Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Mar 31.
Published in final edited form as: Aging Ment Health. 2025 Mar 31;29(8):1525–1534. doi: 10.1080/13607863.2025.2484354

Reciprocal Relationship Between Sense of Control and Social Support: A Random-Intercept Cross-Lagged Panel Model

Eunbea Kim 1, Gina Lee 2, Jeong Eun Lee 1, Peter Martin 1
PMCID: PMC12213213  NIHMSID: NIHMS2070957  PMID: 40159908

Abstract

Although prior research has indicated the relationship between sense of control and social support in the context of older adults’ well-being, little attention has been given to examining the inter-relationship between them longitudinally. Shedding light on the relationship between sense of control and social support, this study aimed to examine the reciprocal relationship between them, considering the two dimensions of sense of control (mastery/constraints) and the distinctive effects of different sources of social support (family/friends support). We also explored the gender differences in the associations. This study included 9,540 individuals 50 and older from the Health and Retirement study (HRS; 2006/2008, 2010/2012, 2014/2016, and 2018/2020). To observe intra-individual changes in the relationship between sense of control and social support over time, we used a random-intercept cross-lagged panel model with the four waves. We also conducted multi-group analyses to examine the moderating effect of gender on the associations. We found that sense of control and social support were concurrently related. Although not consistent across waves, mastery and constraints predicted friends support among women. This study suggests that sense of control may emerge as significant for women’s perceived social support from friends, while it may not hold the same importance for men.

Keywords: Control beliefs, social support, well-being, longitudinal studies, gender

Given the established importance of sense of control and social support for the well-being of older adults, theories and empirical studies have addressed the significance of understanding their relationship in the context of aging (Antonucci, 2001; Antonucci & Jackson, 1987; Heckhausen & Schulz, 2010; Lachman, 2006; Lakey & Orehek, 2011). Empirical research suggests that an initial greater sense of control leads to a subsequent increase in social support or more effective use of existing social support in older adults. Conversely, other studies also found that initial higher perceived social support contributes to an increased sense of control (Gerstorf et al., 2011; Green & Rodgers, 2001). Despite these findings, existing research has been confined by short time frames and the use of traditional cross-lagged panel models, which may limit a comprehensive understanding of intuitive relationship between sense of control and social support. To address these limitations, this study employs a random-intercept cross-lagged panel model (RI-CPLM) over an extended time frame to capture a more nuanced dynamics in how these constructs reciprocally influence each other in later life.

Sense of Control and its Role

Sense of control refers to individuals’ belief in their ability to shape their lives and achieve desired goals through their personal efforts, actions, and choices (Pearlin & Schooler, 1978). While sense of control tends to exhibit stability over the life course, it can also change over time (Neupert & Allaire, 2012; Robinson & Lachman, 2017). A high sense of control helps individuals maintain effective stress-coping strategies, adopt healthier habits, and successfully adapt to age-related changes (Baltes & Baltes, 1986; Lachman, 2006; Lachman et al., 2011; Infurna et al., 2011).

Sense of control has been hypothesized to be multidimensional – mastery and constraints (Infurna & Mayer, 2015; Lachman & Weaver, 1998). Mastery represents personal efficacy in achieving desired goals; and constraints are beliefs about external barriers or obstacles that may hinder progress (Pearlin & Schooler, 1978). High mastery drives individuals to facilitate their ability to achieve goals and overcome faced stressors, whereas high constraints lead to self-doubt and hesitation in goal pursuit. That is, constraints exert a stronger effect on functional health than mastery (Infurna et al., 2018; Infurna & Mayer, 2015; Toyama & Hektner, 2023).

In older adulthood, sense of control becomes a particularly significant resource, serving as an adaptive strategy over age-related health declines (Gerstorf et al., 2011; Infurna & Mayer, 2015; Lachman, 2006). Older adults with greater sense of control are more likely to engage in physical activity and regular exercise, which in turn contributes to better health. Given these implications, it is important to understand how sense of control in older adults relates to other domains of well-being in the context of successful aging. By examining both dimensions, this study aims to address a more nuanced impact of how different facets of control associated with social well-being, particularly social support, in older adults.

Origins of Social Support and its Differential Role

Social support is the perception of receiving support from others and social networks (Taylor, 2011). Such support enhances individuals’ well-being through two mechanisms - main and buffering effects. The main effect suggests that social support directly enhances well-being, while the buffering effect proposes that it protects against stressors, leading to positive health outcomes (Cohen & Willis, 1985; Lakey & Orehek, 2011). It is often categorized as received support and perceived social support (Uchino, 2009). Received social support refers to the actual provision or exchange of support. Perceived social support involves individual’s perception of availability of support. Literature demonstrates that perceived support is more strongly associated with better well-being outcomes than received support (Eagle et al., 2019; Uchino, 2009).

Previous research has identified several types of social support, including instrumental, informational, tangible, and emotional, each of which contributes to positive health outcomes in distinctive ways (Krause, 1986). Particularly, emotional support provides concern, empathy, and encouragement through interpersonal interactions, encouraging psychological well-being. While other types of support are often tied to specific contexts, emotional support has a positive effect on a range of stressors, ultimately leading to overall positive health outcomes (Cohen & Wills, 1985; Thoits, 2011).

The significance of social support varies depending on its source (Merz & Huxhold, 2010). Although both family and friends support are the main sources, friends support tends to be more voluntary, whereas family support is perceived as an obligation in nature (Fingerman et al., 2004; Pinquart & Sörensen, 2000). Greater friends support often leads to positive outcomes, while family support has complex mechanisms to be perceived as either positive, negative, or mixed and sometimes yields unexpected or mixed outcomes depending on other factors such as relationship quality and one’s historical/cultural background (Huxhold et al., 2014).

In older adulthood, the importance of social support becomes increasingly pronounced as older adults often place a higher value on emotional and meaningful social relationships (Carstensen et al., 1992). Family support in later life often occurs within caregiving contexts, which may enhance effectiveness of support also pose challenges to autonomy and independence (Merz & Huxhold, 2010). In contrast, friends support may provide voluntary, flexible, and mutually enriching, fostering meaningful interactions in parallel relationships. Accordingly, the current study operationalizes social support as perceived emotional support and focuses on support derived from family and friends in older adults.

Theoretical Framework

To address the relationship between sense of control and social support in older adults, this study draws two established theoretical models - the Process Model of Control (Lachman, 2006) and the Motivational Theory of Lifespan Development (Heckhausen et al., 2010). The Process Model of Control suggests the antecedent and consequential role of sense of control in the relationship with social support. The model underlines that sense of control facilitates motivational and adaptive behaviors when faced with age-related changes. Greater sense of control enables individuals to leverage their abilities to actively seek available social support and attract others who are the sources of support, ultimately enhancing their social support (Lang et al., 1997; Pearlin & Schooler, 1978). Sense of control thus may be a requisite condition to receive greater social support (Gerstorf et al., 2011; Lachman, 2006). In turn, the proposed model by Lachman and Firth (2004) indicates that supportive social experiences and exchanges contribute to an increase in sense of control. Greater emotional social support helps individuals feel less helpless and more capable of addressing environmental challenges (Antonucci & Jackson, 1987; Bandura, 1977). Thus, individuals who have greater social support would be more encouraged to achieve their desired goals and foster greater sense of control (Antonucci & Jackson, 1987; Gerstorf et al., 2011; Zhou & Yao, 2020).

The Motivational Theory of Lifespan Development explains the relationship between sense of control and social support with the motivational system. The system posits that primary and secondary controls derive adaptive behaviors aimed at optimizing life course development. Social well-being, an important domain in successful aging, is influenced by individuals’ motivation system to cultivate strong social networks and seek social support. Thus, individuals with greater sense of control may be likely to be motivated to seek opportunities for social engagement, enhancing their received social support. The model also suggests that individuals invest efforts and time in multiple common life domains, including social relations with others, with a motivation to achieve greater perceived sense of control. Accordingly, individuals with greater social relations with others may foster their sense of control (Antonucci & Jackson, 1987; Zhou & Yao, 2020).

Relationship Between Sense of Control and Social Support

Within the theoretical backgrounds, empirical studies have explored the reciprocal relationship between sense of control and social support in older adults using longitudinal methods. Yet, there are some gaps in the literature. First, studies have focused more on the mastery aspect of perceived control than examining mastery and constraints simultaneously. For example, Green and Rodgers (2001) studied the association between mastery, stress, and three types of social support (advice, belonging, and tangible) from others among low-income African American mothers. They found that higher mastery predicted higher advice and tangible support in the subsequent wave, in turn, only tangible support positively predicted mastery in the subsequent wave. Second, previous studies have used infrequent assessments. Particularly, Gerstorf et al. (2011) tested the relationships between perceived control, social support, and health among middle-aged and older adults with two waves of data collected 9 years apart from the Midlife in the United States study. They revealed a bidirectional relationship between perceived control and social support. However, there are several gaps to be developed further in methods. The cross-lagged panel model they employed can be biased with between-individuals variance in estimating within-person dynamics from longitudinal perspective. Additionally, the study employs two-time points with a long interval, possibly limiting its ability to capture within-person fluctuations in intuitive inter-relationship. The potential variability in the association between sense of control and social support based on different sources of support has not been also examined. Thus, given these gaps in the literature, this study intends to examine the relationship between sense of control (mastery and constraints) and social support (family and friends support) using four waves of datasets within the RI-CLPM.

Gender Difference in Sense of Control and Social Support

Theoretical model and literature have implied that the associations between the two constructs may vary by gender. The model proposed by Antonucci et al. (2002; 2014) underlines that women are more likely to have an extensive social network and exchange emotional support with others such as friends compared to men. They gain significant psychological benefits, such as enhanced self-efficacy and sense of control, from social support within diverse social network (Antonucci, 2001). Conversely, men’s masculinity norms may discourage seeking or receiving social support beyond their spouses/partners, as it may be perceived as a threat to their independence (Addis & Mahalik, 2003; Ross & Mirowsky, 2013). Additionally, women are more likely to leverage their personal resources (e.g., sense of control) to seek and receive more social support, reflecting their emphasis on the value of social relationships over the life span (Antonucci et al., 2014). However, men’s sense of control may be less relevant in social contexts and more focused on task-oriented domains, such as work performance. An empirical study by Cheng et al. (2021) explored how the patterns of social support is related to personal mastery, finding that women tend to have mutual support exchange patterns. In contrast, men exhibited receiving support pattern in addition to exchange patterns. Men in the receiving pattern tend to report lower personal mastery. With the recognition of gendered dynamics, this study aims to investigate further how sense of control and social support are associated for men and women in a longitudinal context.

The Current Study

The current study provides a unique contribution to the literature by employing the RI-CLPM to investigate the bidirectional relationship between sense of control and social support over a 16-year period, utilizing four waves of longitudinal data. The application of the RI-CLPM allows for the separation of within-individual differences from between-individual differences, yielding more precise estimates of within-person changes in the reciprocal relationship between sense of control and social support. We also explored the distinct relationships of mastery and constraints with social support by adopting a two-factor model of sense of control. Furthermore, given the variability in the effects of the source of social support, we explored how two different sources of social support (family support and friends support) are related to the sense of control. Moreover, this study tests the moderating effect of gender on the relationship between social support and sense of control. The hypotheses of our study are below:

H1: Sense of control and social support would be concurrently associated.

H1a: Mastery would be positively associated with social support (family/friends).

H1b: Constraints would be negatively associated with social support (family/friends).

H2: Sense of control would predict subsequent change in social support.

H2a: Mastery would predict a subsequent increase in social support.

H2b: Constraints would predict a subsequent decrease in social support.

H3: Social support would predict subsequent change in sense of control.

H3a: Social support (family/friends) would predict a subsequent increase in mastery.

H3b: Social support (family/friends) would predict a subsequent decrease in constraints.

H4: Women would exhibit stronger associations between sense of control and social support longitudinally than men.

Methods

Sample

Data were drawn from the Health and Retirement Study (HRS), a biannual longitudinal panel data by the University of Michigan, sampling noninstitutionalized adults aged 50 and older in the United States. The current study utilized variables from the HRS Psychosocial and Lifestyle Questionnaire, obtained by using a leave-behind questionnaire (Smith et al., 2017). Half of the participants were given the questionnaire every four years, and the remaining half received it every four years but with a two-year delay following the first group. The participants from 2006, 2010, 2014, and 2018 and another half of participants from 2008, 2012, 2016, and 2020 were included in the sample.

Participants were limited to married and had at least one child since this study intends to focus on family support from immediate family, specifically from a married partner/spouse and children, and friends support. Accordingly, this study examined friends support of individuals who were married and had at least one child. The final sample included 9,540 individuals aged 50 and older (Appendix A). Compared to un-selected participants (i.e., not in married status and/or had not children), the selected sample reported greater level of mastery (t (7,477) = 5.31, p < .001) and lower level of constraints (t (7,467) = −5.30, p < .05). For friends support, the selected sample had a lower level of friends support than compared to un-selected participants (t (6,936) = −6.09, p < .001).

To account for missing data biases, we conducted a sensitivity analysis. Using independent t-test, we compared the characteristics of the group that completed the responses all four times with the group that did not for the key variables. Participants who completed all four waves were younger (t (10,474) = −25.93, p < .001), had a higher education (t (8,040) = 13.08, p < .001), and reported fewer physical difficulties (t (8,060) = −10.48, p < .001) than those missed waves. They also had higher mastery (t (10,378) = 10.33, p < .001), lower constraints (t (10,376) = −15.98, p < .001), and greater family and friends support (family: t (10,449) = 3.454, p = .001; friends: t (9,643) = 9.506, p <.001). To handle the instances of missing data, full Information Maximum Likelihood (FIML) was applied to estimate parameters using all available data.

Measures

Sense of Control: Mastery and Constraints

Sense of control consisted of two dimensions, mastery and constraints, and they were measured using five items each using a previously validated measure (Mastery Scale; Infurna & Mayer, 2015; Pearlin & Schooler, 1978). Mastery measured beliefs on how much participants perceive that they have the ability to do what they want (e.g., “whether or not I am able to get what I want is in my own hands”). Constraints captured the extent to which participants perceive that they do not have the ability to change their lives and handle life problems (e.g., “what happens in my life is often beyond my control”). The items used a 6-point Likert scale (1 = strongly disagree to 6 = strongly agree), and average scores were computed for each dimension. Higher scores indicate greater mastery and constraints (Mastery: α = .89; Constraints: α = .86).

Social Support: Family Support and Friends Support

Social support was measured using three items that indicated the degree of perceived positive social support scale (Walen and Lachman, 2000). An example item is: “How much do they really understand the way you feel about things?” All three items asked about the respondents’ perceived support from three different subjects: spouse/partner, children, and friends. The items used a 4-point Likert scale (1=A lot to 4 = Not at all), and responses were reversed coded. Thus, higher scores indicate higher perceived support. Three items for spouse/partner support and three items for child support were averaged (α = .75) for family support. Friends support variable was computed as average with the three items for support from friends (α = .84).

Statistical Analysis

Before conducting the main analyses, descriptive statistics and bivariate correlations were computed for all variables. For the main analyses, four sets of the RI-CLPM were conducted to explore associations between (1) mastery and family support, (2) mastery and friends support, (3) constraints and family support, and (4) constraints and friends support. The Mplus version 8.7 was used for the RI-CLPM computation (Muthén & Muthén. 2017).

The RI-CLPM allows determining the leading variable in associations between two variables in within-person difference by partialling out between-person difference over time within a cross-lagged panel model (CLPM). The critique of the traditional CLPM assumes individual scores vary around the same population means, not differentiating the between-person difference from within-person fluctuations (Hamaker et al., 2015). The RI-CLPM addresses this by including a random intercept, which separates time-invariant factors (e.g., gender or socio-economic status) from intra-individual longitudinal changes at the within-level. This is captured with a latent factor at each measurement time point, signifying individuals’ temporal deviations from the expected individual scores at each time point.

The autoregressive pathways of RI-CLPM signify the carry-over effect of a variable, where if the pathway from xt-1 to xt is positively significant (x = mastery/constraints, y = family/friends social support). The cross-lagged pathways of the RI-CLPM provide information on spill-over effects of variable x to y. The terms carry-over and spill-over effects were introduced by Mulder and Hamaker (2020). Given the large sample size of the current study, we employed a more conservative significance level, p < .01.

To test gender differences in the associations between sense of control and social support, we conducted multiple group analyses within the RI-CLPM (Mulder & Hamaker, 2020). These analyses involve the comparison of a multiple-group version of the RI-CLPM with no constraints on the cross-lagged pathways across the groups (i.e., men vs. women) to a model with constrained cross-lagged regression coefficients using chi-square values (χ2). If the chi-square difference test indicates the constrained model is not tenable, the lagged regression coefficients differ across groups.

Results

Bivariate Correlation

The bivariate correlation analysis provided insight into the relationships between age, sense of control, and social support (Appendix B). The results revealed that age was negatively associated with sense of control across all four waves. In addition, age was positively related to social support at T2 and T3. Finally, the correlations between sense of control and social support were consistently positive across all four waves.

Random Intercept Cross-Lagged Panel Model

The findings from the RI-CLPMs indicated that all four models demonstrated a good fit with the data (Table 1). Regarding the carry-over effects, the autoregressive paths of mastery over four time points were positive in the models testing mastery with family support and mastery with friends support (Appendix C; Models 1 and 2). The constraints of carry-over effects were also significant over time, and the same for the social support in the models with constraints (Models 3 and 4). Furthermore, the magnitudes of carry-over effects of family support were larger than friends support in relationships with both mastery and constraints over time. In all models, the carry-over effects of sense of control and social support demonstrated stronger effect sizes over time.

Table 1.

Model Fit of RI-CLPM and Multi Group RI-CLPM for Gender

χ2 df p AIC BIC CFI RMSEA
Model 1 44.859 9 <.001 117120.75 117374.73 .997 .02
 Free model 50.980 18 <.001 116944.80 117452.72 .998 .02
  Male 24.197
  Female 26.783
 Constrained Model 73.761 30 <.001 116943.58 117364.43 .997 .02
 Free vs. Constrained 22.781 12 .030
Model 2 42.041 9 <.001 122244.32 122498.27 .997 .02
 Free model 51.906 18 <.001 119519.64 120028.85 .998 .02
  Male 23.701
  Female 28.205
 Constrained Model 80.254 30 <.001 122010.00 122430.79 .996 .02
 Free vs. Constrained 28.348 12 <.01
Model 3 27.725 9 <.001 117614.84 117868.82 .999 .01
 Free model 35.597 18 <.01 117501.16 118009.08 .999 .01
  Male 18.145
  Female 17.451
 Constrained Model 51.913 30 <.01 117493.48 117914.33 .999 .01
 Free vs. Constrained 16.316 12 .177
Model 4
 Free model 37.591 18 .004 123179.89 123687.73 .998 .01
  Male 17.291
  Female 20.300
 Constrained Model 63.005 30 <.001 125577.68 125999.53 .997 .02
 Free vs. Constrained 25.414 12 .01
Open in a new tab

Note. Model 1 = mastery and family support, Model 2 = mastery and friends support, Model 3 = constraints and family support, and Model 4 = constraints and friends support

For the between-person differences, random intercepts of sense of control and social support were significantly correlated across all four models (see Appendix C). In Models 1 and 2, the correlations between intercepts of mastery and both family and friends support were positively significant. These results imply that, at the trait level, individuals with higher levels of mastery tend to experience higher levels of family and friends support over the four waves. In Models 3 and 4, the intercept of constraints was negatively associated with the intercepts of both family and friends support. At a trait level, individuals with higher levels of constraints would experience lower levels of family and friends support during the four waves.

Concurrent Associations

The results supported hypotheses 1a and 1b. The results of the RI-CLPM of mastery and family support revealed the positive within-person associations throughout T1 to T4 (i.e., concurrent associations; Appendix C; Model 1). Within-person friends support and mastery at T3 and T4 were also positive correlated (Model 2). The magnitudes of the concurrent associations between mastery and support from family and friends were small throughout, ranging from β = .04 to .08. The results of the constraints and family support model (Model 3) denoted that there were negative concurrent associations at all the time points T1 through T4, with somewhat stronger effect sizes compared to the paths between mastery and social support. Although not consistent throughout, there were negative concurrent associations at T4 between constraints and friends support (Model 4).

Spill-Over Effects

According to the results of the mastery and social support models, there were no significant predictive paths with family support (Appendix C; Model 1). However, mastery at T3 significantly predicted an increase in friends support at T4 (β = .07, p < .05; Model 2). For the models with constraints, there were no significant spill-over effects from constraints to subsequent family support (Model 3). With friends support, a higher level of constraints in T1 significantly predicted a higher level of friends support at the next wave (β = .05, p < .05; Model 4). The effect sizes of these spill-over effects were small. With the conservative significance level criteria in the parameters, p <.01, these findings did not support the hypotheses 2 and 3.

Moderating Effect of Gender

The findings testing a moderating effect of gender denoted significant gender differences in spill-over effects (Table 2). There were significant gender differences in the association between mastery and family support, as well as friends support. However, the model of mastery and family support showed only a small effect size of spill-over effects among women (family support T3 to mastery T4: β = −.07, p < .05). With friends support, the model for women signified a spill-over effect from mastery T3 to friends support T4, β = .12, p = .001. Furthermore, the model indicated a spill-over effect from friends support T3 to mastery T4, β = −.07, p <.05., but the effect size was small. Additionally, no spill-over effect was displayed among men (Table 2).

Table 2.

Multigroup Comparison Parameter Estimates of RI-CLPM

Model 1
MS and FA		 Model 2
MS and FR		 Model 4
CS and FR
Male Female Male Female Male Female
Parameter β (SE) β (SE) β (SE) β (SE) β (SE) β (SE)
Carry-over effect
 SC1→SC2 .06 (.03)* .04 (.03) .06 (.03) .04 (.03) .09 (.03)* .10 (.03)**
 SC2→SC3 .12 (.03)*** .13 (.03)*** .12 (.03)*** .14 (.03)*** .05 (.04) .11 (.03)**
 SC3→SC4 .15 (.03)*** .28 (.03)*** .16 (.03)*** .28 (.03)*** .12 (.04)** .20 (.04)***
 SS1→SS2 .17 (.04) .11 (.03)*** .10 (.04)* .11 (.03)** .09 (.04)* .12 (.03)***
 SS2→SS3 .11 (.04)** .22 (.04)*** .19 (.04)*** .10 (.03)* .19 (.04)*** .09 (.04)**
 SS3→SS4 .23 (.04)*** .30 (.03)*** .17 (.04)*** .17 (.04)*** .18 (.04)*** .17 (.04)***
Spill-over effect
 SC1→SS2 .06 (.03) .02 (.03) .02 (.03) −.04 (.03) −.01 (.03) .10 (.03)**
 SC2→SS3 .01 (.03) −.02 (.03) −.01 (.03) −.02 (.03) .07 (.04)* −.04 (.04)
 SC3→SS4 .01 (.03) −.02 (.03) .01 (.03) .12 (.03)** .00 (.04) −.06 (.04)
 SS1→SC2 −.01 (.03) −.02 (.05) .01 (.03) −.06 (.03) .04 (.03) −.01 (.03)
 SS2→SC3 .05 (.03) .03 (.03) .03 (.04) −.01 (.03) .04 (.04) .05 (.03)
 SS3→SC4 .01 (.03) −.07 (.03)* .06 (.04) −.07 (.03)* −.05 (.04) .00 (.04)
Correlation
 RI SC - RI SS .24 (.03)*** .29 (.03)*** .22 (.03)*** .30 (.03)*** −.26 (.03) *** −.30 (.03)***
 SC1 - SS1 .14 (.03)*** .03 (.03) .08 (.03)** .02 (.03) −.10 (.03) −.04 (.03)
 SC2 - SS2 .11 (.03)*** .06 (.03) .06 (.03) .03 (.03) −.03 (.03) .00 (.03)
 SC3 - SS3 .03 (.03) .08 (.03)** .03 (.03) .08 (.03)** .02 (.03) −.07 (.03)*
 SC4 - SS4 .11 (.03)*** .07 (.03)* .11 (.03)** .05 (.03) −.12 (.04)** −.12 (.03)***
Intercept
 SC 1 4.44 (.05)*** 4.16 (.04)*** 4.44 (.05)*** 4.16 (.04)*** 1.81 (.02)*** 1.83 (.02)***
 SC 2 4.15 (.05)*** 4.08 (.05)*** 4.15 (.05)*** 4.08 (.05)*** 1.85 (.03)*** 1.86 (.02)***
 SC 3 4.18 (.06)*** 4.05 (.06)*** 4.17 (.06)*** 4.04 (.06)*** 1.86 (.03)*** 1.88 (.03)***
 SC 4 4.15 (.07)*** 3.85 (.06)*** 4.14 (.07)*** 3.85 (.06)*** 1.93 (.03)*** 1.91 (.03)***
 SS 1 5.97 (.06)*** 5.87 (.06)*** 4.58 (.05)*** 4.98 (.05)*** 4.58 (.05)*** 4.97 (.05)***
 SS 2 6.10 (.07)*** 5.78 (.06)*** 4.65 (.06)*** 5.07 (.06)*** 4.65 (.06)*** 5.07 (.06)***
 SS 3 6.01 (.08)*** 5.90 (.07)*** 4.72 (.07)*** 4.95 (.07)*** 4.71 (.07)*** 4.94 (.07)***
 SS 4 5.75 (.08)*** 5.79 (.08)*** 4.80 (.08)*** 4.85 (.08)*** 4.78 (.08)*** 4.85 (.08)***
Variance
 RI SC .48 (.02)*** .53 (.02)*** .48 (.02)*** .53 (.02)*** .72 (.03)*** .70 (.03)***
 RI SS .17 (.01)*** .18 (.01)*** .26 (.01)*** .20 (.01)*** .26 (.01)*** .19 (.01)***
 SC I .69 (.03)*** .76 (.03)*** .69 (.03)*** .76 (.03)*** .69 (.03)*** .77 (.03)***
 SS I .14 (.01)*** .15 (.01)*** .20 (.01)*** .22 (.01)*** .20 (.01)*** .22 (.01)***
Open in a new tab

Note. SC = Sense of control, indicating either mastery (MA) or constraints (CS). SS = Social support, indicating either family support (FA) or friends support (FR). RI = random intercept.

*

p <.05

**

p <.01

***

p <.001

There were also significant gender differences in the association between constraints and friends support. There was a positive spill-over effect from constraints at T2 to friends support at T3 among men, β = .07, p < .05. The group for women showed a significant spill-over effect from constraints T1 to friends support T2, β = .10, p = .001, with its effect size being a bit stronger compared to the men’s group’s in Model 4. Overall, given the conservative significance criteria, mastery and constraints predicted subsequent levels of friends support among the female middle-aged to older adults. The findings partially supported the Hypothesis 4.

Discussion

Using the RI-CLPM on the four waves of data from the HRS, this study aimed to examine the reciprocal relationship between sense of control and social support over time throughout four models: (1) mastery and family support, (2) mastery and friends support, (3) constraints and family support, and (4) constraints and friends support. We also purposed to explore the moderating effects of gender in the relationship.

Stability of Sense of Control and Social Support

Both sense of control and social support were stable over time within the same construct within the framework of individual-level change by separating between-individual differences from the within-individual changes (i.e., random intercept). Individuals with a strong belief in their ability to get through challenges would tend to reinforce the belief over time. On the other hand, individuals underestimating their abilities and being overwhelmed by external obstacles would amplify negative thoughts regarding their abilities and situations over time. These findings underscore the enduring nature of sense of control as a stable trait throughout the life course, aligning with prior research (Robinson & Lachman, 2017).

Regarding social support, individuals perceiving higher social support from family and friends tended to experience higher social support family and friends over time. Particularly, compared to friends support, family support showed relatively stronger effect sizes over time, addressing its stable nature. This finding affirms the strong family obligation, while losing or making friends often happens due to life events such as relocation or retirement (Antonucci et al., 2014; Fuller et al., 2020).

Relationship Between Sense of Control and Social Support

Between-person differences of sense of control and social support were correlated. We also identified within-person associations between sense of control and social support across 4 waves (i.e., concurrent associations), particularly models including constraints and family support showed the stronger effect sizes. Consistent with the theories and previous findings (Heckhausen et al., 2010; Lachman, 2006), these results imply that mastery and family/friends support are closely linked, which increase simultaneously over time.

In addition, the stronger effect sizes of constraints in the relationship with family support suggest that perceived external factors would have stronger relationships with how individuals perceive support from their family compared to internal beliefs (i.e., mastery). These findings are in line with previous research on the stronger effects of constraints than mastery (Infurna et al., 2018; Infurna & Mayer, 2015; Toyama & Hektner, 2023). The findings suggest that family support would be more consistent and stronger concurrent associations with sense of control compared to friends support (Antouncci et al., 2014).

Relationship Between Sense of Control and Friends Support

With incorporating multiple group analyses for gender differences, our findings showed spill-over effects between sense of control and friends support among women. Women’s higher mastery predicted an increase in their perceived support from friends (Figure 1). This suggests that women’s internal belief as a personal resource promotes the utilization of social resources (i.e., social support) by reaching out to their friends, and accordingly, leading to increased perceived support from friends. However, although men reported higher mastery than women (Infurna et al., 2011; Lachman et al., 2011; Robinson & Lachman, 2017; Ross & Mirowsky, 2002; 2013), their mastery did not predict friends support. This finding aligns with the expectation that men’s mastery might be less relevant to fostering social support (Antonucci, 2001; Antonucci et al., 2014; Green & Rodger, 2001). Instead, their mastery may be more closely associated with accomplishment or success in a society such as the workplace (Reynolds et al., 2007). These results imply that mastery serves as a significance resource for enhancing women’s social support from friends but has limited relevance for men in this context.

Figure 1.

Figure 1.

Open in a new tab

RI-CLPM: Relationship Between Mastery and Friends Support Among Women

Note. Nonsignificant paths are indicated in gray dash arrows, and paths p <.05 are indicated in full black arrows. Bolder black lines indicate stronger effect sizes. Standardized values. MS = Mastery. FR = Friends support.

In addition, higher constraints predicted an increase in friends support among women (Figure 2). Although this finding is unexpected, the features of social relationships among women might explain the finding. Women tend to have a broad social network and strong bond with others outside of family, such as friends (Antouncci et al., 2004; Liebler & Sandefur, 2002). They actively exchange emotional support (e.g., talk support) with friends and ask for help from friends compared to men (Fischer & Beresford, 2015; Prezza & Pacilli, 2002). Thus, when they experience feelings of constrain from external factors, they may leverage their social resources by asking for help and support from their friends. In contrast, for men, we did not find significant spill-over effects between constraints and friends support. Men generally have fewer social network and feel less responsible in social relationships with others except with their spouses/partners (Antonucci et al., 2004, 2014; Liebler & Sandefur, 2002; Stronge et al., 2019). This may explain the weak association between sense of control and friends support among men. Accordingly, despite feeling constrained, their constraints may be less intertwined with social contexts such as relationship with others and social support. Overall, the findings support the gendered dynamics in the relationship between sense of control and social support. They highlight that sense of control would significantly predict friends support among women, while the same associations appear to hold less relevance for men.

Figure 2.

Figure 2.

Open in a new tab

RI-CLPM: Relationship Between Constraints and Friends Support Among Women

Note. Nonsignificant paths are indicated in gray dash arrows, and paths p <.05 are indicated in full black arrows. Bolder black lines indicate stronger effect sizes. Standardized values. CS = Constraints. FR = Friends support.

Moreover, the findings highlight distinct outcomes with different sources of social support in relation to sense of control. While friends support was predicted by individuals’ mastery and constraints, family support did not show significant spill-over effects with sense of control. Given the multifaceted nature of family support, which can have both positive and negative impacts, its influence on sense of control may depend on contextual factors such as relationship quality and family history (Fingerman et al., 2004; Huxhold et al., 2014; Merz & Huxhold, 2010). It is possible that this factor contributed to the weaker association between family support and sense of control. Additionally, the strong concurrent association between sense of control and family support possibly indicates that the exchange of social support within a family may occur immediately and frequently as needed in real-time rather than over time.

Limitations and Future Directions

Several limitations of this study are noted. First, the sample only included individuals who were married and had at least one child; thereby, the findings may not be generalizable to other populations, such as unmarried individuals or those without child. Future research should replicate these findings using diverse longitudinal datasets. Additionally, the dynamics of sense of control and social support may vary across different age groups. For example, friends support becomes increasingly important for older adults, whereas family support tends to be more significant for younger adults. The differences in the meaning of different support resources may lead to varying dynamics with sense of control. It is suggested to exploring age-related variations to determine how different types of support contribute to well-being throughout the lifespan.

In terms of study design, although this study expanded the understanding in the longitudinal relationship between the two constructs by using a 4-year timescale, this timescale might not be appropriate to address the mechanism between the constructs. It is recommended to employ other quantitative methods and different timescales that encourage more sophisticated insight into understanding the relationship and mechanism between the relationship.

As limitations in study measures, grouping spouse and child support under the umbrella of family support overlooks the distinct roles each plays in individuals’ lives. This study also did not consider other types of support, which may yield different effects on sense of control. For example, while instrumental support may lead to lower levels of autonomy and self-efficacy in older adults due to a lack of independence. Additionally, relationship quality is a significant factor in shaping the effects of social support, further influencing its impact on sense of control. We suggest in-depth explorations of how different types of support, relationship quality, and diverse support sources (e.g., siblings, coworkers, or grandparents) are associated with sense of control in future research.

Lastly, although statistically significant, the spill-over effects between sense of control and friends support were not consistently revealed across all four waves. The spill-over effects between the variables need to be confirmed in future studies. Additionally, the small effect values limit the practical application of findings. Using a conservative approach in significance level criteria (p < .01) due to the large sample size may increase the risk of Type II error (false negatives), leading to the rejection of truly significant results.

Conclusions

This study expands theoretical models of sense of control in older adulthood by examining its longitudinal relationship with social support across 16 years. Addressing within-individual changes in the relationship also offers a more comprehensive understanding of how changes in associations unfold over time. Furthermore, by shedding light on the two different dimensions of sense of control (mastery and constraints) and examining their associations with different sources of social support, this study provides deeper insights into knowledge of these constructs among older populations. Our study thus provides further impetus to examine how sense of control and social well-being are related longitudinally, considering additional sources and types of social support, and to explore if this relationship influences other aspects of well-being (physical or cognitive functioning) in older adults, while accounting for potential moderators (e.g., age, race, or socioeconomic status).

The findings suggest that sense of control may emerge as significant for women’s perceived social support from friends, whereas it may not hold the same importance for men. This underscores the importance of understanding gender differences in social well-being in research and interventions aimed at strengthening social connections in later life. As a practical implication, programs enhancing social engagement or connections may help enhance sense of control in older adults, particularly for women.

Supplementary Material

Supp 1

Funding details

Second author, Gina Lee, was supported by the National Institute on Aging (training grant T32AG000129), awarded to the Center for Demography of Health and Aging at the University of Wisconsin-Madison by the National Institute on Aging.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

Statement of Ethics

Since the data used in this study were collected as part of HRS of the University of Michigan, no further ethical approval was required for this study.

Generative AI

This manuscript was prepared with the assistance of Chat GPT-4o (February 2024 version) for language refinement to improve clarity and fluency in writing.

Data availability statement

The data that support the findings of this study are openly available in the University of Michigan Institute for Social Research at http://hrsdata.isr.umich.edu/., reference number [NIA U01AG009740]

References

  1. Addis ME, & Mahalik JR (2003). Men, masculinity, and the contexts of help seeking. American Psychologist, 58(1), 5–14. doi: 10.1037/0003-066X.58.1.5 [DOI] [PubMed] [Google Scholar]
  2. Antonucci TC (2001). Social relations: An examination of social networks, social support, and sense of control. In Birren JE & Schaie KW (Eds.), Handbook of the psychology of aging (pp.427–453). Academic Press. [Google Scholar]
  3. Antonucci TC, Ajrouch KJ, & Birditt KS (2014). The convoy model: Explaining social relations from a multidisciplinary perspective. The Gerontologist, 54(1), 82–92. 10.1093/geront/gnt118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Antonucci TC, & Jackson JS (1987). Social support, interpersonal efficacy, and health: A life course perspective. In Carstensen LL & Edelstein BA (Eds). Handbook of clinical gerontology. Pergamon Press. [Google Scholar]
  5. Antonucci TC, Lansford JE, Akiyama H, Smith J, Baltes MM, Takahashi K, Fuhrer Rebecca & Dartigues JF (2002). Differences between men and women in social relations, resource deficits, and depressive symptomatology during later life in four nations. Journal of Social Issues, 58(4), 767–783. [Google Scholar]
  6. Bandura A (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215. doi: 10.1037/0033295X.84.2.191. [DOI] [PubMed] [Google Scholar]
  7. Carstensen LL (1992). Social and emotional patterns in adulthood: Support for socioemotional selectivity theory. Psychology and Aging, 7(3), 331–338. 10.1037/0882-7974.7.3.331 [DOI] [PubMed] [Google Scholar]
  8. Chen Y, & Feeley TH (2014). Social support, social strain, loneliness, and well-being among older adults: An analysis of the Health and Retirement Study. Journal of Social and Personal Relationships, 31(2), 141–161. 10.1177/0265407513488728 [DOI] [Google Scholar]
  9. Cheng GH-L, Ang S, & Chan A (2021). Patterns of family support exchange and personal mastery in later life: A longitudinal study. Aging & Mental Health, 25(4), 686–694. 10.1080/13607863.2019.1701627 [DOI] [PubMed] [Google Scholar]
  10. Cohen S, & Wills TA (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310–357. 10.1037/0033-2909.98.2.310 [DOI] [PubMed] [Google Scholar]
  11. Eagle DE, Hybels CF, & Proeschold-Bell RJ (2019). Perceived social support, received social support, and depression among clergy. Journal of Social and Personal Relationships, 36(7), 2055–2073. 10.1177/0265407518776134 [DOI] [Google Scholar]
  12. Fingerman KL, Hay EL, & Birditt KS (2004). The best of ties, the worst of ties: Close, problematic, and ambivalent social relationships. Journal of Marriage and Family, 66(3), 792–808. 10.1111/j.0022-2445.2004.00053.x [DOI] [Google Scholar]
  13. Fischer CS, & Beresford L (2015). Changes in support networks in late middle age: The extension of gender and educational differences. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 70(1), 123–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fuller HR, Ajrouch KJ, & Antonucci TC (2020). The convoy model and later-life family relationships. Journal of Family Theory & Review, 12(2), 126–146. 10.1111/jftr.12376 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gadalla TM (2009). Sense of Mastery, Social Support, and Health in Elderly Canadians. Journal of Aging and Health, 21(4), 581–595. 10.1177/0898264309333318 [DOI] [PubMed] [Google Scholar]
  16. Gerstorf D, Röcke C, & Lachman ME (2011). Antecedent–consequent relations of perceived control to health and social support: Longitudinal evidence for between domainassociations across adulthood. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 66(1), 61–71. 10.1093/geronb/gbq077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Green BL, & Rodgers A (2001). Determinants of social support among low–income mothers: A longitudinal analysis. American Journal of Community Psychology, 29(3), 419–441. 10.1023/A:1010371830131 [DOI] [PubMed] [Google Scholar]
  18. Hamaker EL, Kuiper RM, & Grasman RP (2015). A critique of the cross-lagged panel model. Psychological Methods, 20(1), 102–116. 10.1037/a0038889 [DOI] [PubMed] [Google Scholar]
  19. Heckhausen J, & Schulz R (1995). A life-span theory of control. Psychological Review, 102(2), 284–304. 10.1037/0033-295X.102.2.284 [DOI] [PubMed] [Google Scholar]
  20. Heckhausen J, Wrosch C, & Schulz R (2010). A motivational theory of life-span development. Psychological Review, 117(1), 32–60. doi: 10.1037/a0017668 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Huxhold O, Miche M, & Schüz B (2014). Benefits of having friends in older ages: Differential effects of informal social activities on well-being in middle-aged and older adults. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 69(3), 366–375. 10.1093/geronb/gbt029 [DOI] [PubMed] [Google Scholar]
  22. Infurna FJ, Gerstorf D, Ram N, Schupp J, & Wagner GG (2011). Long-term antecedents and outcomes of perceived control. Psychology and Aging, 26(3), 559–575. 10.1037/a0022890 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Infurna FJ, Kappes C, & Fraire N (2018). Long-term antecedents of constraints and mastery: Findings from the Health and Retirement Study. Psychology and Aging, 33(6), 965–974. 10.1037/pag0000281 [DOI] [PubMed] [Google Scholar]
  24. Infurna FJ, & Mayer A (2015). The effects of constraints and mastery on mental and physical health: Conceptual and methodological considerations. Psychology and Aging, 30(2), 432–448. 10.1037/a0039050 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Krause N (1986). Social support, stress, and well-being among older adults. Journal of Gerontology, 41(4), 512–519. 10.1093/geronj/41.4.512 [DOI] [PubMed] [Google Scholar]
  26. Lachman ME (2006). Perceived control over aging-related declines: Adaptive beliefs and behaviors. Current Directions in Psychological Science, 15(6), 282–286. 10.1111/j.1467-8721.2006.00453.x [DOI] [Google Scholar]
  27. Lachman ME, & Firth KMP (2004). The adaptive value of feeling in control during midlife. In Brim OG, Ryff CD, & Kessler RC (Eds). How healthy are we? A national study of well-being at midlife (pp. 320–349). The University of Chicago Press. [Google Scholar]
  28. Lachman ME, Neupert SD, & Agrigoroaei S (2011). The relevance of control beliefs for health and aging. Handbook of the psychology of aging (pp. 175–190). Academic Press. [Google Scholar]
  29. Lachman ME, & Weaver SL (1998). The sense of control as a moderator of social class differences in health and well-being. Journal of Personality and Social Psychology, 74(3), 763–773. 10.1037/0022-3514.74.3.763 [DOI] [PubMed] [Google Scholar]
  30. Lakey B, & Orehek E (2011). Relational regulation theory: A new approach to explain the link between perceived social support and mental health. Psychological Review, 118(3), 482– 495. [DOI] [PubMed] [Google Scholar]
  31. Lang FR, Featherman DL, & Nesselroade JR (1997). Social self-efficacy and short-term variability in social relationships: The MacArthur successful aging studies. Psychology and Aging, 12(4), 657–666. doi: 10.1037/0882-7974.12.4.657. [DOI] [PubMed] [Google Scholar]
  32. Liebler CA, & Sandefur GD (2002). Gender differences in the exchange of social support with friends, neighbors, and co-workers at midlife. Social Science Research, 31(3), 364 391. 10.1016/S0049-089X(02)00006-6 [DOI] [Google Scholar]
  33. Merz EM, & Huxhold O (2010). Wellbeing depends on social relationship characteristics: Comparing different types and providers of support to older adults. Ageing & Society, 30(5), 843–857. 10.1017/S0144686X10000061 [DOI] [Google Scholar]
  34. Mulder JD, & Hamaker EL (2020). Three extensions of the random intercept cross-lagged panel model. Structural Equation Modeling: A Multidisciplinary Journal, 28(4), 638–648. [Google Scholar]
  35. Muthén B, & Muthén L (2017). Mplus. Handbook of item response theory (pp. 507–518). Chapman and Hall/CRC. [Google Scholar]
  36. Neupert SD, & Allaire JC (2012). I think I can, I think I can: Examining the within-person coupling of control beliefs and cognition in older adults. Psychology and Aging, 27(3), 742–749. 10.1037/a0026447 [DOI] [PubMed] [Google Scholar]
  37. Pearlin LI, & Schooler C (1978). The structure of coping. Journal of Health and Social Behavior, 19(1), 2–21. 10.2307/2136319 [DOI] [PubMed] [Google Scholar]
  38. Pinquart M, & Sörensen S (2000). Influences of socioeconomic status, social network, and competence on subjective well-being in later life: A meta-analysis. Psychology and Aging, 15(2), 187–224. 10.1037/0882-7974.15.2.187 [DOI] [PubMed] [Google Scholar]
  39. Prezza M, & Giuseppina Pacilli M (2002). Perceived social support from significant others, family and friends and several socio-demographic characteristics. Journal of Community & Applied Social Psychology, 12(6), 422–429. [Google Scholar]
  40. Reynolds JR, Burge SW, Robbins CL, Boyd EM, & Harris B (2007). Mastery and the fulfillment of occupational expectations by midlife. Social Psychology Quarterly, 70(4), 366–383. 10.1177/019027250707000407 [DOI] [Google Scholar]
  41. Robinson SA, & Lachman ME (2017). Perceived control and aging: A mini-review and directions for future research. Gerontology, 63(5), 435–442. 10.1159/000468540 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Ross CE, & Mirowsky J (2002). Age and the gender gap in the sense of personal control. Social Psychology Quarterly, 65(2), 125–145. 10.2307/3090097 [DOI] [Google Scholar]
  43. Ross CE, & Mirowsky J, (2013). The sense of personal control: social structural causes and emotional consequences. In Aneshensel C, Phelan J, Bierman A (Eds.), Handbook of the sociology of mental health (pp.379–402). Springer. [Google Scholar]
  44. Smith J, Ryan LH, Fisher GG, Sonnega A, & Weir DR (2017). HRS psychosocial and lifestyle questionnaire 2006–2016. Ann Arbor, MI: Survey Research Center, Institute for Social Research, University of Michigan. Retrieved from https://hrs.isr.umich.edu/sites/default/files/biblio/HRS%2020062016%20SAQ%20Documentation_07.06.17_0.pdf [Google Scholar]
  45. Stronge S, Overall NC, & Sibley CG (2019). Gender differences in the associations between relationship status, social support, and wellbeing. Journal of Family Psychology, 33(7), 819–829. 10.1037/fam0000540 [DOI] [PubMed] [Google Scholar]
  46. Taylor SE (2011). Social support: A review. In Friedman HS (Ed.), The handbook of health psychology (pp.189–214). Oxford University Press. [Google Scholar]
  47. Thoits PA (2011). Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior, 52(2), 145–161. [DOI] [PubMed] [Google Scholar]
  48. Toyama M, & Hektner JM (2023). Longitudinal associations of perceived mastery and constraints with coping and their implications for functional health for aging adults. The International Journal of Aging and Human Development, 97(2), 135–156. 10.1177/00914150221112284 [DOI] [PubMed] [Google Scholar]
  49. Uchino BN (2009). Understanding the links between social support and physical health: A life span perspective with emphasis on the separability of perceived and received support. Perspectives on Psychological Science, 4(3), 236–255. 10.1111/j.1745-6924.2009.01122.x [DOI] [PubMed] [Google Scholar]
  50. Walen HR, & Lachman ME (2000). Social support and strain from partner, family, and friends: Costs and benefits for men and women in adulthood. Journal of Social and Personal Relationships, 17(1), 5–30. doi: 10.1177/0265407500171001 [DOI] [Google Scholar]
  51. Zhou X, & Yao B (2020). Social support and acute stress symptoms during the COVID-19 outbreak: Deciphering the roles of psychological needs and sense of control. European Journal of Psychotraumatology, 11(1), 1779494. 10.1080/20008198.2020.1779 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp 1
NIHMS2070957-supplement-Supp_1.docx (26.4KB, docx)

Data Availability Statement

The data that support the findings of this study are openly available in the University of Michigan Institute for Social Research at http://hrsdata.isr.umich.edu/., reference number [NIA U01AG009740]

RESOURCES