stephenson

(2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10(2), 227–237. https://doi.org/10.1177/1745691614568352

30.

Holt-Lunstad

Uchino

B. N.

(2019). Social ambivalence and disease (SAD): A theoretical model aimed at understanding the health implications of ambivalent relationships. Perspectives on Psychological Science, 14(6), 941–966. https://doi.org/10.1177/1745691619861392

31.

Kagitcibasi

Ataca

(2015). Value of children, family change, and implications for the care of the elderly. Cross-Cultural Research, 49(4), 374–392. https://doi.org/10.1177/1069397115598139

32.

Kahn

R. L.

Antonucci

T. C.

(1980). Convoys over the life course: Attachment, roles and social support. In Baltes

P. B.

Brim

O. G.

(Eds.), Life-span Development and Behavior (Vol. 3, pp. 253–286). Academic Press.

33.

Kang

Michael

Y. L.

(2013). Social integration: How is it related to self-rated health? Advances in Aging Research, 2(1), 10–20. https://doi.org/10.4236/aar.2013.21002

34.

Kılıç

(2010). Gender, family and children at the crossroads of social policy reform in Turkey: Alternating between familialism and individualism. In Ajzenstadt

Gal

(Eds.), Children, gender and families in mediterranean welfare states (pp. 165–179). Springer. https://doi.org/10.1007/978-90-481-8842-0_8

35.

Larsen

B. A.

Strong

Linke

S. E.

(2014). The association between family and friend integration and physical activity: Results from the NHIS. International Journal of Behavioral Medicine, 21(3), 529–536. https://doi.org/10.1007/s12529-013-9318-x

36.

Lee

H. J.

Szinovacz

M. E.

(2016). Positive, negative, and ambivalent interactions with family and friends: Associations with well‐being. Journal of Marriage and Family, 78(3), 660–679. https://doi.org/10.1111/jomf.12302

37.

Løhre

(2012). The impact of loneliness on self-rated health symptoms among victimized school children. Child and Adolescent Psychiatry and Mental Health, 6(1), 20–27. https://doi.org/10.1186/1753-2000-6-20

38.

Lüdecke

Ben-Shachar

M. S.

Patil

Waggoner

Makowski

(2021). performance: An R package for assessment, comparison and testing of statistical models. Journal of Open Source Software, 6(60), 3139. https://doi.org/10.21105/joss.03139

39.

Luo

Hawkley

L. C.

Waite

L. J.

Cacioppo

J. T.

(2012). Loneliness, health, and mortality in old age: A national longitudinal study. Social Science & Medicine, 74(6), 907–914. https://doi.org/10.1016/j.socscimed.2011.11.028

40.

Nemoto

Sakurai

Matsunaga

Murayama

Hasebe

Nishi

Narita

Fujiwara

(2021). Social contact with family and non-family members differentially affects physical activity: A parallel latent growth curve modeling approach. International Journal of Environmental Research and Public Health, 18(5), 2313. https://doi.org/10.3390/ijerph18052313

41.

R Core Team . (2023). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/

42.

Ren

Stavrova

Loh

W. W.

(2022). Nonlinear effect of social interaction quantity on psychological well-being: Diminishing returns or inverted u? Journal of Personality and Social Psychology, 122(6), 1056–1074. https://doi.org/10.1037/pspi0000373

43.

Richardson

Elliott

Roberts

(2017). Relationship between loneliness and mental health in students. Journal of Public Mental Health, 16(2), 48–54. https://doi.org/10.1108/JPMH-03-2016-0013

44.

Ryan

R. M.

Deci

E. L.

(2017). Self-determination theory: Basic psychological needs in motivation, development, and wellness. Guilford Press.

45.

Sabatini

(2009). Social capital as social networks: A new framework for measurement and an empirical analysis of its determinants and consequences. The Journal of Socio-Economics, 38(3), 429–442. https://doi.org/10.1016/j.socec.2008.06.001

46.

Seeman

T. E.

(1996). Social ties and health: The benefits of social integration. Annals Of Epidemiology, 6(5), 442–451. https://doi.org/10.1016/S1047-2797(96)00095-6

47.

Şener

Oztop

Doğan

Guven

(2008). Family, close relatives, friends: Life satisfaction among older people. Educational Gerontology, 34(10), 890–906. https://doi.org/10.1080/03601270802129193

48.

Shor

Roelfs

D. J.

(2015). Social contact frequency and all-cause mortality: A meta-analysis and meta-regression. Social Science & Medicine, 128, 76–86. https://doi.org/10.1016/j.socscimed.2015.01.010

49.

Steel

K. C.

Fernandez-Esquer

M. E.

Atkinson

J. S.

Taylor

W. C.

(2018). Exploring relationships among social integration, social isolation, self-rated health, and demographics among Latino day laborers. Ethnicity and Health, 23(4), 425–441. https://doi.org/10.1080/13557858.2017.1280130

50.

Sun

Harris

Vazire

(2020). Is well-being associated with the quantity and quality of social interactions? Journal of Personality and Social Psychology, 119(6), 1478–1496. https://doi.org/10.1037/pspp0000272

51.

Taylor

H. O.

Taylor

R. J.

Tsuchiya

Chatters

(2024). Social isolation, self-rated health, and self-rated oral health among African Americans. Journal of Health Care for the Poor and Underserved, 35(1), 18–36. https://doi.org/10.1353/hpu.2024.a919805

52.

Turkstat . (2021, February 17). Life satisfaction survey 2021. Turkish Statistical Institution. https://data.tuik.gov.tr/Bulten/Index?p=Life-Satisfaction-Survey-2021-45832&dil=2

53.

Uchino

B. N.

Cawthon

R. M.

Smith

T. W.

Light

K. C.

McKenzie

Carlisle

Gunn

Birmingham

Bowen

(2012). Social relationships and health: Is feeling positive, negative, or both (ambivalent) about your social ties related to telomeres? Health Psychology, 31(6), 789–796. https://doi.org/10.1037/a0026836

54.

Venables

W. N.

Ripley

B. D.

(2013). Modern applied statistics with S-plus (3rd ed.). Springer.

55.

Wilkinson

Nakamura

J. S.

Kim

E. S.

VanderWeele

T. J.

(2023). Social support and social strain from children and subsequent health and well-being among older US adults. Journal of Social and Personal Relationships, 40(9), 2980–3003. https://doi.org/10.1177/02654075231164900

56.

Yalcin

(2015). Relationships between well-being and social support: A meta-analysis of studies conducted in Turkey. Turkish Journal of Psychiatry, 26(1), 1–11. https://doi.org/10.5080/u7769

57.

Yavuz

(2018). Future living arrangement preferences of middle-aged individuals in Turkey. Senex: Journal of Aging Studies, 2(2), 3–18. https://doi.org/10.24876/senex.2018.8

58.

Yelpaze

İ.

Deniz

M. E.

Satıcı

(2021). Association between social connectedness and well-being: A study of the mediating role of psychological vulnerability. Turkish Psychological Counseling and Guidance Journal, 11(62), 367–382. https://doi.org/10.17066/tpdrd.1001712

59.

Yilmaz

(2015). “Strengthening the family” policies in Turkey: Managing the social question and armoring conservative–neoliberal populism. Turkish Studies, 16(3), 371–390. https://doi.org/10.1080/14683849.2015.1067863