Abstract
Amid growing disagreement about the benefits of visible and invisible support, we tested daily associations among support visibility and changes in individual and relational well-being in young adult newlyweds (Study 1) and older married couples (Study 2). To extend past research, we assessed emotional and practical support visibility in three contexts (context-general, stress-related, and goal-related) each day. In both samples, reporting context-general or goal-related support receipt predicted increases in personal and relational well-being day-to-day. Further, direct comparison between visible and invisible support days revealed relative personal and relational benefits associated with visible support days. In contrast, reporting stress-related support receipt was related to decreases in personal well-being in both samples, despite increases in relational well-being. This relative personal benefit associated with invisible stress-related support is consistent with past work showing benefits of invisible support for people experiencing major stressors. The current research highlights the need for a nuanced approach to understanding support visibility. Although invisible support may be preferable to protect self-efficacy and prevent distress in some situations (high stress or stressor-related support contexts), its benefits may be less wide-ranging than originally thought. In non-stress contexts and for relational outcomes, visible support may instead prove advantageous.
Keywords: close relationships, social support, invisible support, well-being, lifespan
Close relationships are a prominent topic of study for social psychologists, in part because of robust links between relationships and individual health and well-being (e.g., Braithwaite & Holt-Lunstad, 2017; Holt-Lunstad, Smith, & Layton, 2010). One likely mechanism underlying the association between involvement in close relationships and well-being is social support; individuals who maintain close relationships have support available to them that allows them to cope with stress and to embrace opportunities for growth (Feeney & Collins, 2015). However, social support is a broad construct, and although the perception that support is available (i.e. perceived support availability) is consistently associated with health and well-being, specific supportive actions (termed enacted support or received support) produce variable outcomes for support-recipients (e.g., Cohen, 2004; Gleason, Iida, Shrout, & Bolger, 2008; Lakey, 2013; Uchino, 2009).
Enacted Support Receipt
There are theoretical reasons to expect that receiving support should provide individual and relational benefits. Attachment theory (Bowlby, 1969/1982, 1988; Mikulincer & Shaver, 2009) and a recent perspective on thriving through supportive relationships (Feeney & Collins, 2015) each propose that receiving support enables individuals to thrive during adversity and relative calm. During adversity, close others offer a safe haven by providing emotional support (comfort and reassurance) and practical support (advice, direct problem solving) that restores security and buffers stress (e.g., Bowlby, 1969/1982; Collins & Feeney, 2000). During periods of calm, close others instead offer a secure base for exploration by providing emotional support (encouragement) and practical support (advice and assistance) that allow individuals to accept life opportunities and make progress toward specific goals (e.g., Feeney, 2004, 2007; Feeney & Thrush, 2010; Feeney, Van Vleet, Jakubiak, & Tomlinson, 2017).
The empirical research has not consistently supported these theorized benefits of support receipt. Many studies have documented benefits of support receipt to improve mood, perceived coping, and self-esteem (e.g., Biehle & Mickelson, 2012; Collins & Feeney, 2000; Feeney, 2004; Siewert, Antoniw, Kubiak, & Weber, 2011; Thorsteinsson & James, 1999), yet other studies have demonstrated negative personal outcomes associated with reports of received support. Specifically, received support can, in some cases, be costly for self-esteem (e.g., Fisher, Nadler, & Whitcher-Alagna, 1982; Nadler, Fisher, & Streufert, 1976) and lead to psychological distress (e.g., Bolger et al., 2000; Gleason et al., 2008; Shrout, Herman, & Bolger, 2006; Shrout et al., 2010).
Researchers have offered several explanations for these support receipt costs (Bolger & Amarel, 2007; McClure et al., 2014; Nurullah, 2012; Rafaeli & Gleason, 2009; Vangelisti, 2009). Bolger and Amarel (2007) provided experimental evidence that receiving support can be costly when recipients infer that support-providers believe they are incapable of coping alone. Receiving support is also problematic when it creates inequity (i.e., the support-recipient feels over-benefitted by and indebted to the support-provider; Bar-Kalifa, Pshedetzky-Shochat, Rafaeli, & Gleason, 2018; Gleason et al., 2003; Rafaeli & Gleason, 2009; Ryon & Gleason, 2018). Finally, support receipt may produce costs because it focuses attention on the stressor for which one is receiving support and therefore enhances stress (e.g., Rafaeli & Gleason, 2009).
Dyadic Perspective on Support
Because received support can have costs, researchers have explored the potential benefit of support that is provided by a support-provider but goes unnoticed or is not interpreted as support by the recipient (i.e., “invisible support”; Bolger & Amarel, 2007; Bolger, Zuckerman, & Kessler, 2000). This “invisible support” can be contrasted with “visible support”: enacted support that recipients perceive as support (Bolger et al., 2000; Bolger & Amarel, 2007). Invisible support may be preferable to visible support because it allows a support-recipient to receive assistance without feeling incapable or indebted.
Comparing invisible and visible support means taking a dyadic perspective on support and recognizing that support transactions include behaviors and perceptions combined across two dyad-members. First, an individual’s partner may either provide support or not (partner support-provision effect). Second, the individual may either perceive that he/she received support or not perceive support from the partner (perceived support-receipt effect). Although correlated, these two effects are theoretically separable and empirically distinguishable, resulting in four support classifications (e.g., Belcher et al., 2011; Bolger et al., 2000; Gable, Reis, & Downey, 2003). As shown in Figure 1, couples-members may agree that support was provided/received (visible support) or agree that no support was provided/received (support absent). People can also can fail to notice support that their partners provided (invisible support) or perceive support when their partners do not report providing it (false positive). This classic definition of invisible support as a “miss” (i.e., partners report providing support that recipients do not report receiving; e.g., Bolger et al., 2000) was later expanded to include subtle, indirect support that de-emphasizes the roles of support-provider and recipient (Girme, Overall, & Simpson, 2013; Howland, 2015; Howland & Simpson, 2010). In the expanded definition of invisible support, people may be aware of specific supportive behaviors but they presumably do not interpret the behaviors as overt support.
Figure 1.
Visual depiction of the dual effects of support receipt and partner support provision organized into four support-visibility classifications. The labels in parentheses are based on traditional signal-detection terminology.
Support visibility and individual outcomes.
Some past research has shown a relative benefit of invisible (compared to visible) support for recipients’ mood and self-efficacy (Bolger et al., 2000; Bolger & Amarel, 2007; Shrout et al., 2006). In one—now classic—study, Bolger and colleagues (2000) recruited people preparing to take the bar exam (a stressful context) and their partners to complete daily questionnaires about their support exchanges leading up to and immediately following the exam. Each day, examinees reported whether they received emotional support (whether partners “listened to and comforted” them), and partners reported whether they provided emotional support. Results revealed additive main effects of support receipt and provision in the week before the exam. Specifically, examinees reported less depression after their partners reported providing support (regardless of whether they reported receiving it), but examinees reported more depression and anxiety when they reported receiving support (regardless of whether their partners reported providing it). Examinees experienced the lowest increases in distress when they did not report receiving support but their partners reported providing support, consistent with a relative benefit of invisible support.
Several other studies have also demonstrated benefits of invisible support receipt for individual outcomes (e.g., Bolger & Amarel, 2007; Girme et al., 2013, 2018; Howland & Simpson, 2010; Lüscher et al., 2015). For example, participants who were randomly assigned to receive indirect/invisible support from a stranger prior to giving a stressful speech reported less distress than participants who were assigned to receive direct/visible support or no support (Bolger & Amarel, 2007). In an observational study, participants who received greater invisible emotional support (e.g., “subtle physical contact…off-topic humor…using one’s own or another’s similar troubles and difficulties to provide reassurance”) while discussing a personal goal made more progress on their goal over the next year than people who received less invisible support (Girme et al., 2013, p. 1445). In another study, participants who received subtle practical or emotional support (observationally-coded) but perceived low levels of received support while discussing a desired personal change experienced less anxiety and anger than participants who perceived the subtle support or received overt support (Howland & Simpson, 2010). Participants in this study also reported greater increases in self-efficacy when they received subtle practical support that they did not recognize (Howland & Simpson, 2010). Perceiving practical support may be especially problematic for self-efficacy because task-related assistance may undermine self-efficacy more than expressing care or reassurance.
Recent reviews tout these benefits of invisible support (e.g., Lakey, 2013; Taylor, 2012). However, despite the attention these studies have received, invisible support is far from consistently beneficial (e.g., Belcher et al., 2011; Biehle & Mickelson, 2012; Gable, Reis, & Downey, 2003; Gleason et al., 2008; Hinnen, Ranchor, Baas, Sanderman, & Hagedoorn, 2009; Kirsch & Lehman, 2015). Positive main effects of received support (i.e., benefits of visible and invisible support) occur often, and the impact of received social support is still poorly understood (Nurullah, 2012). In a more recent replication of Bolger and colleagues’ (2000) original study with a larger sample size and more precise analyses, Shrout and colleagues (2010) did not replicate the benefit of invisible support for individuals preparing for the bar exam. Although partners’ reports of emotional support provision were associated with examinees’ worse mood, examinees’ own reports of support receipt were unrelated to mood. Practical support receipt and provision also were unrelated to daily mood (Shrout et al., 2010). Other studies suggest that invisible support is only superior to visible support in select contexts or for select outcomes (Girme et al., 2013, 2018; Gleason et al., 2008; Lüscher et al., 2015; Maisel & Gable, 2009). For instance, invisible support predicted lower distress but more cigarettes smoked among individuals trying to quit smoking (Lüscher et al., 2015). This finding suggests that unawareness of support receipt may undermine goal progress. Additionally, Maisel and Gable (2009) found that visible support and invisible support were both beneficial when the recipient felt understood and valued. Expanding out to consider the effects of support visibility on relational outcomes adds further complexity.
Support visibility and relationship outcomes.
Theoretically, perceiving support receipt should enhance relationship well-being (i.e., satisfaction, commitment, intimacy). Support provision is one of many relationship-oriented behaviors that partners can enact to demonstrate their care for the support-recipient and commitment to the relationship (Cutrona, Russell, & Gardner, 2005; Finkel & Rusbult, 2008). Thus, people who perceive support may perceive their relationships more positively and increase their trust in and closeness to their partners (Holmes & Rempel, 1989; Murray, Holmes, & Collins, 2006). Consistent with this idea, people reported greater perceived closeness to their partners on days when they reported receiving emotional support from them (Gleason et al., 2008). This study also showed the traditional costs of received support for mood, indicating that the benefit for relational outcomes occurred despite personal costs (Gleason et al., 2008). These results are consistent with other empirical research demonstrating relational benefits of support-receipt (e.g., Feeney et al., 2017; Feeney & Collins, 2003; Feeney & Thrush, 2010; Girme et al., 2018; Overall, Fletcher, & Simpson, 2010).
Still other research has demonstrated costs of visible support receipt for relational outcomes. Support-recipients reported lower connectedness and relationship security on days when they received visible support than on days when they received invisible support (Maisel & Gable, 2009). Perceiving support receipt may be costly for relationship well-being in situations where individuals feel that their partners’ support attempts were intrusive or unresponsive (e.g., Feeney & Thrush, 2010; Maisel & Gable, 2009). Practical visible support may be especially likely to produce relational costs because advice or task assistance may be perceived as intrusive or may make an individual feel like a burden.
Goals of the Current Research
Although there has been a great deal of research on support receipt and support visibility, past research has not provided clear or consistent results. As described next, we aimed to add clarity to this literature by considering the role of support context, in keeping with the distinctions between support during adversity and relative calm (e.g., Feeney & Collins, 2015). Additionally, past research has primarily focused on a restricted sample of young adults in early stage relationships; we aimed to expand our understanding of support visibility by sampling adult couples across the adult lifespan.
Support context.
One key limitation of past work that may help account for inconsistency of results is the lack of attention to support context. In many prior studies, participants reported only context-general emotional or practical support exchanges by completing general (across-context) items like “Today, I received emotional support from my partner” (Lüscher et al., 2015, p. 712). These context-general support items may obscure substantial variability in the types of support experiences individuals had. Additionally, because several studies specifically recruited people who were experiencing a stressful life event (e.g., Bolger et al., 2000; Gleason et al., 2003, 2008; McClure et al., 2014; Shrout et al., 2010, 2006), context-general support items could actually represent stress-related support. In other studies, researchers assessed stress-related support directly by assessing whether one’s partner “listened to and comforted” them on a given day (Bolger et al., 2000, p. 955) or “tried to help [them] solve a problem” (Girme et al., 2018, p. 885), though context was not explicitly considered.
In the current research, we recruited participants without the criteria that they be experiencing a major life stressor to sample support experiences more broadly. Critically, we assessed daily context-general support, stress-related support, and goal-related support to investigate outcomes associated with support visibility in each significant context in which support is likely to occur. Consistent with past work, we assessed emotional and practical forms of support separately in each context. Support processes (what forms of support are desired and effective) differ between these stress-related and goal-related (exploration-focused) contexts. Therefore, the outcomes associated with support visibility may similarly differ by support context.
According to attachment theory, individuals instinctively seek support from close others when they experience stress, and partner availability and enacted support restores security and strengthens the relationship between the support-provider and recipient (e.g., Bowlby, 1969/1982; Mikulincer & Shaver, 2009). Therefore, people may experience better outcomes when they perceive stress-related support receipt than when they fail to notice it. This idea is consistent with past work showing that people who were distressed reported better discussion success when they perceived support receipt but people who were not distressed reported worse discussion success when they perceived support (Girme et al., 2013). Although past research has shown costs of visible support during high stress (the week before a major exam; Bolger et al., 2000), perceiving stress-related support may not be costly in everyday life where stressors are likely milder.
Alternatively, in context-general and goal-related contexts, people may benefit more from invisible compared to visible support. In non-adverse contexts, and goal contexts in particular, people aim to develop individual competence and achieve personal growth by exploring autonomously (e.g., Bowlby, 1988; White, 1959). Thus, people who receive and notice support—especially practical support—in these contexts may infer that their partner believes they cannot achieve their goals alone, or they may perceive the support (and their partner) to be intrusive. This is consistent with the concept of a secure base for exploration as being available yet non-interfering, and with research showing that intrusive partner behavior predicts a host of negative personal and relational outcomes (e.g., Feeney, 2004; Feeney & Thrush, 2010).
Although we believe it is important to distinguish support by context, we acknowledge that this distinction can be unclear and imprecise. Some people may experience goal pursuit as stressful, and others may experience stressors as opportunities for growth. Despite the potential for overlap, we expect that distinguishing context-general, stress-related, and goal-related support contexts will incrementally advance our understanding of the consequences of support visibility.
Adult lifespan.
People receive support throughout the lifespan, but past research has primarily sampled young adults in early-stage relationships. Developmental shifts in relationship dynamics and the need for support suggest that the impact of support visibility may differ in early and late adulthood. As people age, their social networks shrink, and they prioritize relationships with close others, like spouses (Carstensen, 1992; Lang & Carstensen, 1994). Couples’ interpersonal dynamics also shift as couples age together. Over time, couple-members amass a great deal of knowledge about one another and gain experience identifying and responding to one another’s needs (Berg, Meegan, & Deviney, 1998; Rauer, Sabey, & Jensen, 2014). Thus, older couples may be able to provide support to one another more sensitively and in ways that are more coordinated with support needs. Couples also tend to have more overlapping stressors and goals in later life due to lessened career roles and greater behavioral interdependence (Berg & Upchurch, 2007; Rauer et al., 2014). When couple-members have shared stressors and goals, support receipt may not be perceived as intrusive or threaten self-efficacy.
Late adulthood also presents unique challenges as physical and cognitive declines necessitate increased reliance on close others (e.g., Berg et al., 1998; Berg & Upchurch, 2007). Given the additional need for instrumental assistance and the increased emphasis on spousal relationships in late adulthood, older adults may especially desire and be receptive to visible spousal support. If stressors become more common and goals become more difficult to attain due to challenges associated with aging, visible partner support may be particularly useful to maintain state security and to pursue goals. Alternatively, it is possible that older adults may be especially resistant to visible spousal support because it may threaten their already waning independence and signal a loss of autonamy and control (Liang, Krause, & Bennett, 2001; Smith, Braunack-Mayer, Wittert, & Warin, 2007). Indeed, past research has shown costs of receiving support in late adulthood (despite benefits of support provision; e.g., Brown, Nesse, Vinokur, & Smith, 2003; Liang, Krause, & Bennett, 2001). In that case, invisible support could be optimal for older adults because they may benefit from supportive behaviors without experiencing the threats to autonomy that may accompany perceiving support-receipt.
In one study that assessed older adults’ support receipt, receiving (visible) “affectionate” support (e.g., positive feedback) was associated with lower depressive symptoms whereas receiving (visible) practical support (e.g., material aid) was associated with greater depressive symptoms (Reinhardt, Boerner, & Horowitz, 2006). In another study, aging adults who reported receiving greater daily spousal support for their goals reported greater goal progress on the same day and the following day, which led to better daily psychological, physical, and relational well-being (Jakubiak & Feeney, 2016). Because few studies have investigated older adults’ support receipt experience, and none have compared effects of support visibility, it is difficult to anticipate how older adults respond to visible and invisible support receipt. The current research will begin to answer this question by sampling and comparing younger and older adult married couples.
Overview of the Current Studies and Hypotheses
We assessed links between daily support visibility and personal and relational outcomes in a sample of young-adult newlyweds and aging (older-adult) married couples. Consistent with past research, we assessed participant’s self-reported distress, self-efficacy, and relationship quality because each of these are key outcomes of support exchanges. Specifically, support is intended to reduce distress and to offer encouragement, though it may increase distress and undermine self-efficacy by suggesting that the support-provider views the support-recipient as inefficacious (e.g., Bolger & Amarel, 2007; Bolger et al., 2000; Feeney, 2004; Howland & Simpson, 2010). Additionally, receiving support should theoretically enhance relationship quality but has been shown to predict reduced relationship connectedness in some research (Maisel & Gable, 2009). In addition to these traditional outcomes, we assessed daily coping as an outcome of stress-related support and goal progress as an outcome of goal-related support (as stress-related support and goal-related support are intended to facilitate coping and goal progress, respectively).
To build on past research, we considered both practical and emotional forms of support in context-general, stress-related, and goal-related contexts. Our hypotheses focused on the comparison between visible support (spouse-provided support that recipients recognized) and invisible support (spouse-provided support that recipients did not recognize) because we aimed to test the current conceptualization of invisible support as an optimal form of support. Given the lack of consistency in past research, we were unable to form strong hypotheses; however, we pre-registered and tested tentative hypotheses (https://osf.io/wrtjs/register/5730e99a9ad5a102c5745a8a) based on the theory that invisible support avoids support-receipt costs (e.g., Bolger & Amarel, 2007; Bolger et al., 2000).1 We limited our predictions regarding benefits of invisible support to the particular types of support and particular contexts that may be most likely to generate support costs: context-general practical support and goal-related practical support. We made other theoretically motivated predictions where possible, and we made no predictions where no theory provided a strong justification for a specific hypothesis. A summary of all hypotheses is presented in Table 1.
Table 1.
Predictions and daily support definitions
| Practical support | Emotional support | |
|---|---|---|
| Context-general support |
Prediction: (H1) Better individual and relational WB when support is invisible (not perceived) |
Prediction: (H2) Better relational when support is visible; No prediction for individual WB |
|
Support item: “Made it easy for me to do things” |
Support item: “Expressedconfidence in me” |
|
| Stress-related support |
Prediction: (H3) Better individual and relational WB and better coping when support is visible |
Prediction: (H3) Better individual and relational WB and better coping when support is visible |
|
Support item: “Gave me advice about a problem” |
Support item: “Tried to comfort or reassure me” |
|
| Goal-related support |
Prediction: (H4) Better individual and relational WB and greater goal progress when support is invisible |
Prediction: N/A |
|
Support item: “Gave me advice about my goals” |
Support item: “Encouraged me to work on my goals ” |
Note: WB = well-being. Support items are the items that measured support-receipt. Items were reworded to measured support-provision.
Context-general practical support (H1).
Consistent with the proposed benefits of invisible support, we predicted that participants would experience better individual and relational well-being when they received invisible (compared to visible) practical support on a given day. Because practical support may communicate inefficacy, we expected that individuals who failed to notice their partner’s context-general practical support provision would maintain better individual and relationship well-being.
Context-general emotional support (H2).
Contrary to the proposed benefits of invisible support but consistent with past research (Gleason et al., 2008), we predicted that participants would report greater relationship well-being on days when they received visible, compared to invisible, context-general emotional support. Perceiving emotional support should enhance relationship satisfaction because it communicates concern and affection. We did not make predictions regarding links between emotional support visibility and personal outcomes because although perceiving emotional support may have individual costs, support-recipients tend to be more receptive to emotional support, which would minimize differences between visible and invisible support. Past studies have shown costs of visible emotional support for personal outcomes, but these studies either sampled individuals in a high stress context or involved support from strangers (e.g., Bolger & Amarel, 2007; Bolger et al., 2000; Gleason et al., 2008). In marital relationships, general emotional support should be expected and therefore not costly.
Stress-related support: practical and emotional (H3).
Consistent with attachment theory’s proposition (and past work showing) that people desire and benefit from overt support during adversity (e.g., Bowlby, 1969/1982; Girme et al., 2013; Mikulincer & Shaver, 2009), we predicted that visible support for stress would predict better individual and relational well-being and greater perceived coping than invisible support for stress. We expected to observe this same pattern of results for emotional and practical forms of stress-related support receipt because both forms of support may be perceived as responsive during stress.
Goal-related support: practical and emotional (H4).
We predicted that individuals benefit more from invisible than visible goal-related practical support (e.g., advice) in terms of their personal and relational outcomes and goal progress. Practical support during goal pursuit may communicate inefficacy, be perceived as intrusive, and heighten support-recipients’ concern about achieving their goals. Although past research has shown benefits of perceived goal-related support to facilitate goal progress (Jakubiak & Feeney, 2016), we expected that there would be greater benefits associated with invisible goal-related practical support than visible practical support. This is consistent with research showing benefits of non-intrusiveness during exploration (Feeney, 2004; Feeney & Thrush, 2010). We made no strong hypotheses for goal-related emotional support because emotional support may produce fewer costs and therefore be equally beneficial when it is visible and invisible in a goal context.
We first tested each of these predictions in a sample of young adult newlyweds (Study 1). Then, we attempted to replicate our findings in a sample of older married couples to assess the generalizability of the results and test for differences in support processes across the adult lifespan (Study 2). We expected that aging participants could be more receptive to visible support because of shared stressors and goals and greater experience with support-exchanges, or we expected that aging participants could be less receptive to visible support because support could signal reduced autonomy.
Study 1: Young-Adult Newlywed Sample
Method
Participants.
Participants were recruited as part of a larger study of newlywed relationships in early adulthood (N = 229 couples; sample size determined a priori based on power analysis for larger study). To be eligible to participate in the larger study, couples had to be married for less than a year in both partners’ first marriage, and both couple-members had to be no older than 40 years of age and fluent in English. The current study includes data from the couples who completed the daily diary portion of the larger study (N = 197 couples; diary data collected March 2005 through June 2007).2 These participants were on average 27.4 years old (SD = 4.4) and primarily White (82.1%). Couples reported that their mean earned annual household income was between US$40,000 and US$60,000.
Procedure and measures.
After couples completed laboratory sessions as part of the larger investigation of newlywed relationships, they completed a seven-day daily diary. Participants completed diary items on Palm Pilots and were instructed to complete them immediately before bed for seven consecutive days. Although these questionnaires measured many aspects of daily relationship functioning and individual and relational well-being, the current investigation focused on the measures described next. The full set of daily diary items is available at https://osf.io/68av9/files. Two articles have previously been published with this daily diary dataset (Feeney & Lemay, 2012; Jakubiak & Feeney, 2016).
Daily support exchanges and visibility.
Participants saw a list of several possible partner behaviors and indicated which behaviors their partner had enacted that day by selecting a checkbox next to the behavior. Support items (summarized in Table 1) were embedded among many other kinds of interpersonal and individual behaviors. Of particular relevance to the current investigation, we used face-valid items that represented emotional and practical support in each context of interest: stress, goal, and context-general. Specifically, participants indicated whether their partner “made it easy for me to do things” (context-general practical support), “expressed confidence in me” (context-general emotional support), “gave me advice about a problem” (stress-related practical support), “tried to comfort or reassure me” (stress-related emotional support), “gave me advice about my goals” (goal-related practical support), and “encouraged me to work on my goals” (goal-related emotional support).3 We used responses to these items to measure participants’ reported support receipt.
Similarly, each participant saw a list of several possible behaviors and indicated which behaviors they had enacted themselves that day (toward their spouse) by selecting a checkbox next to the behavior. Of particular relevance to the current investigation and as summarized in Table 1, participants indicated whether they “made it easy for my spouse to do things” (context-general practical support), “expressed confidence in my spouse” (context-general emotional support), “gave my spouse advice about a problem” (stress-related practical support), “tried to comfort or reassure my spouse” (stress-related emotional support), “gave my spouse advice about his/her goals” (goal-related practical support), and “encouraged my spouse to work on his/her goals” (goal-related emotional support). We used each participants’ spouse’s response to these provision items as a measure of partner reported support provision. See Table 2 for descriptive statistics and inter-correlations of all support items.
Table 2.
Support receipt and partner support provision: Descriptives and zero-order correlations
| Variable | Mean (SD) for daily observations | Mean (SD) for person averages | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Study 1: Young adult sample | ||||||||||||||
| 1. CG - practical receipt | −.01 (.50) | .−.02 (.30) | -------- | 40** | 51** | .29** | .28** | .18** | .33** | .30** | .34** | 14** | ,47** | .20** |
| 2. CG - practical prov. | −.05 (.50) | −.05 (.29) | .22** | -------- | .32** | .48** | .16** | .38** | .18** | .36** | .18** | .35** | 23** | .37** |
| 3. CG - emotional receipt | .02 (.50) | .02 (.32) | .33** | .17** | -------- | .48** | 40** | .24** | 44** | .32** | .43** | .16** | .59** | .32** |
| 4. CG - emotional prov. | .06 (.50) | .05 (.31) | 19** | 29** | .29** | -------- | .22** | .41** | .28** | .53** | 19** | 40** | .28** | .57** |
| 5. SR - practical receipt | −.22 (.45) | −.22 (.25) | 15** | .07** | .27** | .16** | -------- | .35** | .54** | .29** | .64** | .28** | .42** | .20** |
| 6. SR - practical prov. | −.22 (.45) | −.22 (.25) | .08** | 19** | .16** | .26** | .27** | -------- | .32** | .52** | .30** | .67** | .26** | .50** |
| 7. SR - emotional receipt | −.10 (.49) | −.10 (.29) | 19** | .09** | .26** | 19** | .36** | .23** | -------- | .42** | .40** | .25** | .47** | .22** |
| 8. SR - emotional prov. | −.06 (.50) | −.06 (.29) | 14** | 14** | 19** | .34** | 19** | .37** | 25** | -------- | 19** | 48** | .25** | .47** |
| 9. GR - practical receipt. | −.23 (.44) | −.23 (.24) | .21** | .10** | .26** | 17** | .49** | .25** | .26** | .18** | -------- | .30** | .61** | .27** |
| 10. GR - practical prov. | −.21 (.46) | −.21 (.24) | .09** | .18** | 14** | .25** | .24** | .49** | 19** | .32** | .26** | -------- | .26** | .58** |
| 11. GR - emotional receipt | −.12 (.48) | −.12 (.28) | .24** | .15** | .39** | .21** | .34** | .22** | .31** | .18** | .46** | .23** | -------- | .33** |
| 12. GR - emotional prov. | −.12 (.49) | −.12 (.28) | .12** | 19** | .20** | .38** | .18** | .38** | .18** | .29** | .24** | .46** | 23** | -------- |
| Study 2: Aging sample | ||||||||||||||
| 1. CG - practical receipt | .09 (.49) | .06 (.34) | -------- | .21** | .34** | .24** | .06 | .02 | .10* | .04 | −.00 | −.07 | .09* | .01 |
| 2. CG - practical prov. | .11 (.49) | .14 (.32) | .13** | -------- | .18** | .32** | .06 | −.03 | −.02 | −.09 | −.02 | −.04 | .04 | −.11* |
| 3. CG - emotional receipt | .09 (.49) | .07 (.36) | .25** | .11** | -------- | .35** | −.03 | −.04 | 23** | .03 | −.31** | −.10** | .17** | .02 |
| 4. CG - emotional prov. | .16 (.48) | .07 (.36) | .06** | .22** | .22** | -------- | −.02 | −.22** | .03 | .20** | −.07 | −.10** | .13** | .20** |
| 5. SR - practical receipt | −.18 (.47) | −.17 (.33) | .03 | .02 | −.03 | .01 | -------- | .30** | .26** | .11** | .27** | 15** | .05 | .05 |
| 6. SR - practical prov. | −.23 (.45) | −.25 (.29) | −.02 | −.04 | −.08** | − 14** | 19** | -------- | .12** | .31** | .05 | .34** | .04 | .20** |
| 7. SR - emotional receipt | −.17 (.47) | −.20 (.31) | .08** | −.04 | −.24** | .01 | .24** | .12** | -------- | .21** | −.02 | 14** | .13** | .19 |
| 8. SR - emotional prov. | −.25 (.43) | −.24 (.29) | ..00 | − 14** | −.04 | 14** | .08** | .22** | .16** | -------- | .08 | .26** | .13** | 19** |
| 9. GR - practical receipt. | −.31 (.39) | −.28 (.28) | .00 | −.01 | −.21** | −.01 | .23** | .09** | −.08** | .05* | -------- | .21** | .32** | .16** |
| 10. GR - practical prov. | −.28 (.41) | −.31 (.25) | −.02 | −.01 | −.05 | − 14** | 11** | .27** | .05* | .17** | .13** | -------- | .16** | .31** |
| 11. GR - emotional receipt | −.26 (.43) | −.33 (.23) | −.01 | −.02 | .12** | .05* | .13** | .02 | .05* | −.02 | .28** | .07** | -------- | .33** |
| 12. GR - emotional prov. | −.29 (.41) | −.29 (.26) | −.00 | −.12** | .00 | 14** | .05 | .16** | −.01 | .07** | .13** | .29** | .17** | -------- |
Note: Numbers below the diagonal represent correlations between daily reports (without correcting for repeated measures). Numbers above the diagonal represent correlations among the person-level summaries of these variables. For each type of support, responses were effect coded (−.5 = no; .5 = yes). CG=context-general; SR=stress-related; GR=goal-related
p < .05;
p < .01
To assess support-visibility for each form of support, we compared one couple-members’ support receipt response on a given day with his/her partner’s support provision response on the same day. We classified a person as receiving invisible support when they reported that they did not receive a form of support that their spouse reported providing; we classified a person as receiving visible support when they reported receiving a form of support that their spouse also reported providing; we classified a person as experiencing a support absent day when they reported not receiving a form of support that their spouse also reported not providing; and we classified a person as experiencing a false positive when they reported that they received a form of support that their spouse did not report providing. See Supplementary Table 1, in the supplementary materials, for the percent of participant days classified as each type of support-visibility.
Daily outcome measures.
Participants completed several items using slider scales ranging from 1 (not at all) to 100 (very much). We measured daily self-efficacy with one item: “How competent and capable do you feel today.” To measure daily distress, participants completed three items related to anxiety: “How nervous or anxious do you feel today,” “How tense do you feel today,” and “How calm or relaxed do you feel today” (reverse-scored) and three items related to depression: “How sad or blue do you feel today,” “How depressed do you feel today,” and “How cheerful do you feel today?” (reverse-scored). The reliability for the daily distress measure was good (between-person reliability = .86; reliability of change = .83; Cranford et al., 2006; Shrout & Lane, 2012). Participants completed eight items to assess their daily relationship quality. An example item is “How close or connected do you feel to your spouse today” (between-person reliability = .85; reliability of change = .83). Finally, measured daily coping with the item, “To what extent did you feel that you could effectively cope with stressful life events today?,” and we measured daily goal progress with the item, “How much progress did you make on achieving your goals for today?”
Number of stressors.
Because the actual experience of stress could impact participants’ responses to the stress-related support items, we assessed and controlled for number of stressors. Participants were asked whether 43 experiences happened to them on each day, ranging from common stressors (“Did you get bad news today,” “Did you perform poorly on a task today”) to potentially positive experiences (“Did you visit a new place today”). For each experience, participants indicated whether it had happened and classified experiences they selected as either very negative, mildly negative, neutral, mildly positive, or very positive. To assess the degree of stress an individual experienced on a given day, we computed a sum by assigning 1 point for experiences rated “mildly negative” and 2 points for experiences rated “very negative.”
Data analytic strategy.
To account for the non-independence in these data (i.e., data were collected from couple-members at multiple time-points), we constructed a multilevel model. Following recommendations for data in which time and partner are crossed (i.e., couple-members provided their reports at the same time-points; e.g., Kenny, Kashy, & Cook, 2006), this model allowed for correlated errors between spouses (distinguished by sex) and allowed for different error variances for the two couple-members (set to be equal at each time point). We analyzed the data using the mixed models function in SPSS 24.0. We modeled separate intercepts for husbands and wives and allowed for them to be correlated (random statement). We did not include random slopes for the support processes because the results were numerically unstable with only seven time points (i.e., the models did not converge). We also specified the two-level cross-structure of these data, in which time and partner are crossed (repeated statement). Thus, we were able to estimate the interclass correlation as the time-specific correlation between couple-members’ outcome scores (see supplementary materials for model specification syntax).
Our primary hypotheses concerned changes in outcomes as a function of whether a participant reported receiving support and whether the participants’ spouse reported providing support on a given day. We included the previous day’s outcome in each model, and we controlled for average support receipt and spouse support provision throughout the week by including the individual’s average (grand-mean centered). Of primary interest, we included the individual’s report of support receipt (−.5 = no; .5 = yes), the spouse’s report of support provision (−.5 = no, .5 = yes), and their interaction. These daily support predictors were person-centered so that we could separate between-person and within-person effects. In each context (i.e., context-general support, stress-related support, goal-related support), we evaluated practical and emotional support in separate models due to shared variation in between- and within-person emotional and practical support (see Table 2).
In sum, these analyses allowed us to predict how an individual’s own report of support receipt, their spouse’s report of support provision, and their interaction on a given day (T1) were related to changes in outcomes from the previous day (T0) to the current day (T1). In this analysis, main effects indicate an overall effect of perceiving support receipt or partner reporting support provision, and an interaction between support receipt and partner support provision may indicate a difference between visible and invisible support specifically (see Figure 1). When we observed significant interactions, we probed the interaction to assess differences in outcomes on visible and invisible support days.
To further test this direct comparison, we also conducted quasi-signal-detection analyses in which we predicted changes in outcomes from the participant’s support visibility classification (i.e., visible, invisible, absent, or false positive) for the same day.4 Although the main effect analysis is consistent with most past research (e.g., Belcher et al., 2011; Bolger et al., 2000; Gleason et al., 2008; Shrout et al., 2010), there is also precedent for the direct comparison of the four support classifications (e.g., Gable et al., 2003; Maisel & Gable, 2009; Girme et al., 2018). Full results of these quasi-signal-detection analyses (and their model specification information) are presented in supplementary materials.
Finally, we conducted lagged analyses in which we predicted changes in outcomes from the previous day (T0) to the current day (T1) from the previous day’s (T0) support receipt, partner support provision, and their interaction. Although the concurrent analyses directly tested our hypotheses (and fit with our expectation that support would impact well-being immediately), these additional lagged analyses can more directly test directionality (e.g., Laurenceau & Bolger, 2005). Lagged models included the outcome on the previous day (T0), an individual’s average support receipt and partner support provision throughout the week (grand-mean centered), and the change in support receipt and partner support provision from the previous day (T0) to the current day (T1) to predict current day outcomes (see supplementary materials for complete model specification syntax). It is necessary to include the change in predictors as a control because person-centering predictors across a limited number of observations creates an unintended correlation between the person-centered predictor on a given day and the change in the predictor from the previous to the current day. Specifically, higher-than-average predictor days are more likely to be followed by lower-than-average predictor days than other options because the weekly average is based on a limited number of observations (see Jakubiak & Feeney, 2016, Lemay, Clark, & Feeney, 2007; Lemay & Neal, 2013).
Results
Descriptive statistics and zero-order correlations.
Descriptive statistics and zero-order correlations among support variables are presented in Table 2. Descriptive statistics and zero-order correlations among all outcome and control variables are presented in Table 3. Results from the primary analyses are described next and presented in Tables 4–9.
Table 3.
Descriptive statistics and zero-order correlations between outcome and control variables
| Variable | Mean (SD) for daily observations | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|---|
| Study 1: Young adult sample | |||||||
| 1. Daily self-efficacy | 79.5 (19.1) | -------- | −.70** | .48** | .71** | .42** | −.26** |
| 2. Daily distress | 29.8 (20.0) | −.60** | -------- | −.50** | −.75** | −.34** | .38** |
| 3. Daily relationship quality | 89.0 (13.4) | .41** | −.46** | -------- | .49** | .20** | −.18** |
| 4. Daily coping | 72.7 (21.8) | .56** | −.63** | .41** | -------- | .44** | −.29** |
| 5. Daily goal progress | 57.8 (26.5) | .33** | −.30** | .20** | .35** | -------- | −.13** |
| 6. Daily number of stressors | 3.4 (4.5) | −.34** | .46** | −.22** | −.38** | −.18** | -------- |
| Study 2: Aging sample | |||||||
| 1. Daily self-efficacy | 86.7 (17.7) | -------- | −.67** | .54** | .73** | .32** | −.24** |
| 2. Daily distress | 18.0 (17.8) | −.56** | -------- | −.51** | −.72** | −.27** | .29** |
| 3. Daily relationship quality | 91.8 (13.4) | .44** | −.45** | -------- | .45** | .20** | −.12** |
| 4. Daily coping | 83.8 (20.0) | .54** | −.58** | .37** | -------- | .35** | −.24** |
| 5. Daily goal progress | 60.1 (28.0) | .24** | −.25** | .17** | .29** | -------- | −.02 |
| 7. Daily number of stressors | 2.1 (3.8) | −.33** | .45** | −.20** | −.35** | −.10** | -------- |
Note: Numbers below the diagonal represent correlations between daily reports (without correcting for repeated measures). Numbers above the diagonal represent correlations among the person-level summaries of these variables
p < .05;
p < .01
Table 4.
Context-general support receipt and provision results (same-day)
| Self-efficacy | Distress | Relationship quality | ||||
|---|---|---|---|---|---|---|
| Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | |
| Study 1: Young adult sample | B | B | B | B | B | B |
| Outcome lagged | .15** [.10, .19] | .14** [.10, .18] | .31** [.27, .35] | .34** [.30, .38] | .38** [.33, .42] | .38* [.33, .42] |
| Person average (GMC) | ||||||
| Support receipt | 6.29** [1.81, 10.77] | 7.78** [3.39, .18] | −4.73* [−8.67, −.79] | −4.69* [−8.43, −.95] | 6.06** [3.74, 8.38] | 4.92** [2.59, 7.25] |
| Partner support prov. | 3.21 [−1.23, 7.66] | −2.26 [−6.73, 2.21] | −4.46* [−8.40, −.52] | 1.44 [−2.31, 5.19] | 2.96* [.65, 5.28] | 1.15 [−1.16, 3.46] |
| Daily (PC) | ||||||
| Support receipt | 4.36** [2.70, 6.02] | 4.10** [2.34, 5.86] | −5.23** [−6.80, −3.66] | −4.01** [−5.71, −2.31] | 3.60** [2.52, 4.68] | 4.09** [2.94, 5.23] |
| Partner support prov. | 1.28 [−.36, 2.93] | .28 [−1.44, 2.01] | −3.43** [−4.99, −1.87] | −.65 [−2.31, 1.01] | 1.99** [.92, 3.06] | 1.57** [.45, 2.69] |
| Receipt*provision | .48 [−3.82, 4.77] | 1.50 [−3.09, 6.08] | 3.30 [−.76, 7.36] | 1.31 [−3.08, 5.71] | −.68 [−3.43, 2.08] | .95 [.−1.98, 3.90] |
| Variance Components | ||||||
| Residual var. for men | 233,6** [211.9, 257.4] | 234.0** [212.4, 257.8] | 233.8** [211.4, 258.6] | 240.0** [216.6, 265.9] | 125.2** [111.9, 140.0] | 124.9** [111.9, 139.5] |
| Residual var. for women | 212.1** [191.8, 234.5] | 211.9** [191.6, 234.3] | 178.3** [160.6, 197.9] | 190.4** [171.2, 211.7] | 82.8** [74.5, 92.1] | 82.1** [73.8, 91.5] |
| Partial ICC between spouses | .12** [.05, .18] | .12** [.05, .19] | .17** [.10, .23] | .19** [.13, .26] | .27** [.20, .33] | .25** [.18, .32] |
| Intercept var. for men [1,1] | 98.5** [70.1, 138.3] | 99.7** [71.4, 139.3] | 90.3** [62.2, 131.0] | 79.13** [52.3, 119.7] | 30.2** [18.8, 48.5] | 32.5** [20.5, 51.6] |
| Intercept var. for women [2,2] | 115.9** [83.8, 160.4] | 123.0** [89.4, 169.2] | 70.7** [45.5, 110.1] | 70.8** [44.2, 113.5] | 15.5* [6.3, 37.8] | 18.0* [8.1, 40.1] |
| Intercept covariance [1,2] | 5.51 [−16.0, 27.0] | 8.7 [−13.2, 30.6] | 18.1* [.32, 35.9] | 22.4* [4.5, 40.3] | 7.4* [.4, 14.4] | 8.6* [1.2, 16.0] |
| Study 2: Aging sample | B | B | B | B | B | B |
| Outcome lagged | .23** [.18, .27] | 41** [.36, .46] | .12** [.08, .16] | .20** [.15, .25] | .45** [.41, .48] | .61** [.57, .66] |
| Person average (GMC) | ||||||
| Support receipt | 4.59** [1.28, 7.90] | 6.24** [2.98, 9.50] | −8.64** [−12.51, −4.78] | −12.60** [−16.65, −8.55] | 3.83** [2.26, 5.40] | 3.67** [2.13, 5.21] |
| Partner support prov. | −.12 [−3.54, 3.30] | 1.76 [−1.53, 5.05] | 0.23 [−4.15, 3.69] | −.36 [−4.49, 3.78] | 0.04 [−1.59, 1.50] | .64 [−.89, 2.16] |
| Daily (PC) | ||||||
| Support receipt | 3.88** [2.27, 5.48] | 3.62** [1.41, 5.84] | −4.54** [−6.01, −3.08] | −3.95** [−5.99, −1.92] | 1.12** [.35, 1.88] | 1.92** [.81, 3.03] |
| Partner support prov. | 2.85** [1.22, 4.49] | −.89 [−3.07, 1.30] | −2.41** [−3.89, −.93] | −1.26 [−3.29, .76] | 1.29** [.50, 2.07] | .88 [−.24, 2.00] |
| Receipt*provision | .62 [−4.02, 5.25] | −4.46 [−10.74, 1.81] | 3.23 [−1.06, 7.53] | 8.09** [2.10, 14.08] | −2.65* [−4.86, −.45] | −3.16* [−6.26, −.05] |
| Variance Components | ||||||
| Residual var. for men | 178.0** [159.5, 198.6] | 203.7** [174.0, 238.4] | 146.4** [131.1, 163.5] | 177.1** [153.9, 203.8] | 56.2** [48.5, 65.0] | 72.4** [64.6, 81.0] |
| Residual var. for women | 132.6** [117.4, 149.7] | 163.2** [138.2, 192.7] | 108.5** [97.5, 120.8] | 129.8** [113.4, 148.5] | 25.3** [21.4, 29.9] | 27.6** [24.0, 31.8] |
| Partial ICC between spouses | .02 [−.07, .11] | .03 [−.08, .14] | .11* [.02, .20] | .11† [−.00, .21] | .04 [−.07, .14] | .12† [−.01, .23] |
| Intercept var. for men [1,1] | 79.4** [54.9, 114.7] | 32.1* [12.6, 83.7] | 143.6** [112.2, 183.7] | 109.4** [78.9, 151.8] | 22.9** [13.1, 39.9] | 8.13** [4.3, 15.2] |
| Intercept var. for women [2,2] | 81.9** [56.0, 118.9] | 47.7* [21.7, 104.5] | 139.5** [106.7, 182.5] | 131.9** [94.2, 184.7] | 16.1* [6.4, 40.1] | -- |
| Intercept covariance [1,2] | 21.45* [4.1, 39.0] | 14.3† [−1.8, 30.5]] | 53.8** [27.9, 79.7] | 46.1** [17.5, 74.8] | 14.3** [6.0, 22.5] | -- |
Note: Primary results are bolded. Estimates are pooled across husbands/wives; unstandardized beta weights; person average predictors are grand-mean-centered (GMC); daily predictors are person-centered (PC); 95% confidence intervals for significant effects are shown in brackets. Where only one variance component is included, the model included only an overall random intercept because the results were numerically unstable when separate intercepts were modeled for men and women.
< .10;
p < .05;
p < .01
Table 9.
Goal-related support receipt and provision results (lagged)
| Self-efficacy | Distress | Relationship quality | Goal progress | |||||
|---|---|---|---|---|---|---|---|---|
| Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | |
| Study 1: Young adult sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .15** [.10, .19] | .15** [.10, .19] | .39** [.35, .43] | .39** [.35, .43] | .56** [.52, .60] | .55** [.51, .59] | .26** [.21, .30] | .24** [.20, .28] |
| Person average (GMC) | ||||||||
| Support receipt | 2.60 [−2.82, 8.02] | 3.02 [−1.81, 7.85] | 1.97 [−2.25, 6.20] | −3.14 [−6.97, .69] | 1.84† [−.25, 3.93] | 3.40** [1.48, 5.32] | 15.05** [9.04, 21.06] | 14.9** [9.47, 20.3] |
| Partner support prov. | −4.67† [−10.1, .72] | −1.11 [−573, 3.50] | 3.53 [−.70, 7.77] | .93 [−.2.68, 4.54] | .58 [−1.52, 2.67] | .07 [−1.70, 1.85] | −5.44† [−11.36, .39] | −.17 [−5.27, 4.93] |
| Daily (PC) | ||||||||
| Change in support receipt | 1.05 [−1.03, 3.13] | 2.42** [.61, 4.23] | −.83 [−2.85, 1.18] | −1.50† [−3.25, .25] | 2.16** [.73, 3.59] | 1.81** [.57, 3.05] | 2.71† [−32, 5.74] | 8.56** [5.96, 11.16] |
| Change in support provision | .15 [−1.87, 2.18] | .38 [−1.48, 2.25] | −.03 [−2.00, 1.94] | .96 [−.84, 2.76] | 1.52* [.12, 2.91] | .59 [−.68, 1.86] | 2.27 [−2.41, 6.09] | 2.83* [.14, 5.52] |
| Support receipt lagged | 1.15 [−1.77, 4.09] | 1.42 [−1.17, 4.00] | 2.75† [−.10, 5.59] | .49 [−2.02, 3.00] | 1.59 [−.42, 3.60] | .51 [−1.27, 2.29] | 1.84 [−2.41, 6.09] | 8.57** [4.85, 12.30] |
| Partner support prov. lagged | −.37 [−3.25, 2.51] | .65 [−2.05, 3.36] | −.19 [−2.99, 2.60] | .22 [−2.39, 2.84] | .77 [−1.21, 2.74] | 1.26 [−.59, 3.11] | 3.11 [−1.07, 7.27] | 1.82 [−2.07, 5.71] |
| Receipt*provision lagged | −3.67† [−8.03, .70] | −.14 [−4.40, 4.13] | −1.31 [−5.50, 2.87] | −.60 [−4.7−, 3.49] | −1.08 [−3.93, 1.77] | .62 [−2.16, 3.39] | −4.60 [−10.79, 1.60] | −2.83 [−8.81, 3.15] |
| Variance Components | ||||||||
| Residual var. for men | 237.4** [215, 262] | 237.7** [215, 262] | 249.4** [224, 278] | 249.6** [224, 278] | 142.2** [129, 156] | 140.6** [128, 154] | 491.2** [444, 544] | 473.1** [427, 524] |
| Residual var. for women | 217.5** [196, 241] | 216.7** [196, 240] | 197.2** [176, 221] | 199.5** [178, 223] | 97.5** [88.1, 198] | 99.0** [89.5, 109] | 473.7** [427, 524] | 461.6** [417, 511] |
| Partial ICC between spouses | .13** [.07, .20] | .13** [.07, .20] | .21** [.13, .27] | .22** [.15, .29] | .25** [.19, .32] | .26** [.19, .32] | .07† [−.00, .13] | .04 [−.03, .11] |
| Intercept for men [1,1] | 102.0** [72.3, 144.0] | 104.2** [74.0, 146.7] | 63.2** [37.9, 105.7] | 65.9* [38.8, 108.5] | 6.49* [2.7, 15.5] | 5.9* [2.3, 15.0] | 82.9** [47.6, 144.4] | 95.2** [57.6, 157.4] |
| Intercept for women [2,2] | 125.0** [90.2, 173.1] | 124.5** [89.6, 173.0] | 52.9** [28.3, 98.7] | 51.8** [27.3, 98.4] | -- | -- | 122.4** [78.4, 191.2] | 118.6** [76.4, 184.3] |
| Intercept covariance [1,2] | 8.5 [−13.3, 31.3] | 9.0 [−.13.9, 31.9] | 16.2* [.3, 32.2] | 15.1† [−.83, 31.1] | -- | -- | 26.6† [−2.8, 56.0] | 26.5† [−2.6, 55.7] |
| Study 2: Aging sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .43** [.37, .50] | .30** [.25, .35] | .27** [.21, .33] | .16** [.11, .20] | .65** [.60, .70] | .59** [.55, .63] | .18** [.11, .24] | .09** [.05, .14] |
| Person average (GMC) | ||||||||
| Support receipt | −3.22 [−8.28, 1.83] | .74 [−3.89, 5.38] | −4.06 [−10.33, 2.21] | −3.17 [−9.02, 2.67] | .28 [−1.76, 2.32] | .22 [−1.52, 1.96] | 7.61 [−2.31, 17.54] | 21.25** [12.34, 30.15] |
| Partner support prov. | −1.40 [−6.49, 3.70] | −1.67 [−5.95, 2.62] | 3.32 [−2.89, 9.53] | −.19 [−5.50, 5.13] | −.13 [−2.11, 1.86] | −.44 [−2.04, 1.15] | −.23 [−10.23, 9.78] | −3.73 [−11.92, 4.46] |
| Daily (PC) | ||||||||
| Change in support receipt | .11 [−2.64, 2.86] | .04 [−2.05, 2.13] | −1.78 [−4.24, .69] | .53 [−1.33, 2.38] | 1.07 [−.26, 2.41] | −.58 [−1.54, .38] | 8.47** [4.21, 12.73] | 7.50** [4.18, 4.46] |
| Change in support provision | 2.25 [−.55, 5.04] | .91 [−1.06, 2.89] | −.97 [−3.46, 1.53] | −1.51† [−3.36, .14] | .66 [−.66, 1.99] | −.47 [−1.37, .43] | .20 [−4.15, 4.56] | 3.90* [.73, 7.07] |
| Support receipt lagged | −.88 [−4.94, 3.19] | 1.02 [−2.08, 4.11] | −1.24 [−4.89, 2.40] | −.76 [−3.50, 1.98] | .41 [−1.54, 2.37] | −.52 [−1.93, .90] | 14.72** [8.40, 21.05] | 9.80** [4.87, 14.74] |
| Partner support prov. lagged | 2.52 [−1.57, 6.61] | .23 [−2.74, 3.19] | −3.09† [−6.77, .58] | 01.26 [−3.90, 1.38] | 1.02 [−.91, 2.96] | −1.21† [−2.55, 1.36] | 1.53 [−4.89, 7.95] | 3.63 [−1.15, 8.42] |
| Receipt*provision lagged | −6.09 [−14.15, 1.97] | −.78 [−6.43, 4.87] | −2.42 [−9.91, 5.08] | 1.98 [−3.06, 7.03] | −1.56 [−5.34, 2.22] | −.89 [−3.38, 1.59] | −1.09 [−13.96, 11.78] | −5.31 [−14.51, 3.88] |
| Variance Components | ||||||||
| Residual var. for men | 206.6** [167, 236] | 191.6** [169., 216] | 163.7** [137, 196] | 154.9** [137, 176] | 87.5** [74.8, 102] | 58.1** [51.9, 65.2] | 397.6** [336, 471] | 403.22** [357, 455] |
| Residual var. for women | 179.7** [143, 225] | 144.1** [125, 166] | 137.5** [117, 162] | 111.3** [98.5, 126] | 34.2** [26.8, 43.5] | 26.0** [22.8, 29.7] | 503.2** [430, 589] | 435.3** [386, 491] |
| Partial ICC between spouses | −.03 [−.17, .12] | −.01 [−.19, .10] | .15* [.00, .29] | .13* [.02, .23] | .27** [.08, .45] | .02 [−.12, .15] | .12† [−.01, .25] | .14** [.04, .23] |
| Intercept for men [1,1] | 32.2 [7.0, 149.0] | 61.6** [37.7, 100.7] | 110.1** [71.8, 168.7] | 149.3** [112. 1, 198.7] | 2.06 [.07, 65.3] | 7.0** [3.9, 12.5] | 213.8** [131.9, 346.4] | 244.6** [174.8, 342.1] |
| Intercept for women [2,2] | 35.6 [8.6, 147.3] | 55.2** [32.3, 94.2] | 128.3** [82.1, 200.6] | 142.0** [104.5, 193.0] | -- | -- | 275.2** [179.1, 423.0] | 279.2** [203.7, 382.7] |
| Intercept covariance [1,2] | 12.3 [−6.7, 31.2] | 18.3* [1.5, 35.0] | 57.8** [20.6, 85.0] | 67.8** [36.3, 99.2] | -- | -- | 48.2 [−28.7, 125.1] | 40.0 [−19.6, 99.5] |
Note: Primary results are bolded. Estimates are pooled across husbands/wives; unstandardized beta weights; person average predictors are grand-mean-centered (GMC); daily predictors are person-centered (PC); 95% confidence intervals for significant effects are shown in brackets. Where only one variance component is included, the model included only an overall random intercept because the results were numerically unstable when separate intercepts were modeled for men and women.
< .10;
p < .05;
p < .01
Is context-general support visibility related to daily personal and relational outcomes?
Based on the theory that receiving practical support has costs that invisible support can avoid, we predicted that people would report better individual and relational well-being when they did not report receiving support. Contrary to the proposed benefits of invisible practical support, participants reported greater increases in self-efficacy and relationship quality and greater decreases in distress on days when they reported receiving high levels of context-general practical support, compared to their typical level of support receipt (see Table 4). Participants reported the same uniformly positive individual and relational outcomes when they reported receiving context-general emotional support (see Table 4). Additionally, spouses’ context-general practical support provision independently predicted decreases in distress, and both practical and emotional support provision predicted increases in relationship quality. The secondary quasi-signal-detection analysis also showed a relative benefit of visible compared to invisible support days for all outcomes (see supplementary tables). In summary, young adults experienced better individual and relational well-being on days that they reported receiving greater-than-average levels of context-general practical and emotional support. Contrary to the proposed benefits of invisible support, these results suggest benefits of perceiving context-general support.
Lagged analyses.
In addition to the same-day analyses, we evaluated links between context-general support on one day and changes in individual and relational outcomes on the following day. Young adult participants reported greater self-efficacy and relationship quality and less distress on days after they reported receiving high levels of context-general practical support (see Table 5). Additionally, participants reported greater self-efficacy and marginally greater relationship quality on days after they reported receiving context-general emotional support (see Table 5).
Table 5.
Context-general support receipt and provision results (lagged)
| Self-efficacy | Distress | Relationship quality | ||||
|---|---|---|---|---|---|---|
| Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | |
| Study 1: Young adult sample | B | B | B | B | B | B |
| Outcome lagged | .15** [.10, .19] | .14** [.09, .18] | .36** [.32, .40] | .38** [.34, .42] | .53** [.49, .57] | .55** [.51, .58] |
| Person average (GMC) | ||||||
| Support receipt | 6.45** [1.92, 10.99] | 7.50** [3.02, 11.99] | −4.56* [−8.26, −.86] | −4.24* [−7.80, −.67] | 5.28** [3.46, 7.10] | 3.64** [1.86, 5.42] |
| Partner support prov. | 3.13 [−1.38, 7.63] | −2.13 [−6.69, 2.44] | −3.96* [−7.68, −.25] | 1.69 [−1.88, 5.26] | 1.86* [.09, 3.63] | .27 [−1.46, 2.00] |
| Daily (PC) | ||||||
| Change in support receipt | 4.33** [2.64, 6.01] | 3.90** [2.10, 5.69] | −5.09** [−6.71, −3.47] | −3.56** [−5.31, −1.81] | 3.50** [2.34, 4.65] | 3.74** [2.51, 4.97] |
| Change in support provision | 1.47† [−.22, 3.16] | .40 [−1.34, 2.14] | −3.35** [−4.97, −1.73] | −.99 [−2.69, .71] | 1.82** [.66, 2.98] | 1.67** [.47, 2.87] |
| Support receipt lagged | 3.96** [1.49, 6.43] | 3.45* [.81, 6.08] | −3.68** [−6.05, −1.30] | −.81 [−3.38, 1.76] | 2.65** [.96, 4.33] | 1.65† [−.16, 3.46] |
| Partner support prov. lagged | .89 [−1.69, 3.48] | 1.25 [−1.28, 3.79] | −1.77 [−4.26, .71] | −1.15 [−3.62, 1.33] | 1.44 [−.33, 3.22] | 1.59† [−.16, 3.33] |
| Receipt*provision lagged | 2.72 [−1.57, 7.02] | −3.12 [−7.59, 1.34] | −2.08 [−6.15, 1.99] | −.04 [−4.33, 4.26] | 2.60† [−.17, 5.38] | −1.49 [−4.41, 1.43] |
| Variance Components | ||||||
| Residual var. for men | 235.2** [213.1, 259.5] | 234.6** [212.6, 258.9] | 240.3** [216.2, 267.0] | 246.0** [221.0, 273.8] | 136.2** [124.3, 149.1] | 136.9** [124.9, 149.9] |
| Residual var. for women | 212.5** [191.7, 235.5] | 212.3** [191.7, 235.1] | 183.2** [163.9, 204.9] | 193.9** [173.2, 217.0] | 93.9** [85.0, 103.7] | 93.3** [84.3, 103.1] |
| Partial ICC between spouses | .12** [.05, .18] | .12** [.06, 19] | .16** [.09, .23] | .19** [.12, .26] | .23** [.17, .29] | .22** [.15, .28] |
| Intercept var. for men [1,1] | 100.0** [70.9, 141.1] | 103.6** [74.1, 144.9] | 72.3** [45.5, 115.1] | 65.0** [39.3, 107.5] | 6.0* [2.5, 14.1] | 6.9* [3.1, 15.2] |
| Intercept var. for women [2,2] | 117.4** [84.3, 163.6] | 127.1** [92.1, 175.4] | 53.5** [29.9, 95.9] | 54.6** [29.8, 100.3] | -- | -- |
| Intercept covariance [1,2] | 5.6 [−16.5, 27.7] | 8.89 [−14.0, 31.8] | 13.4† [−2.24, 29.0] | 17.0* [0.9, 33.1] | -- | -- |
| Study 2: Aging sample | B | B | B | B | B | B |
| Outcome lagged | .30** [.25, .35] | .37** [.03, .43] | .14** [.09, .18] | .26** [.20, .32] | .57** [.53, .61] | .60** [.55, .65] |
| Person average (GMC) | ||||||
| Support receipt | 4.84** [1.46, 8.23] | 6.75** [2.45, 11.05] | −10.43** [−14.63, −6.22] | −12.85** [−17.64, −8.05] | 3.21** [1.91, 4.51] | 3.88** [.55, .65] |
| Partner support prov. | .17 [−3.34, 3.68] | 1.30 [−3.08, 5.68] | −1.21 [−5.47, 3.06] | 1.39 [−3.54, 6.31] | .65 [−.68, 1.98] | −.02 [−1.75, 1.71] |
| Daily (PC) | ||||||
| Change in support receipt | 3.94** [2.13, 5.76] | 3.57** [.88, 6.26] | −4.27** [−5.91, −2.64] | −2.95* [−5.37, −.53] | .93* [.10, 1.77] | .85 [−.41, 2.11] |
| Change in support provision | 3.51** [1.63, 5.38] | −.68 [−3.33, 1.97] | −2.89** [−4.57, −1.21] | −1.33 [−3.74, 1.09] | 1.27** [.39, 2.15] | .08 [−1.21, 1.37] |
| Support receipt lagged | 2.64† [−.01, 5.30] | 1.59 [−2.36, 5.54] | −3.27** [−5.67, −.87] | −1.17 [−4.75, 2.40] | .43 [−.79, 1.66] | .55 [−1.31, 2.41] |
| Partner support prov. lagged | 5.12** [2.34, 7.91] | .30 [−3.60, 4.19] | −3.31* [−5.8−, −.81] | −2.10 [−5.66, 1.45] | 1.94** [.64, 3.24] | .71 [−1.20, 2.62] |
| Receipt*provision lagged | .75 [−4.03, 5.52] | .67 [−6.06, 7.41] | 5.31* [.96, 0.65] | −2.51 [−8.80, 3.77] | −2.06† [−4.19, .08] | 1.74 [−1.43, 4.91] |
| Variance Components | ||||||
| Residual var. for men | 187.6** [166, 212] | 193.1** [158, 235] | 144.7** [128, 164] | 162.9** [136, 195] | 56.0** [50.0, 62.8] | 83.0** [71.1, 96.9] |
| Residual var. for women | 138.2** [120, 159] | 171.0** [139, 211] | 109.7** [97.1, 124] | 135.7** [115, 160] | 25.6** [22.4, 29.2] | 32.1** [25.8, 40.0] |
| Partial ICC between spouses | −.03 [−.12, .07] | −.02 [−.17, .12] | .15** [.01, .29] | .01 [−.12, .14] | .24** [−.06, .41] | |
| Intercept var. for men [1,1] | 60.8** [37.6, 98.5] | 42.66† [14.5, 125.5] | 132.0** [106.7, 189.0] | 93.4** [60.4, 144.4] | 6.9** [3.9, 12.3] | 4.2 [.8, 22.8] |
| Intercept var. for women [2,2] | 53.5** [31.5, 90.9] | 48.2* [18.3, 127.0] | 130.8** [96.3, 177.9] | 112.6** [70.8, 179.2] | -- | -- |
| Intercept covariance [1,2] | 16.7* [.65, 32.8] | 16.0 [−5.2, 37.1] | 59.1** [29.8, 88.3] | 48.9** [15.8, 82.0] | -- | -- |
Note: Primary results are bolded. Estimates are pooled across husbands/wives; unstandardized beta weights; person average predictors are grand-mean-centered (GMC); daily predictors are person-centered (PC); 95% confidence intervals for significant effects are shown in brackets. Where only one variance component is included, the model included only an overall random intercept because the results were numerically unstable when separate intercepts were modeled for men and women.
< .10;
p < .05;
p < .01
Is stress-related support visibility related to daily personal and relational outcomes?
Prior to assessing our hypotheses, we evaluated links between daily stressors and personal and relational outcomes. On days when young adult participants reported experiencing greater/more severe daily stressors, they reported decreases in self-efficacy, relationship quality, and perceived coping, and they reported increases in distress (see Table 6). Therefore, we conducted our primary analyses controlling for daily stressors.
Table 6.
Stress-related support receipt and provision results, controlling for daily stressors (same-day)
| Self-efhcacy | Distress | Relationship quality | Coping | |||||
|---|---|---|---|---|---|---|---|---|
| Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | |
| Study 1: Young adult sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .11** [.07, .15] | .12** [.08, .16] | .23** [.20, .27] | .23** [.20, .27] | .31** [.27, .36] | .31** [.26, .35] | .14** [.10, .18] | .14** [.11, .18] |
| Stress | −1.54** [−1.71, −1.38] | −1.51** [−1.67, −1.34] | 1.79** [1.64, 1.94] | 1.79** [1.64, 1.94] | −.72** [−.82, .−.61] | −.74** [−.85, −.63] | −1.85** [−2.04, −1.66] | −1.84** [−2.02, −1.65] |
| Person average (GMC) | ||||||||
| Support receipt | 5.15† [−.08, 10.38] | 3.08 [−1.53, 7.69] | .30 [−4.31, 4.92] | .73 [−3.37, 4.83] | 4.77** [1.7−, 7.85] | 4.52** [1.77, 7.28] | 2.10 [−3.39, 7.60] | .87 [−3.96, 5.69] |
| Partner support prov. | −4.69† [−9.59, 10.38] | −1.66 [−6.15, 2.83] | 3.75† [−.67, 8.18] | −.26 [−4.26, 3.75] | 1.22 [−1.73, 4.16] | 2.67† [−.02, 5.36] | −1.66 [−6.90, 3.59] | 1.87 [−2.82, 6.56] |
| Daily (PC) | ||||||||
| Support receipt | .49 [−1.39, 2.28] | −1.61* [−3.21, −.01] | .60 [−1.02, 2.23] | .69 [−.77, 2.14] | 3.16** [1.97, 4.34] | 2.93** [1.87, 4.00] | −1.53 [−3.63, .57] | −1.69† [−3.57, .19] |
| Partner support prov. | .29 [−1.54, 2.13] | −.98 [−2.55, .59] | .92 [−.76, 2.59] | 1.42† [−.01, 2.86] | .35 [−.87, 1.57] | .49 [−.56, 1.54] | −.82 [−2.98, 1.34] | −1.84† [−3.69, .01] |
| Receipt*provision | 3.56 [−1.29, 8.81] | −4.49* [−8.63, −.35] | .62 [−3.96, 5.20] | .52 [−.33, 4.29] | .47 [−2.85, 3.79] | .67 [−2.09, 3.42] | .95 [−4.95, 6.85] | −1.57 [−6.41, 3.27] |
| Variance Components | ||||||||
| Residual var. for men | 197.7** [179.5, 217.7] | 196.5** [178, 216] | 181.5** [164, 200 | 180.8** [164, 200] | 114.9** [103, 127] | 112.7** [101, 125] | 297.1** [269.6, 327.4] | 296.2** [268.8, 326.4] |
| Residual var. for women | 183.8** [166.3, 203.1] | 184.3** [166, 204] | 139.7** [127, 154] | 139.9** [127, 155] | 73.4** [66.3, 81.3] | 73.1** [66.0, 80.9] | 234.6** [212.7, 258.8] | 234.8** [212.9, 259.1] |
| Partial ICC between spouses | .06† [−.00, .13] | .07† [−.00, .13] | .10** [.03, .17] | .11** [.04, .17] | .26** [.19, .32] | .25** [.18, .31] | .04 [−.03, .11] | .05 [−.02, .12] |
| Intercept for men [1,1] | 98.1** [81.7, 134.3] | 97.7** [71.3, 133.9] | .89.9** [66.2, 122.0] | 91.4** [67.3, 124.0] | 42.2** [29.6, 60.2] | 43.2** [30.4, 61.3] | 96.4** [69.1, 134.5] | 94.7** [67.6, 132.7] |
| Intercept for women [2,2] | 122.5** [91.2, 164.5] | 122.0** [91.0, 163.6] | 90.8** [65.1, 126.5] | 90.5** [64.8, 126.3] | 31.5** [19.6, 50.7] | 31.2** [19.6, 49.7] | 127.4** [91.9, 176.8] | 126.1** [90.8, 175.2] |
| Intercept covariance [1,2] | 15.6 [−5.9 37.1] | 12.8 [−8.2, 33.8] | 27.3** [9.1, 45.6] | 25.3** [6.8, 43.7] | 11.3** [2.7, 19.9] | 9.8* [1.1, 18.1] | 17.7 [−6.6, 41.9] | 13.9 [−9.9, 37.6] |
| Study 2: Aging sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .47** [.3, .46] | .32** [.27, .36] | .20** [.15, .24] | .15** [.10, .19] | .74** [.71, .78] | .57** [.53, .61] | .20** [.14, .25] | .14** [.09, .19] |
| Stress | −1.00** [−1.21, −.80] | −1.23** [−1.42, −1.04] | 1.53** [1.33, 1.74] | 1.50** [1.32, 1.68] | −.24** [−.36, −.12] | ** | −1.65** [−1.92, −1.38] | −1.46** [−1.70, −1.22] |
| Person average (GMC) | ** | |||||||
| Support receipt | 2.06 [−1.09, 5.21] | .22 [−3.08, 3.52] | −.21 [−4.13, 3.70] | 5.76** [1.55, 9.97] | 2.06* [.42, 3.70] | 2.17** [.58, .3.74] | 3.02 [−1.42, 7.46] | −7.33** [−11.89, −2.78] |
| Partner support prov. | −2.80 [−6.35, 5.21] | −4.09* [−7.56, −.63] | 4.41† [.00, 8.81] | 4.29† [−.13, 8.71] | −.80 [−2.62, 1.02] | −.87 [−2.47, 3.74] | −6.65* [−11.67, −1.63] | −3.06 [−7.85, 1.2] |
| Daily (PC) | ||||||||
| Support receipt | .56 [−1.41, 2.53] | −1.14 [−3.05, .76] | .26 [−1.43, 2.04] | 2.29** [.65, 3.93] | .87 [−.37, 2.11] | .97† [−.01, 1.95] | −.22 [−2.79, 2.35] | −2.30† [−4.63, .02] |
| Partner support prov. | −1.10 [−3.17, .98] | −2.29* [−4.25, −.32] | 2.28* [.38, 4.17] | 1.39 [−.32, 3.11] | −.34 [−1.64, .97] | .29 [−.74, 1.31] | 1.39 [−1.32, 4.10] | −1.74 [−4.14, .65] |
| Receipt*provision | 1.28 [−4.29 6.85] | −1.51 [7.10, 4.09] | 2.47 [−2.83, 7.77] | 1.14 [−3.91, 6.19] | 2.69 [−.62, 6.01] | −1.57 [−3.35, 1.32] | −.98 [−8.30, 6.33] | −3.85 [−10.77, 3.08] |
| Variance Components | ||||||||
| Residual var. for men | 194.6** [167, 227] | 188.8** [164., 217] | 163.7** [142, 188] | 137.6** [121, 157] | 81.3** [72.5, 91.3] | 63.8** [54.9, 74.3[ | 265.3** [230, 305] | 235.5** [206, 269] |
| Residual var. for women | 144.6** [125, 167] | 142.5** [123, 164] | 1138** [100, 129] | 105.4** [93.2, 119] | 63.4** [56.9, 70.6] | 27.7** [23.6, 32.2] | 287.1** [253, 326] | 245.4** [217, 278] |
| Partial ICC between spouses | .06 [−.06, .18] | .01 [−.10, .11] | .16** [.06, .27] | .13* [.03, .24] | .21** [.11, .31] | .07 [−.05, .18] | .10† [−.01, .20] | .01 [−.01, .13] |
| Intercept for men [1,1] | 32.2* [16.4, 62.9] | 50.0** [27.9, 89.5] | 95.1** [69.3, 130.5] | 113.9** [85.9, 151.1] | .00 a | 10.1** [5.8, 17.7] | 76.1** [46.3, 124.9] | 138.3** [98.8, 193.5] |
| Intercept for women [2,2] | 23.3 [7.07, 77.0] | 28.8* [12.4, 71.6] | 79.3** [54.1, 116.4] | 122.6** [90.5, 155.2] | -- | -- | 65.1** [36.9, 114.8] | 78.8** [51.1, 121.7] |
| Intercept covariance [1,2] | 15.2* [2.90, 27.4] | 12.9† [−1.43, 27.3] | 32.5** [11.7, 53.4] | 38.3** [12.4, 64.2] | -- | -- | 15.8 [−7.1, 38.8] | 43.7* [1.51, 63.9] |
Note: Primary results are bolded. Estimates are pooled across husbands/wives; unstandardized beta weights; person average predictors and daily stressors are grand-mean-centered (GMC); daily predictors are person-centered (PC); 95% confidence intervals for significant effects are shown in brackets. Where only one variance component is included, the model included only an overall random intercept because the results were numerically unstable when separate intercepts were modeled for men and women.
< .10;
p < .05;
p < .01
We predicted that there would be a relative benefit of visible over invisible stress-related support because stress activates the desire for active, responsive spousal support. For stress-related practical support, we observed no effects for personal outcomes, but stress-related practical support receipt was related to greater increases in relationship quality (see Table 6). The quasi-signal-detection analysis also indicated a relative relational benefit for visible stress-related support days compared to invisible stress-related support days (see supplementary tables).
Similarly, young adults reported greater increases in relationship quality when they reported receiving high levels of stress-related emotional support. However, regarding individual outcomes, participants reported marginally greater decreases in coping and greater decreases in self-efficacy on days when they reported receiving emotional support (see Table 6). The main effect predicting self-efficacy was qualified by an interaction with spouse emotional support provision. When spouse support provision was high (1 SD above the mean), reporting support receipt predicts less self-efficacy (t(1690.6) = −2.75, p < .01), but on days when spouse support provision was low (1 SD below the man), reporting support receipt was unrelated to self-efficacy (t(1668.8) = −.31, p = .755). In other words, this interaction provides evidence that invisible (compared to visible) stress-related emotional support better protects self-efficacy. The quasi-signal-detection analysis showed similar personal costs and relational benefits associated with visible (compared to invisible) stress-related emotional support.5
In summary, young adults experienced poorer personal outcomes when they received visible stress-related emotional support and experienced no personal outcomes of stress-related practical support. Relationally, young adults benefitted (rated their relationships more favorably) on days when they received and noticed both stress-related practical and emotional support.
Lagged analyses.
In lagged models, participants who reported high levels of stress-related emotional support receipt on one day reported greater increases in relationship quality on the following day (see Table 7). Additionally, when spouses reported providing stress-related emotional support on one day, participants reported lower levels of coping on the following day (see Table 7). No other lagged effects reached significance.
Table 7.
Stress-related support receipt and provision results, controlling for daily stressors (lagged)
| Self-efficacy | Distress | Relationship quality | Coping | |||||
|---|---|---|---|---|---|---|---|---|
| Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | |
| Study 1: Young adult sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .11** [.07, .15] | .11** [.07, .15] | .24** [.20, .27] | .24** [.20, .28] | .52** [.48, .56] | .52** [.48, .56] | .15** [.11, .19] | .14** [.10, .18] |
| Stress | −1.55** [−1.71, −1.38] | −1.50** [−1.67, −.134] | 1.77** [1.62, 1.92] | 1.78** [1.63, 1.93] | −.61** [−.71, −.51] | −.63** [−.73, −.53] | −1.83** [−2.02, −1.64] | −1.86** [−2.05, −1.67] |
| Person average (GMC) | ||||||||
| Support receipt | 5.43* [.11, 10.77] | 2.71 [−1.99, 7.41] | .25 [−4.41, 4.89] | .74 [−3.36, 4.83] | 3.34** [1.20, 5.49] | 2.99** [1.11, 4.85] | 2.54 [−2.96, 8.05] | .84 [−4.02, 5.70] |
| Partner support prov. | −4.37† [−9.47, .73] | −1.31 [−5.89, 3.27] | 3.75† [−.69, 8.20] | −.21 [−4.21, 3.78] | .71 [−1.31, 2.73] | 1.68† [−.13, 3.49] | −1.44 [−6.67, 3.81] | 1.91 [−2.80, 6.63] |
| Daily (PC) | ||||||||
| Change in support receipt | .52 [−1.31, 2.35] | −1.45† [−3.09, .18] | .75 [−.91, 2.41] | .75 [−.74, 2.24] | 2.92** [1.61, 4.22] | 2.58** [1.40, 3.77] | −1.20† [−2.47, .07] | −.28 [−1.68, 1.11] |
| Change in support provision | −.59 [−1.27, 2.46] | −1.10 [−2.71, .50] | .97 [−.72, 2.67] | 1.45† [−.01, 2.92] | .43 [−.91, 1.77] | .13 [−1.03, 1.30] | .17 [−1.10, 1.44] | .10 [−1.35, 1.56] |
| Support receipt lagged | .82 [−1.78, 2.46] | −.62 [−3.01, 1.77] | 1.98† [−.37, 4.34] | .52 [−1.66, 2.70] | 1.52 [−.34, 3.37] | 2.62** [.89, 4.35] | −.60 [−2.61, 1.41] | .75 [−1.10, 2.61] |
| Partner support prov. lagged | 1.29 [−1.33, 3.92] | −3.12* [−5.48, −.76] | 1.80 [−.59, 4.19] | 1.77 [−.38, 3.92] | .69 [−1.19, 2.57] | −.64 [−2.34, 1.07] | −1.42 [−3.47, .63] | −2.24* [−4.11, −.38] |
| Receipt*provision lagged | −.77 [−5.08, 3.53] | .18 [−3.75, 4.10] | .44 [−3.45, 4.33] | 1.36 [−2.21, 4.92] | −1.60 [−4.53, 1.32] | −1.36 [−4.04, 1.32] | −.30 [−5.31, 4.70] | −1.79 [−6.36, 2.78] |
| Variance Components | ||||||||
| Residual var. for men | 198.5** [180, 219] | 197.2** [179, 218] | 181.2** [164, 200] | 181.9** [165, 201] | 133.1** [121, 146] | 131.7** [120, 144] | 299.1** [271, 330] | 297.3** [269, 328] |
| Residual var. for women | 184.4** [167, 204] | 183.5** [166, 203] | 139.7** [126, 155] | 140.0** [127, 155] | 86.3** [78, 96] | 88.0** [80, 97] | 237.2** [214, 262] | 237.3** [215, 262] |
| Partial ICC between spouses | .06** [−.01, .13] | .07* [.00, .14] | .10** [.04, .17] | .11** [.04, .18] | .20** [.14, .30] | .20** [.13, .26] | .05 [−.02, .12] | .04** [−.03, .11] |
| Intercept for men [1,1] | 101.9** [74.4, 139.6] | 101.3** [73.8, 139.0] | 88.5** [64.8, 120.7] | 88.3** [64.5, 120.8] | 9.7** [5.3, 17.6] | 8.1** [4.1, 15.9] | 93.0** [65.7, 131.7] | 93.4** [66.1, 132.1] |
| Intercept for women [2,2] | 126.0** [93.6, 170.0] | 128.0** [95.3, 171.8] | 88.8** [63.2, 124.8] | 87.2** [61.7, 123.3] | -- | -- | 123.2** [87.7, 173.2] | 125.0** [89.2, 175.3] |
| Intercept covariance [1,2] | 17.0 [−5.4, 39.4] | 15.1 [−7.0, 37.2] | 24.7** [6.5, 42.9] | 22.5 [4.2, 40.7] | -- | -- | 15.2 [−8.7, 39.1] | 12.3 [−11.6, 36.3] |
| Study 2: Aging sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .33** [.27, .38] | .31** [.26, .36] | .19** [.14, .24] | .21** [.16, .26] | .76** [.72, .81] | .61** [.57, .66] | .23** [.16, .30] | .28** [.21, .35] |
| Stress | −1.02** [−1.26, −.78] | −1.24** [−1.45, −1.03] | 1.42** [1.19, 1.65] | 1.42** [1.21, 1.62] | −.21** [−.35, −.07] | −.29** [−.39, −.19] | −1.69** [−2.01, −1.38] | −1.59** [−1.92, −1.26] |
| Person average (GMC) | ||||||||
| Support receipt | 1.10 [−3.33, 5.53] | 1.26 [−2.44, 4.95] | −1.26 [−6.06, 3.55] | 7.05** [2.59, 11.51] | 1.95† [−.07, 3.96] | 1.64† [−.05, 3.32] | 3.42 [−1.99, 8.82] | −4.55† [−9.87, .77] |
| Partner support prov. | −3.09 [−7.96, 1.78] | −5.59** [−9.49, −1.69] | 5.79* [.51, 11.07] | 4.32† [−.36, 8.99] | −.83 [−3.02, 1.32] | −.83 [−2.52, .88] | −5.94* [−11.78, 8.82] | −1.22 [−6.65, 4.21] |
| Daily (PC) | ||||||||
| Change in support receipt | .57 [−1.75, 2.89] | −.36 [−2.57, 1.84] | .54 [−1.53, 2.61] | 1.78† [.14, 3.70] | .52 [−.91, 1.96] | 1.01† [−.08, 2.11] | −1.46 [−3.66, .74] | 1.14 [−1.17, 3.46] |
| Change in support provision | 1.04 [−1.50, 3.57] | −2.10† [−4.37, .17] | .96 [−1.32, 3.25] | .46 [−1.56, 2.47] | −.12 [−1.69, 1.45] | .59 [−.56, 1.75] | .23 [−2.03, 2.50] | −.50 [−3.24, 2.25] |
| Support receipt lagged | −.91 [−4.41, 2.59] | 1.55 [−1.75, 4.85] | 1.22 [−1.91, 4.36] | .67 [−2.21, 3.55] | −.98 [−1.19, 3.14] | .51 [−1.11, 2.12] | −1.55 [−4.68, 1.57] | 2.86 [−.76, 6.49] |
| Partner support prov. lagged | .71 [−2.77, 4.19] | −1.78 [−5.14, 1.58] | −.17 [−3.31, 2.97] | −1.14 [−4.13, 1.84] | −.04 [−2.21, 2.13] | 1.19 [−.51, 2.90] | .71 [−2.79, 4.20] | .72 [−2.92, 4.35] |
| Receipt*provision lagged | −4.63 [−10.97, 1.72] | −3.11 [−9.00, 2.79] | −.83 [−6.65, 4.98] | 2.48 [−2.89, 7.85] | .68 [−3.06, 4.43] | −2.68† [−5.62, .26] | −1.82 [−10.19, 6.55] | −10.42* [−20.19, −.65] |
| Variance Components | ||||||||
| Residual var. for men | 174.6** [144, 211] | 179.3** [153, 210] | 148.9** [126, 176] | 139.0** [118, 163] | 83.1** [71.9, 96.1] | 71.0** [61.3, 82.3] | 274.7** [226, 334] | 272.4** [222, 333] |
| Residual var. for women | 139.6** [115, 169] | 144.2** [123, 168] | 106.4** [90.7, 124.9] | 108.5** [93, 126] | 57.8** [60.6, 66.0] | 29.1** [24.3, 34.8] | 271.4** [229, 321] | 337.1** [284, 400] |
| Partial ICC between spouses | .07 [−.09, .22] | −.02 [−.15, .10] | .19**** [.05, .31] | .14* [.01, .27] | .19* [.01, .35] | −.08 [−.29, .14] | 0.91 [−.18, .15] | .01 [−.15, .18] |
| Intercept for men [1,1] | 58.4 [31.2, 109.5] | 47.5** [25.4, 89.1] | 100.7** [69.7, 145.5] | 88.5** [60.8, 128.7] | -- | 4.26 [1.1, 16.3] | 92.5** [52.8, 162.0] | 36.6 [10.5, 127.8] |
| Intercept for women [2,2] | 51.7* [22.5, 118.8] | 21.0† [6.7, 66.3] | 79.2** [50.7, 123.7] | 89.1** [57.7, 137.6] | -- | -- | 26.0 [5.6, 119.8] | 40.5† [13.3, 123.1] |
| Intercept covariance [1,2] | 20.3* [1.2, 39.4] | 8.7 [−6.2, 23.6] | 27.4* [3.1, 51.7] | 35.2* [8.6, 61.8] | -- | -- | 34.3† [−2.7, 71.3] | 35.4* [3.9, 66.9] |
Note: Primary results are bolded. Estimates are pooled across husbands/wives; unstandardized beta weights; person average predictors and daily stressors are grand-mean-centered (GMC); daily predictors are person-centered (PC); 95% confidence intervals for significant effects are shown in brackets. Where only one variance component is included, the model included only an overall random intercept because the results were numerically unstable when separate intercepts were modeled for men and women.
< .10;
p < .05;
p < .01
Is goal-related support visibility related to daily personal and relational outcomes?
We predicted that participants would report greater individual and relational well-being and greater goal progress when they received invisible practical (compared to visible practical) goal-related support. Contrary to this hypothesis, participants reported greater relationship quality when they reported receiving practical support (see Table 8). Partners’ reports of support provision independently predicted increases in relationship quality as well (see Table 8). Consistent with these results, the quasi-signal-detection analysis showed a relative benefit of goal-related visible support days (compared to invisible support days) for relationship quality (see supplementary tables).
Table 8.
Goal-related support receipt and provision results (same-day)
| Self-efficacy | Distress | Relationship quality | Goal progress | |||||
|---|---|---|---|---|---|---|---|---|
| Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | Practical support | Emotional support | |
| Study 1: Young adult sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .15** [.10, .19] | .15** [.10, .19] | .36** [.32, .40] | .36** [.32, .40] | .44** [.40, .48] | .42** [.38, .36] | .26** [.22, .30] | .24** [.21, .29] |
| Person average (GMC) | ||||||||
| Support receipt | 2.26 [−3.07, 7.59] | 2.89 [−1.87, 7.65] | 1.71 [−2.63, 6.05] | −3.18 [−8.12, .76] | 2.42† [−.18, 5.01] | 4.39** [2.00, 6.78] | 14.9** [9.00, 20.9] | 15.4** [10.01, 20.70] |
| Partner support prov. | −4.90† [−10.22, .41] | −1.35 [−5.90, 3.20] | 3.63 [−.74, 8.00] | 1.17 [−2.54, 4.88] | 1.03 [−1.6, 3.63] | .42 [−1.80, 2.63] | −5.48† [−11.36, .39] | −.32 [−5.35, 4.71] |
| Daily (PC) | ||||||||
| Support receipt | 1.07 [−1.02, 3.15] | 2.70** [.91, 4.47] | −1.20 [−3.21, .82] | −2.27* [−3.99, −.55] | 2.18** [.81, 3.56] | 2.18** [−1.80, 2.63] | 2.34 [−.70, 5.39] | 8.43** [5.86, 11.00] |
| Partner support prov. | .04 [−1.96, 2.04] | .25 [−1.57, 2.08] | .56 [−1.38, 2.50] | .95 [−.80, 2.71] | 1.50* [.17, 2.83] | .38 [−.81, 1.58] | 1.81 [−1.11, 4.73] | 3.11* [.37, 5.76] |
| Receipt * provision | −1.13 [−7.29, 5.02] | 4.52† [−.52, 9.56] | 6.10* [.20, 12.00] | −6.78** [−11.6, −1.99] | −1.37 [−5.37, 2.63] | 1.83 [−1.39, 5.06] | −4.05 [−12.90, 4.79] | 5.34 [−1.76, 12.45] |
| Variance Components | ||||||||
| Residual var. for men | 237.0** [215, 261] | 236.1** [214, 260] | 248.7** [224, 276] | 247.5** [223, 275] | 134.7** [119, 152] | 132.7** [118, 149] | 497.5** [450, 550] | 480.0** [434, 530] |
| Residual var. for women | 216.7** [196, 240] | 215.4** [195, 238] | 184.2** [174, 216] | 194.5** [175, 217] | 87.5** [77.8, 98.3] | 88.1** [78.7, 98.7] | 472.5** [427, 522] | 459.0** [415, 507] |
| Partial ICC between spouses | .14** [.07, .20] | .13** [.07, .20] | .22** [.15, .28] | .22** [.16, .29] | .28** [.22, .35] | .29** [.22, .35] | .07† [−.00, .13] | .04 [−.03, .11] |
| Intercept for men [1,1] | 100.0** [71.1, 140.7] | 101.1** [72.0, 142] | 73.0** [46.6, 114.1] | 74.6** [47.0, 115.0] | 27.7** [14.9, 51.4] | 25.5** [15.3, 49.4] | 83.7** [49.1, 142.5] | 93.1** [57.1, 151.9] |
| Intercept for women [2,2] | 122.7** [88.9, 169.3] | 12.4** [89.4, 171] | 62.8** [37.0, 106.7] | 61.8** [36.2, 105.6] | 11.5 [2.9, 46.5] | 11.8 [3.2, 43.3] | 120.1** [76.9, 187.6] | 114.2** [73.4, 177.6] |
| Intercept covariance [1,2] | 9.1 [−13.1, 31.3] | 9.3 [−12.8, 31.5] | 20.6* [3.4, 37.8] | 20.1* [3.0, 37.2] | 7.9* [.4, 15.4] | 7.0† [−.8, 1.6] | 26.9† [−2.2, 56.0] | 24.6† [−3.8, 52.9] |
| Study 2: Aging sample | B | B | B | B | B | B | B | B |
| Outcome lagged | .50** [.45, .55] | .22** [.18, .27] | .22** [.17, .27] | .13** [.09, .18] | .65** [.61, .69] | .56** [−1.52, 2.36] | .16** [.11, .21] | .11** [.07, .15] |
| Person average (GMC) | ||||||||
| Support receipt | −1.42 [−5.17, 2.34] | .46 [−4.23, 5.15] | −.42 [−5.74, 4.89] | −.88 [−6.31, 4.55] | .16 [−1.71, 2.03] | .42 [−1.52, 2.36] | 9.91* [1.79, 17.83] | 21.3** [13.2, 29.3] |
| Partner support prov. | −.54 [−4.42, 3.35] | .30 [−3.94, 4.53] | 3.43 [−2.08, 8.94] | −1.27 [−6.15, 3.62] | .24 [−1.63, 2.12] | .37 [−1.36, 2.11] | 3.23 [−5.15, 11.62] | 0.96 [−8.25, 6.33] |
| Daily (PC) | ||||||||
| Support receipt | 1.01 [−1.37, 3.40] | .35 [−1.52, 2.21] | −3.20** [−5.37, −1.03] | .11 [−1.57, 1.79] | 1.62** [.41, 2.83] | .12 [−.80, 1.05] | 7.13** [3.55, 10.71] | 7.70** [4.58, 10.83] |
| Partner support prov. | 1.41 [−.95, 3.77] | .70 [−1.02, 2.43] | −.85 [−2.96, 1.27] | −1.59 [−3.15, −.03] | .10 [−1.06, 1.26] | −.42 [−1.28 .43] | 1.06 [−2.49, 4.61] | 3.14* [.21, 6.07] |
| Receipt * provision | .44 [−6.09, 6.97] | −.72 [−6.07, 4.63] | −1.03 [−7.20, 5.13] | .98 [−3.90, 5.85] | −.74 [−4.00, 2.52] | −1.07 [−3.60, 1.36] | −5.32 [−15.55, 4.91] | 3.93 [−5.20, 13.06] |
| Variance Components | ||||||||
| Residual var. for men | 221.3** [187, 262] | 181.3** [162, 202] | 179.3** [155, 217] | 156.9** [141, 175] | 77.7** [69.0, 87.6] | 62.8** [50.4, 78.3] | 388.6** [339, 446] | 413.3** [370, 461] |
| Residual var. for women | 178.8** [149, 214] | 136.2** [121, 154] | 132.3** [117, 154] | 109.9** [98.7, 122] | 37.3** [32.1, 43.3] | 28.7** [21.7, 38.0] | 466.3** [408, 533] | 474.6** [426, 529] |
| Partial ICC between spouses | .02 [−.09, .13] | .04 [−.05, .13] | .11† [−.01, .21] | .14** [.05, .22] | .16* [.04, .27] | .06 [−.04, .15] | .13* [.02, .23] | .12** [.03, .21] |
| Intercept for men [1,1] | 16.4 [2.4, 113.0] | 81.6** [56.4, 118.0] | 120.6** [87.0, 167.2] | 149.2** [116.4, 191.1] | 7.8** [3.9, 15.7] | 13.6 [3.62, 51.4] | 211.0** [143.7, 309.7] | 202.5** [145.4, 282.0] |
| Intercept for women [2,2] | 28.2 [7.4, 107.0] | 85.4** [58.4, 124.7] | 149.2** [107.0, 208.0] | 143.4** [109.2, 188.3] | −− | 5.6 [.2, 200.4] | 316.0** [229.1, 435.8] | 296.9** [223.0, 395.4] |
| Intercept covariance [1,2] | 11.7 [−2.6, 26.1] | 22.5* [4.4, 40.7] | 51.3** [19.4, 83.2] | 59.4** [32.4, 86.3] | −− | 8.8† [−1.6, 19.1] | 58.7† [−4.7, 122.0] | 27.3 [−24.1, 78.6] |
Note: Primary results are bolded. Estimates are pooled across husbands/wives; unstandardized beta weights; person average predictors are grand-mean-centered (GMC); daily predictors are person-centered (PC); 95% confidence intervals for significant effects are shown in brackets. Where only one variance component is included, the model included only an overall random intercept because the results were numerically unstable when separate intercepts were modeled for men and women.
< .10;
p < .05;
p < .01
Additionally, although there were no main effects linking goal-related practical support and self-efficacy, distress, or goal progress, there was a significant interaction between reported practical support receipt and partner reported practical support provision for distress (see Table 8). When partner support provision was high (1 SD above the mean), reporting support receipt was unrelated to distress (t(1395.9) = .19, p = .850), but when partner support provision was low (1 SD below the mean), reporting support receipt predicted less distress (t(1351.4) = −2.32, p = .021). This interaction does not indicate a difference between visible and invisible goal-related practical support days for distress but instead suggests a benefit of a false positive compared to no support. The quasi-signal-detection analysis similarly did not indicate a difference in distress between visible and invisible support days.
Regarding goal-related emotional support, young adults reported greater increases in self-efficacy, relationship quality, and goal progress, as well as greater decreases in distress on days when they reported receiving goal-related emotional support (see Table 8). For distress, this main effect was qualified by an interaction, which was consistent with a relative benefit of visible support over invisible support days. When partner support provision was high (1 SD above the mean), reporting support receipt predicted less distress (t(1505.4) = −3.56, p < .001), but when partner support provision was low (1 SD below the mean), reporting support receipt was unrelated to distress (t(1331.8) = −.35, p = .724). The quasi-signal-detection analyses also showed a relative benefit of visible over invisible goal-related emotional support days for all outcomes (see supplementary tables).
In summary, receiving and noticing goal-related practical support was associated with relational well-being benefits in young adulthood. Goal-related emotional support similarly predicted increases in relational well-being as well as individual benefits.
Lagged analyses.
We evaluated links between goal-related support on one day and changes in individual and relational outcomes on the following day. As shown in Table 9, participants reported greater increases in goal progress after days on which they reported receiving goal-related emotional support. No other lagged effects were observed.6
Discussion
Overall, the results of Study 1 revealed an interesting pattern of results linking support visibility and changes in personal and relational outcomes day-to-day. For context-general support, perceiving both practical support (“made it easy to do things”) and emotional support (“expressed confidence in me”) receipt were related to increases in same-day individual and relational well-being outcomes. These context-general results contradict the theorized costs of visible support and instead suggest that young adults may benefit from perceiving spousal support receipt in daily life. Results suggest a relative benefit of visible compared to invisible context-general support for both practical and emotional support. Lagged analyses were consistent with the concurrent analyses, and therefore suggest that support-visibility may promote individual and relational well-being, rather than the reverse.
The results for goal-related support visibility also provided initial support for a benefit of receiving (or even just perceiving) practical support (“gave advice about goals”) and emotional support (“encouraged to work on goals”). Young adult participants reported greater relationship quality on days when they reported receiving goal-related practical or emotional support, and participants reported greater goal progress on days and following days that they reported receiving goal-related emotional support. They also reported greater individual well-being on days when they reported receiving goal-related emotional support. Distress, in particular, was lower on days when participants received visible compared to invisible emotional support for goals. Other well-being outcomes were optimized when participants merely perceived goal-related emotional support receipt (whether or not partners reported providing it). The potential benefit of merely perceiving support also emerged for practical support; participants reported less distress when they perceived practical goal-related support that their partners did not report providing (false positive) than when both partners reported no support. Consistent with the context-general support results, these results suggest a relative benefit of visible support or generally perceiving goal-related support receipt.
Although individuals are motivated to maintain autonomy when they pursue their goals, people may benefit personally and relationally from perceiving spousal support receipt in a goal context because close others provide a “launching function” to facilitate goal pursuit (Feeney & Collins, 2015). These results are also consistent with the attachment theory perspective that individuals seek support from close others in goal contexts and are able to regulate their emotions and explore most effectively when partners demonstrate their responsiveness by providing active support (Bowlby, 1988; Feeney, 2007).
Finally, for support related to stressors, the results followed a somewhat different pattern. For relationship outcomes, we observed a benefit associated with perceiving stress-related practical (“gave advice about a problem”) or emotional (“tried to comfort or reassure”) support receipt. Stress-related practical support receipt was unrelated to individual outcomes, but perceiving stress-related emotional support receipt predicted decreases in self-efficacy. In particular, visible stress-related emotional support predicted greater decreases in self-efficacy than invisible stress-related emotional support.
Overall, the relational benefit and personal costs of stress-related support that we observed are consistent with past research describing visible support as a “mixed blessing” (Gleason et al., 2008). This pattern of results suggests that support for stress best promotes/protects individual well-being when it is not perceived by support-recipients, consistent with past research demonstrating a benefit of invisible over visible support in high-stress samples (e.g., Bolger et al., 2000). Alternatively, participants may be more aware of the support that they receive when they experience extreme stress, and it may be the extreme stress and greater support-seeking in times of stress that drives changes in outcomes day-to-day (i.e., a support mobilization effect). Although we controlled for participants’ reported daily stressors, we may not have been able to account for the full impact of experiencing a stressful day; thus, stress may predict same-day visible support receipt rather than the reverse.
Across all forms of support, there were very few significant interactions between participants’ reported support receipt and their spouses’ reports of support provision. Therefore, the direct comparisons between visible and invisible support days may be driven primarily by the main effects. That is, participants may benefit from perceiving that they received support, whether their partner actually provided it (received support) or did not actually provide it (false positive). This is consistent with the large literature showing that perceived support availability is strongly linked with positive outcomes for perceivers (e.g., Uchino, 2009).
Study 2: Aging Adult Married Sample
In Study 2, we analyzed data from another 7-day daily diary study with a sample of older adult married couples. Because this sample completed the same personal and relational well-being measures and the same measures of daily support receipt and provision, it was an ideal sample for a replication and for testing whether outcomes of support visibility versus invisibility differ in younger and older adult samples.
Method
Participants.
Participants were recruited as part of a larger study of relationships in late adulthood (N = 271 couples; sample size determined a priori based on power analysis for larger study). To be eligible to participate in the larger study, couples needed to be married and able to visit the research laboratory, and target participants needed to be over age 65. The current investigation includes data from all couples who completed the daily diary portion of the larger study (N = 238 couples; diary data collected May 2011 through June 2013).2 These participants were on average 70.0 years old (SD = 6.7) and primarily White (83.5%). Couples had been married for an average of 40.0 years (SD = 14.3), and the majority of couples (69.0%) reported that they had not been previously married. Couples were diverse with regard to household income; the majority of participants reported that their income ranged between US$20,000 and US$40,000 (22.8%), between US$40,000 and US$60,000 (21.6%), between US$60,000 and US$80,000 (19.0%), or between US$80,000 and US$100,000 (14.3%).
Procedure, measures, and analysis strategy.
The procedure and measures were similar to Study 1. After completing laboratory sessions as part of the larger investigation of older adults’ close relationships, participants received instructions for completing a seven-day daily diary. Participants were instructed to complete the daily questionnaires immediately before bed for seven consecutive days, and participants borrowed iPod Touch devices to complete the questionnaires. The full set of daily diary items is available at https://osf.io/68av9/files. One article has previously been published with this daily diary dataset (Jakubiak & Feeney, 2016).
Participants completed items identical to Study 1 to assess their daily support exchanges and visibility, self-efficacy, distress (between-person reliability = .86; reliability of change = .78), relationship quality (between-person reliability = .95; reliability of change = .87), perceived coping, and goal progress (see Tables 2 and 3 for descriptive information). The measure of daily stressors in this study included five additional items for a total of 48 daily stressors. We used identical statistical models to evaluate whether the results replicated in this second sample.
Additionally, to directly compare the results between this aging sample and the young-adult sample (Study 1), we conducted analyses on the full, combined sample. Specifically, we tested for moderation by sample to identify differences in results between the young-adult and aging participants.
Results
Descriptive statistics and zero-order correlations.
Descriptive statistics and zero-order correlations among all continuous study variables are presented in Table 2 and Table 3. Results from the primary analyses are described next and presented in Tables 4–9.
Is context-general support visibility related to daily personal and relational outcomes?
Aging participants experienced consistently better personal and relational well-being on days when they reported receiving high levels of support. Specifically, participants reported greater increases in self-efficacy and relationship quality and greater decreases in distress on days when they reported receiving practical or emotional support (see Table 4). The secondary quasi-signal-detection analysis showed a consistent relative benefit of visible over invisible support (see supplementary tables).
We also observed three interactions between support receipt and partner support provision in this sample. First, we observed an interaction between context-general practical support receipt and partner support provision predicting relationship quality. When partner support provision was high (1 SD above the mean), receiving support was unrelated to relationship quality (t(680.0) = 0.51, p = .610), but when partner support was low (1 SD below the mean), receiving support predicted greater relationship quality (t(636.8) = 3. 71, p < .01). We also observed interactions between context-general emotional support receipt and partner emotional support provision to predict relationship quality and distress. Specifically, when partner support provision was high (1 SD above the mean), receiving support was related to unrelated to relationship quality (t(847.0) = −0.98, p < .328), but when partner support was low (1 SD below the mean), receiving support predicted greater relationship quality (t(781.7) = 3.06, p < .01). Finally, when partner support provision was high (1 SD above the mean), receiving emotional support was unrelated to distress (t(909.7) = −0.69, p < .490), (t(861.9) = −4.47, p < .01). Although these interaction results are not consistent with a benefit of noticing support-receipt (when partner support provision is high), the quasi-signal-detection analysis showed a benefit of visible compared to invisible support (see supplementary tables).
In summary, aging participants benefitted personally and relationally on days when they reported receiving practical or emotional support. Direct comparisons of visible and invisible support days also indicated a personal and relational benefit for visible support days.
Lagged analyses.
We also evaluated links between context-general support on one day and changes in individual and relational outcomes on the following day. Participants reported marginally greater increases in self-efficacy and greater decreases in distress on days after they received greater-than-average context-general practical support (see Table 5). Additionally, participants reported greater self-efficacy and relationship quality and less distress on days after their partner reported providing context-general practical support (see Table 5). Finally, we observed an interaction between context-general practical support receipt and partner support provision predicting the following day’s distress. When partner support provision was high (1 SD above the mean), practical support receipt did not predict the next day’s distress (t(1082.9) = −0.92, p = .355. However, when partner support provision was low (1 SD below the mean), practical support receipt predicted less next day distress (t(1077.3) = −3.15, p < .01. Neither reports of emotional support receipt no provision predicted next-day outcomes (see Table 5).
Sample comparison.
There were three differences in same-day effects between this sample of aging participants and the young-adult sample (Study 1). The first two differences revealed stronger effects of reporting practical and emotional support receipt for relationship quality in the young-adult sample (Study 1) than the aging sample (Study 2). For practical support, sample interacted with the main effect of support receipt (B = −2.83 95% CI [1.48, 3.00], p < .01) such that the main effect of support receipt to predict greater relationship quality was stronger in the young-adult sample (B = 1.95, 95% CI [.99, 2.91], p < .01) than the aging sample (B = 1.20, 95% CI [.01, 2.39], p = .049). Similarly, for emotional support, sample interacted with the main effect of support receipt (B = −2.11, 95% CI [−4.11, −.12], p = .038) such that although the overall effect in the combined sample was significant (B = 3.09, 95% CI [2.09, 4.09], p < .01), the main effect of support receipt was stronger in the young-adult sample (B = 1.63, 95% CI [.54, 2.72],p < .01) than the aging sample (B = .63, 95% CI [−1.04, 2.30], p = .459). Finally, the interaction between emotional support receipt and partner support provision predicting distress was stronger in the aging sample (B = 10.73, 95% CI [3.49, 17.96], p < .01) than the young-adult sample (B = 1.94, 95% CI −2.29, 6.18], p = .369), B = 8.79, 95% CI [.30, 17.17], p = .040. In other words, the relative benefit of false positives compared to no support was evident only in the older adult sample.
In the lagged analyses, we observed a stronger effect of partner practical support provision on self-efficacy in the aging sample (Study 2; B = 5.52, 95% CI [2.21, 8.81], p < .01) than the young-adult sample (Study 1; B = .83, 95% CI [−1.62, 3.28], p = .507), B = 4.68, 95% CI [.57, 8.79], p = .026. We observed no other significant differences between context-general support effects in Study 1 and Study 2.
Is stress-related support visibility related to daily personal and relational outcomes?
On days when participants reported experiencing greater/more severe daily stressors, they reported decreases in self-efficacy, relationship quality, and perceived coping, and they reported increases in distress (see Table 6). Therefore, we controlled for daily stressors, consistent with Study 1.
We observed no main effects of stress-related practical support receipt, but we observed a main effect of partner practical support provision to predict an increase in distress (see Table 6). In the secondary quasi-signal-detection analysis, results revealed that participants reported greater relationship quality on stress-related visible support days compared to invisible support days (see supplementary tables).
For stress-related emotional support, participants reported greater increases in distress, marginally greater decreases in perceived coping, and marginally greater increases in relationship quality on days when they reported receiving greater-than-average stress-related emotional support (see Table 6). Additionally, partner stress-related emotional support provision predicted decreases in self-efficacy. The quasi-signal-detection analysis also showed a relative benefit associated with invisible stress-related support days compared to visible stress-related support days for distress and perceived coping (see supplementary tables).
In summary, the daily links between stress-related support transactions and personal and relational well-being suggest benefits associated with invisible support. For both stress-related practical and emotional support, results indicated a relative benefit associated with not noticing stress-related support, whether or not a partner reported providing it.6
Lagged analyses.
We observed only one lagged effect for stress-related support in this sample. There was an interaction between yesterday’s stress-related emotional support receipt and partner emotional support provision predicting coping (see Table 7). Specifically, when partner support provision is high (1 SD above the mean), perceiving support receipt is unrelated to distress (t(619.3) = −.291, p = .771). When partner support provision is low (1 SD below the mean), perceiving support receipt predicts lower coping (t(1061.4) = −4.44, p < .01). This result is inconsistent with a difference between visible and invisible support days but suggests a cost of false positives compared to no support.
Sample comparison.
The differences between samples again revealed stronger effects of reporting practical support receipt for relationship quality in the young-adult sample (Study 1) than the aging sample (Study 2). Specifically, sample interacted with the main effect of stress-related practical support receipt (B = −2.10, 95% CI [1.11, 3.09], p < .01) such that although the overall effect in the combined sample was significant (B = 2.10, 95% CI [1.11, 3.09], p < .01), the main effect of support receipt on relationship quality was stronger in the young-adult sample (B = 3.15, 95% CI [1.92, 4.37], p < .01) than the aging sample (B = 1.06, 95% CI [−.49, 2.62], p = .181). We observed no other significant same-day or lagged differences between stress-related support effects in Study 1 and Study 2.
Is goal-related support visibility related to daily personal and relational outcomes?
Aging participants reported lower distress, greater relationship quality, and greater goal progress when they reported receiving goal-related practical support (see Table 8). The secondary, quasi-signal-detection, analysis similarly showed a relative benefit of visible support for distress, although the comparison for goal progress did not reach significance (see supplementary tables).
Regarding goal-related emotional support, aging participants reported greater goal progress on days on which they reported receiving emotional support related to their goals and when their partners reported providing goal-related emotional support (see Table 8). The secondary analyses directly comparing visible and invisible support days similarly indicated a relative benefit for visible emotional support days for goal progress (see supplementary tables).
In summary, older adults participants benefitted personally and relationally when they reported receiving goal-related practical support, and they were able to make greater progress on their goals when they received and noticed goal-related practical and/or emotional support.
Lagged analyses.
We also evaluated links between goal-related support on one day and changes in individual and relational outcomes on the following day. Aging participants reported greater goal progress on days after they received goal related practical and goal-related emotional support (see Table 9). No other lagged effects emerged.
Sample comparison.
In the combined sample, we observed that sample interacted with the main effect of goal-related emotional support receipt (B = −1.89, 95% CI [−3.62, −.16], p = .033) to predict relationship quality. Although the overall effect in the combined sample was significant (B = 1.25, 95% CI [.38, 2.12], p < .01), the main effect of support receipt on relationship quality was stronger in the young-adult sample (B = 2.19, 95% CI [1.13, 3.26], p < .01) than the aging sample (B = .31, 95% CI [−1.06, 1.67], p = .662).
In the lagged analyses, there was a difference in the link between practical support and goal progress between the samples. The positive link between practical support receipt and goal progress was stronger in the aging sample (Study 2; B = 14.13, 95% CI [7.37, 20. 89], p < .01) than the young-adult sample (Study 1; B = 2.40, 95% CI [−1.81, 6.61], p = .264), B = 11.73, 95% CI [3.78, 19.70], p = .004. We observed no other significant differences between goal-related support effects in Study 1 and Study 2.7
Discussion
Study 2 provided an opportunity to replicate and assess the generalizability of Study 1’s findings in a second sample of older adults. Generally, the results from this aging sample paralleled the results from the newlywed sample, with a few notable discrepancies. Consistent with Study 1, the results for context-general support revealed a relative benefit associated with perceiving practical and emotional support receipt and suggested a relative benefit of visible over invisible support. That is, older and younger adults alike experienced greater personal and relational well-being on days that they received visible context-general support. The relational benefits of perceiving both practical and emotional support receipt were stronger for the young-adult participants (Study 1) than the aging participants (Study 2). Additionally, older adults showed a relative benefit of false positive practical or emotional support receipt (perceiving practical or emotional support that the partner did not report providing) predicting greater relationship quality, as well as a benefit of false positive emotional support receipt predicting lower distress. These results suggest that merely perceiving support receipt, whether or not the partner reports providing it, may confer individual and relational benefits, at least in late adulthood.
In the stress-context, the results of Study 2 were also largely consistent with Study 1. In both samples, participants reported decreases in personal well-being on days when they perceived stress-related emotional support receipt, revealing a relative benefit of invisible support days. This result may suggest that perceiving emotional support receipt makes stressed individuals feel worse, consistent with theorized costs of enacted support (e.g., Nurullah, 2012; Rafaeli & Gleason, 2009). However, this result could also indicate a support mobilization effect (a reverse effect) whereby participants who are especially distressed subsequently receive stress-related emotional support. Although we controlled for daily stress to reduce this likelihood, it is nonetheless a possible interpretation of these findings. Stress-related results only diverged between samples for stress-related practical support. Newlyweds (Study 1) experienced higher relationship quality on days when they received visible stress-related practical support, whereas aging participants (Study 1) did not experience a relational benefit when they noticed support receipt.
Finally, in the goal-related context, aging participants’ results again largely replicated the pattern of results we observed in Study 1, with some divergences. Younger and older adults both reported greater goal progress on days and following days that they received visible goal-related emotional support. However, goal-related emotional support predicted increases in relationship quality for newlyweds (Study 1) but did not predict relationship quality for older adults (Study 2). Instead, receiving goal-related practical support showed consistent links to increases in relationship quality in both samples. Goal-related practical support receipt also was associated with less distress and greater same-day and next-day goal progress for older adults. The link between goal-related practical support receipt and next-day goal progress was significantly stronger in the aging sample (Study 2) than the newlywed sample (Study 1).
Given these results, receiving assistance (assessed as goal-related advice) may be a more effective form of support in later life. Perhaps the type of goals individuals pursue in early and late adulthood could explain this difference. Older adults tend to pursue more health-related and interpersonal goals for which spouses might offer useful, responsive practical support, whereas younger adults often pursue personal, work-related goals that may be more removed from their partners’ expertise (Berg & Upchurch, 2008; Rauer et al., 2014). Some evidence suggests that older adults receive less frequent unsolicited advice overall, which could make them more receptive to receiving advice (Smith & Goodnow, 1999). Alternatively, young adults may interpret practical support (e.g., advice) as intrusive in a goal context, which could interfere with goal progress (Feeney & Thrush, 2010), whereas older adults may interpret it as helpful if partners have greater goal interdependence later in life.
General Discussion
This investigation aimed to clarify the contested association between support visibility and personal and relational well-being by evaluating links between different forms of social support and daily well-being. Specifically, we assessed daily support behaviors from the perspectives of individuals and their spouses, and we assessed emotional and practical forms of context-general, stress-related, and goal-related support. To expand on the relatively restricted sampling in extant research, we tested our hypotheses in samples of married couples in early adulthood (Study 1) and late adulthood (Study 2), and we sampled broadly rather than restricting to highly-stressed couples.
We pre-registered and tested predictions based on theorized benefits of invisible support to avoid personal well-being costs (e.g., Bolger et al., 2000; Nurullah, 2012; Rafaeli & Gleason, 2009), but we did not form strong hypotheses because actual benefits of invisible support have been observed inconsistently (Belcher et al., 2011; Biehle & Mickelson, 2012; Bolger et al., 2010; Hinnen et al., 2009; Kirsch & Lehman, 2015). We limited our predictions to the particular contexts and type of support that may be most likely to generate personal costs: context-general practical support and goal-related practical support. In stress contexts, we expected that visible practical support might instead be related to personal benefits. Regarding relational outcomes, we expected that receiving emotional support should generally promote relational well-being, but that visible context-general and goal-related practical support might be associated with declines in relational well-being because these forms of support may be less welcomed and more intrusive (e.g., Cutrona, Cohen, & Ingram, 1990).
Table 10 provides an overview of the results regarding the benefits/costs associated with perceiving support receipt as well as the relative benefit associated with visible versus invisible support. In Table 10, we present results orgaanized by support context (context-general, stress-related, goal-related), support type (emotional, practical), and outcome type (personal, relational), and we note the key differences we observed between samples.
Table 10.
Summary of key results organized by support context, support type, and outcome type
| Practical support | Emotional support | |
|---|---|---|
| Context-general support | Individual well-being: | Individual well-being: |
| • Benefit associated with perceiving support receipt (Visible support benefit) | • Benefit associated with perceiving support receipt (Visible support benefit) | |
| Relational well-being: | Relational well-being: | |
| • Benefit associated with perceiving support receipt (Visible support benefit) | • Benefit associated with perceiving support receipt (Visible support benefit) | |
| • Stronger same-day benefit associated with perceiving support for young-adult sample | • Stronger same-day benefit associated with perceiving support for young-adult sample | |
| Stress-related support | Individual well-being: | Individual well-being: |
| • Not associated with perceiving support receipt | • Cost associated with perceiving support receipt (Invisible support benefit) | |
| Relational well-being: | Relational well-being: | |
| • Young adults only: Benefit associated with perceiving support receipt (Visible support benefit) | • Benefit associated with perceiving support receipt (Visible support benefit) | |
| Goal-related support | Individual well-being: | Individual well-being: |
| • Young adults: False positive benefit compared to no support | • Benefit associated with perceiving support receipt (Visible support benefit) | |
| • Older adults: Benefit associated with perceiving support receipt (Visible support benefit) | ||
| Relational well-being: | Relational well-being: | |
| • Benefit associated with perceiving support receipt (Visible support benefit) | • Young adults: Benefit associated with perceiving support receipt (Visible support benefit) |
Context-General Support Visibility
Overall, the results of both studies were inconsistent with the often-theorized broad benefits of invisible support to protect personal well-being. Both younger and older adults reported greater improvements in personal well-being when they reported receiving context-general support (emotional and practical) compared to when they did not report support receipt. Lagged analyses indicated that perceiving context-general support receipt predicted increases in individual well-being prospectively, which suggests that perceiving support receipt may enhance well-being rather than only the reverse.
A direct comparison of visible and invisible support days revealed that participants reported greater self-efficacy and less distress on days that they received visible support compared to invisible support, which provides further evidence for the benefit associated with perceiving context-general spousal support across the adult lifespan. Moreover, older adults reported greater decreases in distress on days when they perceived false positive emotional support compared to no support days, and they experienced greater decreases in distress on days after they perceived false positive practical support compared to no support days (Study 2). These findings, though unexpected, highlight the importance of merely perceiving support receipt, both when partners report providing support and when they do not.
Our context-general results are inconsistent with past research that showed relative personal benefits associated with invisible support receipt in samples experiencing a major stressor (e.g., Bolger et al., 2000; Gleason et al., 2008) and in a sample of young-adult cohabiting couples (Maisel & Gable, 2009). However, our results are consistent with one study showing that context-general emotional support receipt was associated with better same-day mood in a sample of typical married couples (not sampled for high stress) in the early years of marriage (Biehle & Mickelson, 2012). Given this pattern of findings across studies, it may be the case that support receipt (emotional and practical) is costly in contexts characterized by high stress but welcomed when couple-members are not experiencing a major stressor. In the current work, couples were not recruited with the criteria of experiencing a major stressor, so though they likely had variable levels of typical stress, they were not a highly or uniformly stressed sample overall. Received support in a general (not specifically stressed) context may provide assistance and communicate concern rather than heighten the salience of an already salient stressor. The pattern of findings across studies may also indicate that visible support receipt may be more beneficial in marital relationships (e.g., Biehle & Mickelson, 2012) than in relationships characterized by less commitment (e.g., Maisel & Gable, 2009). In marital relationships, which entail greater interdependence, support receipt may be expected and therefore not communicate inefficacy or threaten personal well-being (e.g., Rusbult & Van Lange, 2003).
Regarding relationship quality, our results are consistent with past research showing benefits of visible support for relationship outcomes. In both samples, we observed positive associations between participants’ reports of context-general practical and emotional support receipt and improvements in relationship quality on the same day. For the newlywed sample (Study 1), we also observed next-day relational benefits associated with perceiving support receipt. A direct comparison between visible and invisible support days revealed greater increases in relationship quality on visible support days and provided converging evidence for a relational benefit of perceiving support receipt. These results are consistent with Gleason and colleagues’ (2008) study in which visible support enhanced daily closeness in a stressed sample. Therefore, our results suggest that the relational benefits of visible support may be more consistent and generalizable across samples than the personal costs of visible support.
Although we observed relational benefits associated with perceiving practical and emotional support receipt in both samples, these associations were stronger for young adults (Study 1) than older adults (Study 2). Young adults’ relational perceptions may be more malleable day-to-day due to the relative newness of their relationships. In general (and in the current research), older adults typically have more established relationships in which perceptions may be more stable overall.8 Finally, older adults (Study 2) reported relational benefits associated with false positive emotional and practical support days (compared to no support days) to improve their relationship quality. These findings are consistent with the personal benefits (i.e., lower distress) associated with false positive support that we observed in our aging sample and reveal the potential importance of merely perceiving that support was received. A great deal of past research has compared visible and invisible support; the current findings suggest that researchers also should consider the consequences of false positives in dyadic support exchanges. False positives may be a form of positive illusion in which people perceive their partner’s support or support intentions idealistically. Like positive illusions, false positives may prove beneficial to improve and protect relationship quality (e.g., Miller, Niehuis, & Huston, 2006; Murray, Holmes, & Griffin, 1996).
Stress-Related Support Visibility
In an effort to clarify inconsistencies in the support visibility literature, we tested associations between specific instances of stress-related practical and emotional support and daily well-being. Consistent with past research showing benefits of invisible support to protect mood and self-efficacy for individuals under high stress (e.g., Bolger et al., 2000; Bolger & Amarel, 2007), we observed decreases in self-efficacy, increases in distressed mood, and marginal decreases in coping when participants reported receiving stress-related emotional support. Described differently, individuals reported greater psychological distress (both samples), lower self-efficacy (Study 1), and lower perceived coping (Study 2) on visible compared to invisible support days. In the aging sample (Study 2), perceiving stress-related emotional support when the partner did not report providing it (false positive) also predicted poorer perceived coping on the following day.
Together, these results may indicate that receiving (and noticing) support tied to specific problems enhances distress and undermines self-efficacy and coping, perhaps because it communicates that one is inefficacious or highlights the stressor’s salience. An alternative explanation is that of support mobilization (a reverse effect). That is, on days on which people are highly distressed about specific stressors, they may be more likely to receive (and notice) stress-related support. Seidman, Shrout, and Bolger (2006) investigated the reason for the association between support receipt and next day distress using simulation, and they concluded that the costs of support receipt cannot be explained by support mobilization or experiencing adverse events in those lagged analyses. Because many of our analyses assess the relationship between support and outcomes on the same day, it is possible that our results could in some cases indicate support mobilization.
Unlike emotional support, we found that stress-related practical support was consistently unrelated to individual well-being outcomes. These null findings could be due to our measurement of stress-related practical support (i.e., “gave me advice about a problem”). Advice is only one of many ways that someone could provide practical stress-related support, and other forms of stress-related practical support like tangible task assistance may be more impactful, for better or worse. Alternatively, these null findings could indicate variability in how people respond to stress-related practical support that we cannot differentiate with our current approach. For instance, the benefits and/or costs associated with stress-related practical support may be restricted to some people or to people coping with a specific type of stressor.
Regarding relational outcomes, we observed that perceiving support receipt (practical and emotional forms) was associated with same-day increases in daily relationship quality in the young adult sample (Study 1) but not in the aging sample (Study 2). Perceiving emotional support receipt also predicted next-day increases in relationship quality in the young adult sample only. However, our direct sample comparison indicated that only the same-day link between perceiving practical support receipt and relationship quality differed significantly between samples. Therefore, it seems that younger and older adults may both perceive their relationships more favorably when they perceive emotional support tied to stressors whereas only young adults are sensitive to stress-related practical support receipt. The relatively stronger links between stress-related support visibility and changes in relational perceptions in the young adult sample compared to the aging sample is consistent with context-general support findings.
Overall, these findings suggest that receiving support during stress may indeed be, as Gleason and colleagues’ (2008) described, “a mixed blessing.” Under stress, individuals who receive visible support may experience relational benefits as well as personal well-being costs. Attachment theory posits that stressors activate the attachment system and that people therefore benefit from receiving support during stress (Bowlby, 1969/1982, 1988; Mikulincer & Shaver, 2009). However, support must be provided responsively for benefits to accrue (e.g., Maisel & Gable, 2009). In a high stress context, responsive support provision may indeed entail indirect, “under the radar,” provision of support, as proponents of invisible support’s benefits have argued (e.g., Bolger et al., 2000; Bolger & Amarel, 2007; Girme et al., 2013; Howland & Simpson, 2010).
Goal-Related Support Visibility
For goal-related support—enacted support specifically related to goal pursuit—we observed a relative benefit associated with receiving and noticing support, inconsistent with the theorized benefits of invisible support. In the young adult sample (Study 1), the positive links between perceiving goal-related support receipt and individual well-being were restricted to emotional support (Study 1). However, aging participants (Study 2) primarily reported personal well-being benefits when they reported receiving goal-related practical support. Specifically, young adults (Study 1) reported greater increases in self-efficacy and goal progress, as well as greater decreases in distress on days when they received visible (compared to invisible) goal-related emotional support. Greater goal-related emotional support also predicted next-day increases in goal progress for young adults. Aging adults (Study 2) similarly experienced same-day and next-day increases in goal-progress when they reported receiving goal-related emotional support, but they also experienced increases in goal progress (same-day and next-day) and decreases in distress (same-day) when they reported receiving goal-related practical support.
The overall consistency of these results underscores the importance of goal-related support across the adult lifespan. These results also suggest that goal-related practical support may be more effective to promote individual well-being for older compared to younger adults. One explanation for this divergence is related to how individual goals change over time (Berg & Upchurch, 2008; Rauer et al., 2014). Older adults have more health and relational goals, for which practical support may be welcomed, whereas young adults may have more individual goals related to their careers for which partner advice may be unhelpful or perhaps even intrusive (e.g., Carstensen, Isaacowitz, & Charles, 1999).
Regarding relational outcomes, results largely paralleled findings in other support contexts. Both younger and older adults reported increases in relationship quality when they received visible support for goals, and these apparent benefits were stronger for young adults (Study 1). Specifically, young adults reported increases in relationship quality on days when they reported receiving practical and emotional support for goals, whereas older adults reported relational benefits associated with goal-related practical support only. These results again suggest that young adults shift their relational perceptions in relation to support interactions to a greater extent than do older adults.
Overall, these goal-related support findings support the theorized individual and relational benefits of receiving (and noticing) relational catalyst support to promote engagement with goals and life opportunities outside of stress contexts (Feeney & Collins, 2015). These results also support a recent theory suggesting that having a partner that is instrumental to one’s goals promotes goal attainment and improves relationship quality (Orehek & Forest, 2016). Inconsistent with theorized benefits of invisible support, younger and older adults alike reported the greatest improvements in personal and relational well-being when they received and noticed goal-related support.
Strengths, Limitations, and Future Directions
A key strength of this work is the focus on practical and emotional forms of social support in three distinct contexts (context-general, stress-related, and goal-related). This multi-dimensional investigation of support is consistent with the multi-dimensional nature of social support (e.g., Feeney & Collins, 2015). The results of this study do help clarify inconsistencies in past work by showing that support visibility is associated with individual well-being differently in stress and non-stress contexts.
A second theoretical strength is that we sampled participants from across the adult lifespan to test the attachment theoretical perspective that enacted support should be associated with individual and relational well-being across the lifespan (Bowlby, 1969/1982, 1988; Mikulincer & Shaver, 2009). In general, we observed similar associations between support receipt and well-being in newlyweds and aging adult couples. We also observed consistently stronger links between perceiving support receipt and relational well-being for young adults. This latter finding suggests the need for future research to directly compare the malleability of relationship quality and the unique predictors of changes in relationship quality across the adult lifespan. In the current work, we cannot differentiate whether participants’ age or relationship length explains sample differences because the samples differ on both factors and age and relationship length are highly correlated (r = .91). However, the current research helps to address the oversampling of young adult couples and adds to recent research showing benefits of support receipt in late adulthood (e.g., Jakubiak & Feeney, 2016).
The current research also has several methodological strengths. First, we included controls in our analyses to enhance the interpretability of our results. Specifically, we controlled for the previous day’s outcome so that we can assess predictors of day-to-day change, and we controlled for average support receipt and provision to isolate within person effects. Researchers have rarely corrected for participants’ typical support receipt to isolate within-person effects of daily support receipt (c.f. Belcher et al., 2011; Lüscher et al., 2015; Shrout et al., 2010). Second, we evaluated associations between support transactions and well-being using the two primary analysis strategies used in past research and report the results of the secondary (quasi-signal-detection) analysis in supplementary materials. Thus, we provide results that can be more directly compared to a broad set of past studies of invisible support and support receipt.
Although the current research has several notable strengths, it also has limitations that should be addressed in future work. A primarily limitation is our inability to determine causality in the analyses that link support transactions and outcomes on the same day. In these concurrent analyses, support transactions may influence daily well-being, daily well-being may influence support transactions, or some third variable may influence both support transactions and well-being to create the observed associations. Despite this potential limitation of same-day analyses, we expect that support exchanges impact outcomes immediately, on the same day, and thus, concurrent models are most informative of support processes. Future research should assess support transactions and participants’ outcomes several times in the same day to include the methodological strengths of prospective analyses without sacrificing the theoretical sensibility of the same-day approach. Additionally, experimental studies in which context-general, stress-related, and goal-related support are individually manipulated would provide a stronger test of the directionality underlying the current results.
Several additional limitations stem from the items we used to measure support. First, the support items may not perfectly differentiate context-general, stress, and goal contexts, despite our efforts to select items that would best capture these contexts. For example, people may experience goal-pursuit as stressful or vice versa, or participants may focus on a stress or goal context when responding to the items intended to be context-general. Second, we measured support exchanges using daily self-report measures that do not provide any information about support sensitivity or responsiveness to needs. From past diary work and observational research (e.g., Collins & Feeney, 2000; Feeney, 2004; Maisel & Gable, 2009), we know that the outcomes of support receipt depend on whether the support is provided responsively and sensitively. Therefore, it is possible that context-general support may have been associated with greater benefits in the current work than in previous research because it was provided more responsively in these samples. Differences in responsiveness may also explain the disparate findings for context-general support, stress-related support, and goal-related support in the current research. It may be more difficult to provide responsive support in stress contexts than other contexts. Future research should explore this possibility. Third, the stress-related support items we used may have captured minor stressors rather than major stressors, which limits our ability to directly compare our stress-related results to past work that samples participants experiencing chronic stressors or a major stressful life event.
Finally, our measurement of support interactions spanned only seven days. Recent methodological recommendations suggest sampling over a longer time so that the first few days—on which participants may react differently due to the new experience of reflecting on their relationships daily—can be excluded (Shrout et al., 2018). Our pattern of results is unchanged by excluding the first two days of daily reports; however, future researchers might assess support visibility over longer periods of time. Relatedly, a small number of dairy days may have prevented estimation of random effects for the support effects. Future studies with a larger number of diary days could investigate heterogeneity in these support effects.
There are many exciting avenues for future research, as the current work represents only an initial attempt to clarify the consequences associated with support visibility. Other researchers have recently attempted to refine our understanding of when and for whom visible and invisible support are preferable by assessing the roles of support-recipients’ attachment orientation (Girme, Overall, & Hammond, 2019), predominant motivation (locomotion versus assessment; Zee, Cavallo, Flores, Bolger, & Higgins, 2018), and perceived control (Ryon & Gleason, 2018). Past research has also consistently underscored the importance of support reciprocity or equity in support receipt. Future researchers and theorists should consider how context and these others factors jointly influence the consequences of support visibility.
Conclusion
In conclusion, across two studies with couples in different life and relationship stages, we demonstrated that the association between support visibility and individual and relational well-being outcomes is complex and nuanced. Invisible support was associated with individual well-being benefits in a stress-context (support for specific problems), but visible support was generally related to greater benefits in goal contexts and in daily life (context-general). We sampled adult couples in earlier and later life and found overall consistency in the well-being outcomes associated with perceiving support receipt. Most importantly, this research suggests that awareness of support receipt may enhance, rather than detract from, well-being, and it motivates additional research to identify the conditions in which invisible versus visible support are most advantageous.
Supplementary Material
Acknowledgments
This work was supported by grants from the National Science Foundation (BCS0424579, Study 1) and the National Institutes of Health (1R01AG032370 – 01A2, Study 2) to the second author. Special thanks to Nataliya Rozinskiy (Study 1) and Bria Toneff (Study 2) to for programming the diaries and managing data collection.
Footnotes
The pre-registered hypotheses pertained to older adults (Study 2).
A sensitivity power analysis indicated that given this sample size, we were approximately 90% powered to detect a small effect (i.e., a 2-point change in our outcome variables, on a 100-point scale).
These items were selected from a larger set of daily items that assessed many aspects of support that are expected to occur in daily life because they best exemplified the constructs of interest.
This “second analysis” is the analysis that we pre-registered. However, after reviewing the results, there were consistent main effects of support receipt; outcomes were consistently similar on visible support and false positive days, and outcomes were consistently similar on invisible support and support absent days. Therefore, we present the main effect results in addition to comparing visible and invisible support days.
Daily stress did not moderate links between daily stress-related support and outcomes.
In all but one case, effects were equal across gender. Perceiving goal-related emotional support predicted greater increases in daily goal progress for women (B = 11.52, 95% CI [7.77, 15.26], p < .01) than men (B = 6.00, 95% CI [2.47, 9.53], p < .01). The pattern of results also held after removing the first two daily diary days, though the significance level of some results changed.
We observed a few gender differences in Study 2. Context-general practical support receipt more strongly reduced same-day distress for wives (B = −6.66, 95% CI [−8.91, −4.41], p < .01) than husbands (B = −3.09, 95% CI [−5.00, −1.18], p < .01) and enhanced same-day relationship quality for wives (B = 2.76, 95 % CI [1.38, 4.14], p < .01) more than husbands (B = .47, 95% CI [−.47, 1.41], p = .325). Stress-related emotional support was more strongly linked to lower coping for wives (B = −4.75, 95 % CI [−7.92, −1.56], p < .01) than husbands (B = .58, 95% CI [−2.86, 3.99], p = .739). Partner goal-related emotional support provision predicted greater decreases in same-day distress for husbands (B = −3.03, 95 % CI [−4.97, −1.09], p < .01) than wives (B = −.25, 95% CI [−2.90, 2.39], p = .693). Additionally, partner goal-related emotional support provision predicted decreases in same-day relationship quality for wives (B = −2.02, 95% CI [−3.65, −.38], p = .016) but not for husbands (B = .14, 95% CI [−.86, 1.14], p = .782). The pattern of results also held after removing the first two daily diary days, though the significance level of some results changed.
In an empty model (no predictors), there is more residual variance in daily relationship quality in the young-adult sample (Study 1, σmen = 107.28, SE = 4.48; σwomen = 78.17, SE = 3.23) than the aging sample (Study 2; σmen = 65.84, SE = 2.48; σwomen = 38.53, SE = 1.50).
Contributor Information
Brittany K. Jakubiak, Syracuse University
Brooke C. Feeney, Carnegie Mellon University
Rebecca A. Ferrer, Basic Biobehavioral and Psychological Sciences Branch, Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, National Institutes of Health
References
- Bar-Kalifa E, Pshedetzky-Shochat R, Rafaeli E & Gleason MEJ (2018). Daily support equity in romantic couples: Response surface analysis of monadic and dyadic data. Social Psychological and Personality Science, 9, 790–801. 10.1177/1948550617725150 [DOI] [Google Scholar]
- Belcher AJ, Laurenceau J-P, Graber EC, Cohen LH, Dasch KB, & Siegel SD (2011). Daily support in couples coping with early stage breast cancer: maintaining intimacy during adversity. Health Psychology, 30(6), 665–73. 10.1037/a0024705 [DOI] [PubMed] [Google Scholar]
- Berg CA, Meegan SP, & Deviney FP (1998). A social-contextual model of coding with everyday problems across the lifespan. International Journal of Behavioral Development, 2, 239–261. [Google Scholar]
- Berg CA, & Upchurch R (2007). A developmental-contextual model of couples coping with chronic illness across the adult life span. Psychological Bulletin, 133, 920–954. 10.1037/0033-2909.133.6.920 [DOI] [PubMed] [Google Scholar]
- Biehle SN, & Mickelson KD (2012). Provision and receipt of emotional spousal support: The impact of visibility on well-being. Couple and Family Psychology: Research and Practice, 1(3), 244–251. 10.1037/a0028480 [DOI] [Google Scholar]
- Birditt KS, Antonucci TC, & Tighe L (2013). Enacted support during stressful life events in middle and older adulthood: An examination of the interpersonal context. Psychology of Aging, 27(3), 728–741. 10.1037/a0026967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bolger N, & Amarel D (2007). Effects of social support visibility on adjustment to stress: experimental evidence. Journal of Personality and Social Psychology, 92(3), 458–475. 10.1037/0022-3514.92.3.458 [DOI] [PubMed] [Google Scholar]
- Bolger N, Zuckerman A, & Kessler RC (2000). Invisible support and adjustment to stress. Journal of Personality and Social Psychology, 79(6), 953–961. 10.1037//0022-3514.79.6.953 [DOI] [PubMed] [Google Scholar]
- Bowlby J (1969/1982). Attachment and loss: Vol. 1 Attachment (2nd ed.) New York, NY: Basic Books. [Google Scholar]
- Bowlby J (1988). A secure base: Parent-child attachment and healthy human development. New York: Basic Books. [Google Scholar]
- Braithwaite S, & Holt-Lunstad J (2017). Romantic relationships and mental health. Current Opinion in Psychology, 13, 120–125. 10.1016/j.copsyc.2016.04.001 [DOI] [PubMed] [Google Scholar]
- Brown SL, Nesse RM, Vinokur AD, & Smith DM (2003). Providing social support may be more beneficial than receiving it: Results from a prospective study of mortality. Psychological Science, 14, 320–327. [DOI] [PubMed] [Google Scholar]
- Carstensen LL (1992). Social and emotional patterns in adulthood. Psychology and Aging, 7(3), 331–338. [DOI] [PubMed] [Google Scholar]
- Carstensen LL, Isaacowitz DM, & Charles ST (1999). Taking time seriously: A theory of socioemotional selectivity. American Psychologist, 54, 165–181. 10.1037/0003-066X.54.3.165 [DOI] [PubMed] [Google Scholar]
- Cohen S (2004). Social relationships and health. American Psychologist, 59, 676–684. [DOI] [PubMed] [Google Scholar]
- Cohen S, & Wills TA (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310–57. [PubMed] [Google Scholar]
- Collins NL, & Feeney BC (2000). A safe haven: An attachment theory perspective on support seeking and caregiving in intimate relationships. Journal of Personality and Social Psychology, 78(6), 1053–1073. 10.1037/0022-3514.78.6.1053 [DOI] [PubMed] [Google Scholar]
- Collins NL, & Feeney BC (2010). An attachment theoretical perspective on social support dynamics in couples: Normative processes and individual differences In Sullivan K & Davila J (Eds.), Support Processes in Intimate Relationships (pp. 89–120). New York, NY: Oxford University Press. [Google Scholar]
- Cranford JA, Shrout PE, Iida M, Rafaeli E, Yip T, & Bolger N (2006). A procedure for evaluating sensitivity to within-person change: Can mood measures in diary studies detect change reliably? Personality and Social Psychology Bulletin, 32, 917–929. 10.1177/0146167206287721 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cutrona C, Cohen S, & Ingram S (1990). Contextual determinants of the perceived supportiveness of helping behaviors. Journal of Social and Personal Relationships, 7, 553–562. 10.1177/0192513X12437708 [DOI] [Google Scholar]
- Cutrona CE, Russell DW, & Gardner KA (2005). The relationship enhancement model of social support In Revenson TA, Kayser K, & Bodenmann G (Eds.), Couples Coping With Stress: Emerging Perspectives on Dyadic Coping (pp. 73–95). Washington, DC: American Psychological Association. [Google Scholar]
- Feeney BC (2004). A secure base: responsive support of goal strivings and exploration in adult intimate relationships. Journal of Personality and Social Psychology, 87(5), 631–48. 10.1037/0022-3514.87.5.631 [DOI] [PubMed] [Google Scholar]
- Feeney BC (2007). The dependency paradox in close relationships: Accepting dependence promotes independence. Journal of Personality and Social Psychology, 92(2), 268–285. 10.1037/0022-3514.92.2.268 [DOI] [PubMed] [Google Scholar]
- Feeney BC, & Collins NL (2003). Motivations for Caregiving in Adult Intimate Relationships: Influences on Caregiving Behavior and Relationship Functioning. Personality and Social Psychology Bulletin, 29(8), 950–968. 10.1177/0146167203252807 [DOI] [PubMed] [Google Scholar]
- Feeney BC, & Collins NL (2015). A new look at social support: A theoretical perspective on thriving through relationships. Personality and Social Psychology Review, 19(2), 113–147. 10.1177/1088868314544222 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Feeney BC, & Thrush RL (2010). Relationship influences on exploration in adulthood: the characteristics and function of a secure base. Journal of Personality and Social Psychology, 98(1), 57–76. 10.1037/a0016961 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Feeney BC, Van Vleet M, Jakubiak BK, & Tomlinson JM (2017). Predicting the pursuit and support of challenging life opportunities. Personality and Social Psychology Bulletin, 43(8), 1171–1187. 10.1177/0146167217708575 [DOI] [PubMed] [Google Scholar]
- Finkel EJ, & Rusbult CE (2008). Prorelationship motivation: An interdependence theory analysis of situations with conflicting interests In Shah JY & Gardner WL (Eds.), Handbook of motivation science. New York: NY: Guilford Press. [Google Scholar]
- Fisher JD, Nadler A, & Whitcher-Alagna S (1982). Recipient reactions to aid. Psychological Bulletin, 91(1), 27–54. 10.1037/0033-2909.91.1.27 [DOI] [Google Scholar]
- Gable SL, Reis HT, & Downey G (2003). He said, she said: A quasi-signal detection analysis of daily interactions between close relationship partners. Psychological Science, 14(2), 100–105. 10.1111/1467-9280.t01-1-01426 [DOI] [PubMed] [Google Scholar]
- Girme YU, Maniaci MR, Reis HT, McNulty JK,. Carmichael CL, Gable SL, Baker LR, & Overall NC (2018). Does support need to be seen? Daily invisible support promotes next day relationship well-being. Journal of Family Psychology, 32, 882–893. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Girme YU, Overall NC, & Hammond MD (2019). Facilitating autonomy in interdependent relationships: Invisible support facilitates highly avoidant individuals’ autonomy. Journal of Family Psychology, 33, 154–165. 10.1037/fam0000488 [DOI] [PubMed] [Google Scholar]
- Girme YU, Overall NC, & Simpson J. a. (2013). When visibility matters: Short-term versus long-term costs and benefits of visible and invisible support. Personality & Social Psychology Bulletin, 39(11), 1441–54. 10.1177/0146167213497802 [DOI] [PubMed] [Google Scholar]
- Gleason MEJ, Iida M, Bolger N, & Shrout PE (2003). Daily supportive equity in close relationships. Personality and Social Psychology Bulletin, 29(8), 1036–1045. 10.1177/0146167203253473 [DOI] [PubMed] [Google Scholar]
- Gleason MEJ, Iida M, Shrout PE, & Bolger N (2008). Receiving support as a mixed blessing: Evidence for dual effects of support on psychological outcomes. Journal of Personality and Social Psychology, 94(5), 824–838. 10.1037/0022-3514.94.5.824. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hinnen C, Ranchor AV, Baas PC, Sanderman R, & Hagedoorn M (2009). Partner support and distress in women with breast cancer: The role of patients’ awareness of support and level of mastery. Psychology and Health, 24(4), 439–455. 10.1080/08870440801919513 [DOI] [PubMed] [Google Scholar]
- Holmes JG, & Rempel JK (1989). Trust in close relationships In Hendrick C (Ed.), Review of Personality and Social Psychology, Vol. 10. Close Relationships (pp. 187–220). Thousand Oaks, CA: Sage Publications, Inc. [Google Scholar]
- Holt-Lunstad J, Smith TB, & Layton JB (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), e1000316 10.1371/journal.pmed.1000316 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Howland M (2015). Reading minds and being invisible: The role of empathic accuracy in invisible support provision. Social Psychological and Personality Science, 7(2), 149–156. 10.1177/1948550615606195 [DOI] [Google Scholar]
- Howland M, & Simpson JA (2010). Getting in under the radar. A dyadic view of invisible support. Psychological Science, 21(12), 1878–85. 10.1177/0956797610388817 [DOI] [PubMed] [Google Scholar]
- Jakubiak BK, & Feeney BC (2016). Daily goal progress is facilitated by spousal support and promotes psychological, physical, and relational well-being throughout adulthood. Journal of Personality and Social Psychology, 111(3), 317–340. 10.1037/pspi0000062.supp [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kenny DA, Kashy DA, & Cook WL (2006). Dyadic data analysis. New York, NY: Guilford Press. [Google Scholar]
- Kirsch JA, & Lehman BJ (2015). Comparing visible and invisible social support: Non-evaluative support buffers cardiovascular responses to stress. Stress and Health, 31(5), 351–364. 10.1002/smi.2558 [DOI] [PubMed] [Google Scholar]
- Lakey B (2013). Social support processes in relationships In Simpson JA & Campbell L (Eds.), The Oxford Handbook of Close Relationships (pp. 711–728). New York, NY: Oxford University Press. [Google Scholar]
- Lang FR, & Carstensen LL (1994). Close emotional relationship in late life: Further support for proactive aging in the social domain. Psychology and Aging, 9(2), 315–324. 10.1037//0882-7974.9.2.315 [DOI] [PubMed] [Google Scholar]
- Laurenceau J-P, & Bolger N (2005). Using diary methods to study marital and family processes. Journal of Family Psychology, 19(1), 86–97. [DOI] [PubMed] [Google Scholar]
- Lemay EP, Clark MS Jr., & Feeney BC (2007). Projection of responsiveness to needs and the construction of satisfying communal relationships. Journal of Personality and Social Psychology, 92, 834–853. 10.1037/0022-3514.92.5.834 [DOI] [PubMed] [Google Scholar]
- Lemay EP Jr., & Neal AM (2013). The wishful memory of interpersonal responsiveness. Journal of Personality and Social Psychology, 104, 653–672. http://dx.doi/10.1037/a0030422 [DOI] [PubMed] [Google Scholar]
- Liang J, Krause NM, & Bennett JM (2001). Social exchange and well-being: Is giving better than receiving? Psychology and Aging, 16, 511–523. 10.1037//0882-7974.16.3.511. [DOI] [PubMed] [Google Scholar]
- Lüscher J, Stadler G, Ochsner S, Rackow P, Knoll N, Hornung R, & Scholz U (2015). Daily negative affect and smoking after a self-set quit attempt: The role of dyadic invisible social support in a daily diary study. British Journal of Health Psychology, 20(4), 708–723. 10.1111/bjhp.12135 [DOI] [PubMed] [Google Scholar]
- Maisel NC, & Gable SL (2009). The paradox of received social support: The importance of responsiveness. Psychological Science, 20(8), 928–932. 10.1111/j.1467-9280.2009.02388.x [DOI] [PubMed] [Google Scholar]
- McClure MJ, Xu JH, Craw JP, Lane SP, Bolger N, & Shrout PE (2014). Understanding the costs of support transactions in daily life. Journal of Personality, 82(6), 563–574. 10.1111/jopy.12061 [DOI] [PubMed] [Google Scholar]
- Mikulincer M, Gillath O, & Shaver PR (2002). Activation of the attachment system in adulthood: Threat-related primes increase the accessibility of mental representations of attachment figures. Journal of Personality and Social Psychology, 83(4), 881–895. 10.1037//0022-3514.83.4.881 [DOI] [PubMed] [Google Scholar]
- Mikulincer M, & Shaver PR (2007). Reflections on security dynamics: Core constructs, psychological mechanisms, relational contexts, and the need for an integrative theory, Psychological Inquiry, 18(3), 197–209. [Google Scholar]
- Mikulincer M, & Shaver PR (2009). An attachment and behavioral systems perspective on social support. Journal of Social and Personal Relationships, 26(1), 7–19. 10.1177/0265407509105518 [DOI] [Google Scholar]
- Miller PJE, Neihuis S, & Huston TL (2006). Positive illusions in marital relationships: A 13-year longitudinal study. Personality and Social Psychology Bulletin, 32, 1579–1594. 10.1177/0146167206292691 [DOI] [PubMed] [Google Scholar]
- Murray SL, Holmes JG, & Collins NL (2006). Optimizing assurance: The risk regulation system in relationships. Psychological Bulletin, 132, 641–666. [DOI] [PubMed] [Google Scholar]
- Murray SL, Holmes JG, & Griffin DW (1996). The benefits of positive illusions: Idealization and the construction of satisfaction in close relationships. Journal of Personality and Social Psychology, 70, 79–98. [Google Scholar]
- Nadler A, Fisher JD, & Streufert S (1976). When helping hurts: Effects of donor-recipient similarity and recipient self-esteem on reactions to aid. Journal of Personality, 392–402. [DOI] [PubMed] [Google Scholar]
- Nurullah A (2012). Received and provided social support: A review of current evidence and future directions. American Journal of Health Studies, 27, 173–188. [Google Scholar]
- Orehek E, & Forest A (2016). When people serve as means to goals: Implications of a motivational account of close relationships. Current Directions in Psychological Science, 25, 79–84. [Google Scholar]
- Overall NC, Fletcher GJO, & Simpson J. a. (2010). Helping each other grow: romantic partner support, self-improvement, and relationship quality. Personality & Social Psychology Bulletin, 36(11), 1496–513. 10.1177/0146167210383045 [DOI] [PubMed] [Google Scholar]
- Rafaeli E, & Gleason MEJ (2009). Skilled support within intimate relationships. Journal of Family Theory & Review, 1, 20–37. [Google Scholar]
- Rauer AJ, Sabey A, & Jensen JF (2014). Growing old together: Compassionalte love and health in older adulthood. Journal of Social and Personal Relationships, 31, 677–696. 10.1177/0265407513503596 [DOI] [Google Scholar]
- Reinhardt JP, Boerner K, & Horowitz A (2006). Good to have but not to use: Differential impact of perceived and received support on well-being. Journal of Social and Personal Relationships, 23(1), 117–129. 10.1177/0265407506060182 [DOI] [Google Scholar]
- Rusbult CE, & Van Lange PAM (2003). Interdependence, interaction, and relationships. Annual Review of Psychology, 54, 351–375. 10.1146/annurev.psych.54.101601.145059 [DOI] [PubMed] [Google Scholar]
- Ryon HS, & Gleason MEJ (2018). Reciprocal support and daily perceived control: Developing a better understanding of daily support transcations across a major life transition. Journal of Personality and Social Psychology, 115, 1034–1053. 10.1037/pspi0000141 [DOI] [PubMed] [Google Scholar]
- Seidman G, Shrout PE, & Bolger N (2006). Why is enacted social support associated with increased distress? Using simulation to test two possible sources of spuriousness. Personality and Social Psychology Bulletin, 32(1), 52–65. 10.1177/0146167205279582 [DOI] [PubMed] [Google Scholar]
- Shrout PE, Bolger N, Iida M, Burke C, Gleason MEJ, & Lane SP (2010). The effects of daily support transactions during acute stress: Results from a diary study of bar exam preparation In Sullivan K & Davila J (Eds.), Support Processes in Intimate Relationships (pp. 175–199). New York: NY: Oxford University Press; 10.1093/acprof:oso/9780195380170.003.0007 [DOI] [Google Scholar]
- Shrout PE, Herman CM, & Bolger N (2006). The costs and benefits of practical and emotional support on adjustment: A daily diary study of couples experiencing acute stress. Personal Relationships, 13(1), 115–134. 10.1111/j.1475-6811.2006.00108.x [DOI] [Google Scholar]
- Shrout PE, & Lane SP (2012). Psychometrics In Mehl MR & Conner TS (Eds.), Handbook of Research Methods for Studying Daily Life (pp. 302–320). New York, NY: Guiford Press. [Google Scholar]
- Shrout PE, Stadler G, Lane SP, McClure MJ, Jackson GL, Clavél FD, … Bolger N (2018). Initial elevation bias in subjective reports. Proceedings of the National Academy of Sciences, 115, E15–E23. 10.1073/pnas.1712277115 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Siewert K, Antoniw K, Kubiak T, & Weber H (2011). The more the better? The relationship between mismatches in social support and subjective well-being in daily life. Journal of Health Psychology, 16(4), 621–631. 10.1177/1359105310385366 [DOI] [PubMed] [Google Scholar]
- Smith JA, Braunack-Mayer A, Wittert G, & Warin M (2007). “I’ve been independent for so damn long!”: Independence, masculinity and aging in a help seeking context. Journal of Aging Studies, 21(4), 325–335. 10.1016/j.jaging.2007.05.004 [DOI] [Google Scholar]
- Smith J, & Goodnow JJ (1999). Unasked-for support and unsolicited advice: Age and the quality of social experience. Psychology and Aging, 14(1), 108–121. 10.1037//0882-7974.14.1.108 [DOI] [PubMed] [Google Scholar]
- Swann WB, De La Ronde C, & Hixon JG (1994). Authenticity and positivity strivings in marriage and courtship. Journal of Personality and Social Psychology, 66(5), 857–869. 10.1037//0022-3514.66.5.857 [DOI] [PubMed] [Google Scholar]
- Taylor SE (2012). Social support: A review In Friedman HS (Ed.), The Oxford Handbook of Health Psychology (pp. 189–214). New York: NY: 10.1093/oxfordhb/9780195342819.013.0009 [DOI] [Google Scholar]
- Thorsteinsson EB, & James JE (1999). A meta-analysis of the effects of experimental manipulations of social support during laboratory stress. Psychology & Health, 14(5), 869–886. 10.1080/08870449908407353 [DOI] [Google Scholar]
- Uchino BN (2009). Understanding the links between social support and physical health of perceived and received rupport, Perspectives on Psychological Science, 4(3), 236–255. [DOI] [PubMed] [Google Scholar]
- Vangelisti AL (2009). Challenges in conceptualizing social support. Journal of Social and Personal Relationships, 26(1), 39–51. 10.1177/0265407509105520 [DOI] [Google Scholar]
- White RW (1959). Motivation reconsidered: The concept of competence. Psychological Review, 66, 297–333. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.