More social support is associated with more positive mood but excess support is associated with more negative mood

Abstract

Support gaps occur when an individual receives less or more of a certain type of social support than the individual desires. Existing research suggests that support gaps are associated with unwanted relationship and psychological outcomes. However, many of the outcomes associated with support gaps are also associated with received support and support availability—and it is not clear that support gaps explain anything about such outcomes beyond what is known from these longer established constructs. In samples of community members (n = 430) and students (n = 755), we examined associations between received support, available support, and support gaps and several positive and negative personal and relationship outcomes. Whereas received support and available support were more strongly associated with increased positive outcomes than decreased negative outcomes, support gaps were more strongly associated with increased negative outcomes than decreased positive outcomes. When all three support constructs were simultaneously included in multilevel structural equation models, they each demonstrated incremental predictive value for most of the outcomes we examined. A full understanding of the social contributions to mood and relationship outcomes requires assessment of received support, available support, and support gaps.

Keywords

Depression mood perceived support psychological well-being received support relationship satisfaction support adequacy support gaps

Introduction

Social support is regarded as a meta-construct because researchers have developed a number of ways of conceptualizing how help from one person might benefit another (Vaux, 1987). Recently, support researchers have begun to emphasize the importance of “support gaps” (McLaren & High, 2015). Support gaps are discrepancies between the amount of support that a support recipient desires and the amount of support that recipient has received. Unlike previous conceptualizations of social support which have generally emphasized problems due to lack of support, the support gap concept suggests that both too little and too much support can be undesirable. The purpose of this study was to test the incremental validity of the support gap concept over these longer established approaches to conceptualizing social support.

Stress and coping theory

One well-established approach to understanding the benefits of social support is stress and coping theory. According to stress and coping theory, the demands of an individual’s environment lead to or expose deficits in the individual’s psychological resources. Receiving social support such as encouragement, advice, concern, and practical assistance can help compensate for those deficits (Lakey & Cohen, 2000). Studies on stress and coping theory usually measure “received support” which is the amount of support an individual has received in the recent past. Stress and coping theory predicts that individuals who have received more support will experience better personal and relational outcomes.

Social cognitive theory

The social cognitive theory of social support is another well-established approach. This theory puts less emphasis on recent experiences of support and more emphasis on persistent beliefs about social support providers (Lakey & Drew, 1997). The focal construct of this theory is perceived support availability which is the belief that support will be available when it is needed. Perceived support availability is sometimes referred to simply as perceived support. Social cognitive theory predicts that people who perceive that support is available to them derive confidence and reassurance from that belief, which in turn promote adaptive psychological and interpersonal functioning (Lakey & Drew, 1997).

Support gaps

Both stress and coping theory and social cognitive theory take a “more is better” approach to social support. However, researchers such as Brock and Lawrence (2010) have proposed that receiving too much support can be aversive. Social support can come in different forms and individuals differ regarding how much of each type of support they prefer (High & Steuber, 2014). The concept of a support gap has been developed to describe discrepancies between the amount of a given type of support the individual desires and the amount the individual actually receives. Gaps between desired and received support are created when the individual receives too much or too little of any type of support. This concept has also been referred to as support inadequacy (Dehle, Larsen, & Landers, 2001). Insufficient support may fail to help the recipient meet the demands of the environment or suggest that the provider lacks interest in the recipient’s well-being. Excessive support may feel intrusive or communicate that the support provider lacks confidence in the recipient’s problem-solving abilities (Brock & Lawrence, 2009).

Although support gaps have been linked to a growing number of unwanted personal and relationship consequences, it is notable that many of these unwanted consequences are the same or similar to the consequences associated with lacking received or available support. Most studies on support gaps have not compared the explanatory power of this construct to that of the longer established conceptualizations of social support (e.g., Bar-Kalifa & Rafaeli, 2013; Brock & Lawrence, 2008, 2009; Dehle et al., 2001; Faw, Harvey, & Feng, 2018; High & Crowley, 2018; High & Steuber, 2014; McLaren & High, 2015; Priem & Solomon, 2015). It is important to know whether support gaps reflect something beyond low amounts of received or available support. This issue is particularly significant given that some studies have suggested that excess support is not harmful (Bar-Kalifa & Rafaeli, 2013) or in some cases even helpful (Faw et al., 2018; High & Crowley, 2018; McLaren & High, 2015; Siewert, Antoniw, Kubiak, & Weber, 2011). Similarly, it is also important to know whether support gaps fully account for any associations between received support, available support, and their putative benefits.

Benefits of social support

Social support can have benefits both for the support recipient and for the recipient’s relationships. These benefits can come in the form of increases in positively valenced feelings and/or decreases in negatively valenced feelings.

Personal benefits

Many studies have focused on the potential of social support to prevent or alleviate symptoms of depression. Previous studies have consistently found that perceived support availability is correlated with lower symptoms of depression, but received support often shows no association, or even a positive association with depressive symptoms (Nurullah, 2012; Uchino, Bowen, Carlisle, & Birmingham, 2012). One study on support gaps in marriage found that greater gaps were associated with more depressive symptoms (Dehle et al., 2001), but this finding has not been consistently replicated (Nylen, O’Hara, & Engeldinger, 2013). Other research has suggested that not all gaps are equally related to negative feelings. Two studies on perceived stress (Siewert et al., 2011) and general negative mood (Bar-Kalifa & Rafaeli, 2013) found that insufficient support was related to increased negative affect whereas excess support was sometimes related to decreased negative affect.

Positive feelings

Researchers in the field of positive psychology have emphasized that most people want more out of life than to avoid feeling depressed (Fredrickson, 1998). Qualities of psychological well-being such as pride, interest, hope, and a sense of purpose are at least partly independent of negative affect (Diener & Emmons, 1984). Unlike the case of depression, both received support and perceived support availability are linked with positively valenced individual outcomes such as a sense of self-efficacy (Kerpelman, Eryigit, & Stephens, 2008; Luszczynska, Mohamed, & Schwarzer, 2005), life satisfaction (Finch et al., 1997; Stålnacke, 2007), and a sense of meaning in life (Krause, 2007). The limited research on support gaps and general positive feelings is less clear. High and Crowley (2018) found that insufficient support was associated with a decreased ability to positively reappraise stressful situations, whereas excess support was generally associated with an increase in positive reappraisals. Bar-Kalifa and Rafaeli (2013) generally failed to find significant associations between support gaps and positive mood.

Relational benefits

In addition to benefitting the individual, social support also appears to have benefits for the individual’s relationships. Multilevel studies on social support in which support is measured from multiple relationships have shown that support and relational outcomes are influenced by both characteristics of the recipient such as gender and personality and characteristics of the provider, such as how the provider and the recipient are related (e.g., Lakey & Scoboria, 2005; Lakey, Orehek, Hain, & VanVleet, 2010; Williamson & O’Hara, 2017). When relationship-specific constructs such as support and relationship satisfaction are measured for multiple providers, it is possible to partition the covariance between those constructs into recipient-level and provider-level correlations. It is possible to determine, for example, if people tend to be more satisfied with the providers who give them the most support (provider-level) apart from the tendency of people who receive more support from their providers in general to be more satisfied with their relationships in general (recipient-level). Although most social support studies do not distinguish between recipient- and provider-level associations, those that do offer some unique insights into the connection between social support and its putative relational benefits.

Positive feelings

One potential benefit of social support is improved relationship satisfaction. Relationship satisfaction has traditionally been conceptualized as single dimension from high dissatisfaction to high satisfaction. Measured in this way, relationship satisfaction is connected with higher received support (Rogge, Fincham, Crasta, & Maniaci, 2017), higher support availability (Mattson, Rogge, Johnson, Davidson, & Fincham, 2013), and fewer support gaps (Dehle et al., 2001; Lawrence et al., 2008). However, a more nuanced approach to relationship evaluations has shown that satisfaction and dissatisfaction are two correlated (r = −.39) but separated dimensions (Fincham & Linfield, 1997). Findings related to support gaps and purely positive relational evaluations have been mixed. Dehle, Larsen, and Landers (2001) found no significant association between support gaps and positive evaluations. Bar-Kalifa and Rafaeli (2013) found that insufficient support was associated with fewer positive evaluations of the support provider but did not replicate this finding for excess support. Faw, Harvey, and Feng (2018) found that excess esteem support was actually related to increased relationship satisfaction.

Other research has considered the moods that support recipients’ experience when with specific support providers. Multilevel studies have shown that people who tend to receive and perceive more support also tend to experience more positive mood with their providers (Lakey & Scoboria, 2005; Lakey et al., 2010). That finding reflects a recipient-level process. At the provider level, people tend to experience the most positive mood when they are with providers who are the most supportive and the most available for support. Similar multilevel studies have not been conducted on support gaps. When examining support gaps from a single provider, McLaren and High (2015) found that insufficient emotional support was related to less esteem improvement, whereas excess emotional support was related to greater esteem improvement.

Negative feelings

Separate from relationship satisfaction, relationship dissatisfaction has been connected to less received support (Rogge et al., 2017) and less available support (Mattson et al., 2013). Support gaps have also been associated with increased dissatisfaction (Dehle et al., 2001). Although insufficient support has been linked with increased negative feelings about the support provider, some forms of excess support have been linked with decreased negative feelings about the support provider (Bar-Kalifa & Rafaeli, 2013; McLaren & High, 2015).

With regard to mood during interactions with support providers, McLaren and High (2015) found that recipients generally felt more hurt after interacting with providers who gave insufficient support but less hurt when providers gave some types of excessive support. In multilevel studies, available support is consistently associated with less negative mood at both the recipient and provider levels (Lakey & Scoboria, 2005). Interestingly, individuals who tend to receive the most support overall tend to experience the most negative mood when they are with their support providers, even though those same recipients tend to experience the least negative affect when they are with the providers who support them the most (Lakey et al., 2010).

Incremental validity of the support gap concept

The above studies have done much to establish the importance of the support gap concept. However, they have not shown that support gaps explain something about an individual's social or psychological problems that cannot be explained by deficits in received or available support. There are several studies that indicate some promise for the support gap as an incrementally valid correlate of putative social support outcomes. For instance, Lawrence et al. (2008) showed that support gaps were associated with decreased marital satisfaction even after controlling for received support. Matsunaga (2011) found that both received support and support gaps were generally predictive of cognitive reappraisals of bullied college students. By contrast, Davis and High (2019) generally failed to find significant associations between support gaps and the outcomes they examined once they controlled for received support. The only study we identified controlling for perceived support found that support gaps were associated with increased negative affect and perceived stress (Siewert et al., 2011). With support gaps included in the model, perceived support showed no significant association with either outcome.

The current study

Both stress and coping theory and social cognitive theory propose that more support is better. However, the support gap concept suggests that too much support can be problematic. Support gaps have been connected with a number of psychological and relational outcomes previously connected with received support and support availability, but it is not clear that support gaps reflect something more than a deficiency in received or available support. A number of findings on support gaps reviewed above are more consistent with stress and coping theory in that they show potential benefits of excess support. For researchers and clinicians interested in the value of the support gap concept, it is important to know whether problems associated with support gaps are likely to be driven by a mismatch between desired and received support and not simply a lack of received or available support.

In examining the incremental value of the support gap concept, we sought to answer two research questions. The first question was whether support gaps show a pattern of correlations with positive and negative outcomes similar to the patterns shown by received or available support. Received support is generally not related to reductions in negative outcomes but substantially connected to increases in positive outcomes (Finch et al., 1997; Nurullah, 2012; Rogge et al., 2017; Uchino et al., 2012). Perceived support has been reliably connected to reduced negative outcomes but also appears more strongly related to positive outcomes (e.g., Aquino, Russell, Cutrona, & Altmaier, 1996; Jenkins, Belanger, Connally, Boals, & Durón, 2013; Lakey et al., 2010). Without any a priori hypotheses, we wanted to know if support gaps would be differentially related to positive and negative outcomes.

In order to answer this first question, we sought to employ a multilevel approach similar to that of Lakey and Scoboria (2005). Some studies on support gaps have examined multiple types of relationships (Davis & High, 2019; High & Crowley, 2018; High & Steuber, 2014), but these studies have not made full use of the multilevel design. Multilevel analyses can show the extent to which support gaps are driven by recipient- versus provider-level processes. They can also show whether relationship outcomes are associated with support gaps at the provider level after accounting for covariance at the recipient level. For instance, a study on support gaps in romantic relationships might show that individuals tend to experience more negative affect around their partners when there are greater gaps in support. A multilevel study on multiple relationships could show that the association between support gaps and negative affect is driven primarily by recipient level process—perhaps people who are broadly prone to discontentment tend to report greater support gaps and more negative affect in all of their relationships. Alternatively, a multilevel study could show that the association between support gaps and negative affect is also manifested at the provider level. A significant association at the provider level would indicate meaningful differences between providers such that support recipients tend to experience more negative affect when interacting with specific providers whose support is least consistent with the recipient’s desires.

Our second research question was whether support gaps explain differences in personal and relationship functioning after accounting for received and available support. We also wanted to know if support gaps would fully account for any associations between received support, available support, and our putative correlates. The limited previous research pertinent to this question suggests that all three support constructs would be incrementally valid, at least for predicting some personal and/or relational outcomes.

Method

Participants

In order to evaluate the generalizability of our findings, we recruited two samples for this study. Only individuals aged 18 or older and fluent in English were allowed to participate.

Community sample

The first sample consisted of community-dwelling adults in the United States recruited from ResearchMatch.org, an online registry of individuals interested in research studies. All constructs were measured through an online survey, and community participants were compensated $10 for taking part in the study. We sent recruitment e-mails to 4,382 participants on the ResearchMatch.org registry and 611 began the survey. It was completed by 461 respondents, of whom 31 gave responses missing 20% or more of the items on one or more measures. All cases were screened for indicators of invalid responding such as unusually rapid completion time and response sets. No invalid responses were identified in the community sample. The final sample thus included 430 participants. Table 1 includes demographic characteristics of the recipients, and Table 2 includes characteristics of their providers.

Table 1.

Demographic characteristics.

	Student (N = 755)	Community (N = 430)
Female (%)	511 (68%)	284 (66%)
Race
Asian or Pacific Islander	76 (10%)	9 (2%)
Black/African American	28 (4%)	44 (10%)
Caucasian, not of Hispanic origin	589 (78%)	352 (82%)
Hispanic of any origin	42 (6%)	12 (3%)
Other	20 (3%)	12 (3%)
Household income
Less than $15,000 per year	157 (21%)	34 (8%)
$15,000–$30,000	37 (5%)	57 (13%)
$30,000–$45,000	58 (8%)	68 (16%)
$45,000–$60,000	82 (11%)	50 (12%)
$60,000–$75,000	70 (10%)	49 (12%)
$75,000–$90,000	82 (11%)	53 (12%)
More than $90,000	250 (34%)	114 (27%)
Education
Did not graduate high school		1 (0.2%)
High school/GED		36 (8%)
Vocational or associate’s degree		50 (12%)
Currently in college		49 (11%)
Bachelor’s degree		129 (30%)
Master’s degree		130 (30%)
Advanced (MD, PhD, JD, etc.)		34 (8%)
Relationship status
Single	426 (56%)	128 (30%)
Dating (or engaged)	326 (43%)	99 (23%)
Married (together)	2 (∼0%)	194 (45%)
Married (but separated)	0 (0%)	9 (2%)
Employment status
Full time (30+ hr/week)	5 (1%)	283 (66%)
Part time	351 (47%)	63 (15%)
Unemployed (looking for a job)	160 (21%)	25 (6%)
Not employed, not looking for a job	238 (31%)	57 (13%)

Note. Not all participants reported all demographic data.

Table 2.

Frequency of provider gender and provider relationship types.

	Student	Community
Male recipients
Male providers	437 (60%)	206 (48%)
Female providers	288 (40%)	228 (52%)
Female recipients
Male providers	454 (30%)	264 (31%)
Female providers	1,074 (70%)	588 (69%)
Friends	929 (41%)	429 (33%)
Coworkers	21 (1%)	130 (10%)
Siblings	225 (10%)	132 (10%)
Parents	728 (32%)	236 (18%)
Other family	83 (4%)	111 (9%)
Partners	239 (11%)	223 (17%)
Other	40 (2%)	27 (2%)
Live together	660 (29%)	276 (22%)

Student sample

The second sample consisted of college students from a large Midwestern university. The students also completed the study online. They took part in the study as one option to fulfill a course requirement in introductory psychology. A total of 967 students began the survey and 793 finished it. Of those 793, 11 gave what appeared to be invalid responses in that they finished the survey substantially faster than our pilot testers and gave the same response to all of the items on one or more questionnaires. Of the 782 valid responses, 27 were excluded because they had 20% or more missing data on one or more measures. The final sample thus included 755 participants. Table 1 includes demographic characteristics of the recipients, and Table 2 includes characteristics of their providers.

Procedures

In completing the study’s survey, participants were asked to identify three support providers by responding to the following prompt: “We want to know about the important people in your life. Who are the three adults (or peers) you talk with and/or spend time with the most?” We asked participants to identify three providers in order to remain consistent with previous multilevel studies on social support (e.g., Lakey et al., 2010). Participants completed measures of support availability, received support, support gaps, relationship-specific affect, and relationship satisfaction for each provider separately.

Measures

Perceived support availability

The social support that participants perceived to be available from each provider was measured using the Support scale of the Quality of Relationships Inventory (Pierce, Sarason, & Sarason, 1991). The Support scale includes 7 items which participants rate on a 4-point scale from “not at all” to “very much.” We computed the internal consistency reliability for all provider-specific measures at both the recipient level and the provider level. These computations were done using the lme4 package in R (Bates, Maechler, Bolker, & Walker, 2015) and formulas provided by Lakey, Orehek, Hain, and VanVleet (2010). They are described in more detail in a previous publication on these data (Williamson & O’Hara, 2017). Internal consistency reliabilities for support availability were >.80 for the recipient and provider components in both samples.

Received support

Received support was measured using the Support in Intimate Relationships Rating Scale—Revised (SIRRS-R; Barry, Bunde, Brock, & Lawrence, 2009). The SIRRS-R consists of emotional support, informational support, instrumental support, and physical comfort subscales which were aggregated for the present analyses. Received support over the preceding month was rated on a 5-point scale (“not at all,” “once or twice,” “about once a week,” “several times a week,” “about every day”). Internal consistencies for total received support were >.80 for the recipient and provider components in both samples.

Support gaps

Support gaps were measured with a second form of the SIRRS-R. For each item, participants were shown how much support they reported receiving and were asked how much support they would have liked to have received (“a lot less” (2), “somewhat less” (1), “about the same” (0), “somewhat more” (1), “a lot more” (2)). Higher scores indicate a greater support gap. Insufficient support was computed by summing responses of “somewhat more” (1) and “a lot more” (2). Excess support was computed by summing responses of “somewhat less” (1) and “a lot less” (2). Internal consistency reliabilities were >.80 for all support gaps, in both samples, at both levels.

Relationship-specific affect

Participants completed the Positive and Negative Affect Scales (PANAS; Watson, Clark, & Tellegen, 1988) rating the affective states they typically experience when interacting with each provider. These scales contain 10 items each to which participants responded on a 5-point scale ranging from “not at all” to “extremely.” Internal consistency reliabilities were >.80 for both positive and negative affect, in both samples, at both levels.

Relationship satisfaction

Participants’ positive and negative feelings about their relationships with each provider were measured with an adapted version of the Positive and Negative Quality in Marriage Scale (PNQMS; Fincham & Linfield, 1997). For use in this study, the wording “your spouse” in the PNQMS items was changed to the names of the support providers identified by the participant. This scale is composed of two 3-item subscales, relationship satisfaction and relationship dissatisfaction. All items were rated on an 11-point scale from “not at all” to “extremely.” In the current study, both scales showed good internal consistency reliability (α > .80) in both samples, at both levels.

Psychological health

Depression (20 items) and psychological Well-being (8 items) were measured with two scales from the Inventory of Depression and Anxiety Symptoms (IDAS; Watson et al., 2007). These scales measure mood over the preceding 2 weeks and ask respondents to rate their mood experiences on a 5-point scale ranging from “not at all” to “extremely.” In both samples, both the Depression and Well-being scales showed good internal consistency reliability (α > .80). In the community sample, the mean Depression score was M = 41.57 (SD = 15.21) and the mean Well-being score was M = 26.44 (SD = 7.59). In the student sample, the mean Depression score was M = 44.31 (SD = 14.21) and the mean Well-being score was M = 28.90 (SD = 6.22).

Data analysis

Missing data

Cases with 20% missing data on one or more measures were excluded from analysis (Graham, Cumsille, & Elek-Fisk, 2003). For the remaining cases, missing item scores were estimated using the mean of the scores for completed items of the same scale (Downey & King, 1998).

Variance components

All analyses were conducted in Mplus version 8 (Muthén & Muthén, 2017). Variance in constructs measured for each provider was partitioned into recipient, provider, and error components using multilevel structural equation modeling. Similar to previous multilevel studies on social support (e.g., Lakey et al., 2010), we averaged the odd and even items of each measure to create two indicators for each construct. These indicators were nested within the providers who were in turn nested within the recipients (see Figure 1). The level of odd and even indicators was created to avoid attributing unsystematic error variance to differences between providers. Differences between variance components were tested using Wald χ² tests generated by the “model test” function.

Figure 1.

Levels in the multilevel design.

Associations between the constructs were evaluated using multilevel structural equation modeling. Associations between provider-specific outcomes were modeled at all three levels (error, providers, recipients) in order to avoid misattributing covariance between constructs at the error level to covariance between those same constructs at the provider level. Depression and well-being were modeled only at the recipient level. Differences between associations were evaluated with Wald χ² tests. In the regression models, correlations among all independent variables were included at all levels such that the models were just-identified (χ² = 0; RMSEA = 0; CFI = 1).

Results

Variance components

The variance components of the provider-specific variables are presented in Table 3. Both the recipient and provider components of all constructs were statistically significant. In the community sample, the provider components were significantly larger than the recipient components for all support variables. With respect to support gaps specifically, this finding indicates that differences in support gaps were driven more by the recipients’ relationships with individual providers than by characteristics of the recipients themselves. In the student sample, the provider component was again larger than the recipient component for available support. The recipient and provider components were roughly equal for received support and support gaps.

Table 3.

Variance components of provider-specific variables.

	Students		Community
	${\hat{σ}}^{2}$	% of Total variance	${\hat{σ}}^{2}$	% of Total variance
Received support
Recipient level	0.22^A	30%	0.40^A	47%
Provider level	0.48^B	66%	0.42^A	50%
Support availability
Recipient level	0.11^A	22%	0.12^A	29%
Provider level	0.26^B	52%	0.19^B	48%
Insufficient support
Recipient level	0.01^A	28%	0.02^A	56%
Provider level	0.005^B	65%	0.02^A	39%
Excess support
Recipient level	0.03^A	29%	0.05^A	43%
Provider level	0.07^B	66%	0.06^A	53%
Total support gap
Recipient level	0.08^A	30%	0.08^A	50%
Provider level	0.04^B	66%	0.07^A	46%
Positive affect
Recipient level	0.38^A	48%	0.40^A	54%
Provider level	0.28^B	35%	0.19^B	26%
Negative affect
Recipient level	0.09^A	35%	0.18^A	55%
Provider level	0.12^A	49%	0.09^B	27%
Satisfaction
Recipient level	0.54^A	28%	0.91^A	48%
Provider level	1.12^B	59%	0.70^A	37%
Dissatisfaction
Recipient level	2.21^A	38%	2.87^A	52%
Provider level	3.16^B	54%	2.13^B	39%

Note. ${\hat{σ}}^{2}$ = raw variance component. All components were significantly different from zero. The recipient and provider components for each variable were compared, and components with different superscripts are significantly different.

Correlations between support constructs

Correlations between the support constructs at the recipient and provider levels are presented in Table 4. Insufficient support, excess support, and total support gaps did not show any associations with received support or available support above .31. Associations of this magnitude indicate that support gaps are empirically distinct from these other social support constructs.

Table 4.

Correlations between social support variables.

Variable	Received support	Support availability	Insufficient support	Excess support	Total support gap
Received support
Recipient level		.44*	−.09	.21*	.12
Provider level		.56*	−.06	−.18*	−.19*
Support availability
Recipient level	.36*		−.31*	−.08	−.23*
Provider level	.58*		−.07	−.19*	−.21*
Insufficient support
Recipient level	−.04	−.17*		.06	—
Provider level	−.05	−.13*		−.08*	—
Excess support
Recipient level	.04	−.15*	.05		—
Provider level	.02	−.16*	.01		—
Total support gap
Recipient level	.03	−.20*	—	—
Provider level	.00	−.20*	—	—

Note. Correlations for the community sample are presented below the diagonal. Correlations for the student sample are presented above the diagonal.

*p < .05.

Comparing associations

The correlations between the support constructs and psychological outcomes are presented in Table 5. Available support was correlated with increased well-being and fewer symptoms of depression—though the associations with well-being were stronger. Received support was correlated with increased well-being but also with increased depression. Support gaps were not consistently related to psychological well-being but excess support in particular was associated with increased depressive symptoms.

Table 5.

Correlations between social support and personal outcomes.

(Community/student)	Positive valence	Negative valence	Δ\|r\|
(Community/student)	Psychological well-being	Depression	Δ\|r\|
Received support	.25/.25	.22/.12	.03/.13*
Perceived support	.53/.36	−.23/−.21	.30/.15
Insufficient support	−.08/−.07	.20/.10	−.13/−.03
Excess support	−.26*/−.02	.42/.35	−.16/−.33
Total support gap	−.27*/−.06	.46/.33	−.20/−.28

Note. Δ|r| = difference between the absolute value of the correlations with the positively valenced outcome and the negatively valenced outcome.

*p < .05.

A similar pattern of findings emerged from the correlations between the support constructs and the provider-specific outcomes (Table 6). More received support, more available support, and fewer support gaps were fairly consistently associated with increased positive mood, increased relationship satisfaction, decreased negative mood, and decreased relationship dissatisfaction. Both insufficient and excess support were generally associated with fewer positive feelings and more negative feelings. The primary exception to the pattern was that recipients who tended to receive the most support overall also tended to experience the most negative affect around their providers. Recipients also experienced the most negative mood around the providers who gave them the most support.

Table 6.

Correlations between social support and relational outcomes.

Received support	Positive valence	Negative valence	Δ\|r\|
	Interpersonal positive affect	Interpersonal negative affect
	(Community/student)	(Community/student)
Recipient	.36/.53	.20/.24	.17/.29*
Provider	.44/.46	.05*/.01	.39/.45
Perceived support
Recipient	.64/.54	−.19/−.28	.45/.26
Provider	.72/.58	−.29/−.13	.43/.46
Insufficient support
Recipient	−.15/−.16	.32/.16	−.17/.00*
Provider	−.19/−.07	.26/.10	−.07/.02*
Excess support
Recipient	−.07/.12	.21/.48	−.14/−.36*
Provider	−.15/−.05	.42/.17	−.27/−.12
Total support gap
Recipient	−.12/−.01*	.31/.48	−.19/−.48*
Provider	−.21/−.08	.49/.20	−.28/−.13
	Relationship satisfaction	Relationship dissatisfaction
Received support	(Community/student)	(Community/student)
Recipient	.24/.21	.13/−.04	.11/.16*
Provider	.38/.38	−.11/−.20	.27/.18
Perceived support
Recipient	.68/.62	−.29/−.38	.39/.23
Provider	.66/.57	−.37/−.32	.29/.25
Insufficient support
Recipient	−.36/−.30	.12/.15	.24/.15
Provider	−.18/−.13	.21/.11	−.02/.02*
Excess support
Recipient	.02/−.07*	.20/.26	−.22/−.18
Provider	−.19/−.10	.36/.23	−.16/−.13
Total support gap
Recipient	−.11/−.23*	.22/.29	−.12/−.06
Provider	−.25/−.16	.41/.26	−.16/−.11

Note. Δ|r| = difference between the absolute value of the correlation with the positively valenced outcome and the correlation with the negatively valenced outcome.

*p < .05.

Again, the positive and negative outcomes showed a distinct pattern of associations with the support constructs. Received support and perceived support were more strongly associated with increased positive mood and increased relationship satisfaction than they were with the negatively valenced constructs. The opposite pattern emerged for excessive support. Excess support was more strongly associated with increased negative outcomes than with decreased positive outcomes.

Incremental validity of support constructs

Next, received support, available support, insufficient support, and excess support were included simultaneously as predictors of depression and well-being (Table 7), affect during interactions with provider (Table 8), and relationship satisfaction and dissatisfaction (Table 9). These analyses showed that almost all of the associations that were significant when estimated without controlling for other support constructs were also significant in the models with multiple independent variables. All of the outcomes were significantly predicted by at least two support constructs, and the majority were predicted by three or more. These findings demonstrate that support gaps, as well as received support and available support, are incrementally valid social support constructs.

Table 7.

Multilevel regression predicting psychological outcomes.

	Community	Student
Well-being	β	β
Received support	.09	.11*
Support availability	.47*	.33*
Insufficient support	.02	.05
Excessive support	−.20*	−.03
R ²	.32	.14
Depression	β	β
Received support	.30*	.17*
Support availability	−.25*	−.26*
Insufficient support	.15*	.01
Excessive support	.37*	.30*
R ²	.31	.18

Note. β = standardized path coefficients. R ² = total variance explained by all independent variables.

*p < .05.

Table 8.

Multilevel regression predicting interpersonal affect.

	Community	Student
Interpersonal positive affect	β	β
Recipient level
Received support	.16*	.34*
Support availability	.57*	.40*
Insufficient support	−.04	−.01
Excessive support	.01	.08*
R ²	.43	.41
Provider level
Received support	.04	.20*
Support availability	.68*	.49*
Insufficient support	−.10*	−.02
Excessive support	−.04	.08*
R ²	.54	.37
Interpersonal negative affect	β	β
Recipient level
Received support	.28*	.33*
Support availability	−.22*	−.38*
Insufficient support	.28*	.05
Excessive support	.15	.38*
R ²	.22	.38
Provider level
Received support	.27*	.14*
Support availability	−.36*	−.16*
Insufficient support	.22*	.10*
Excessive support	.36*	.17*
R ²	.33	.06

Note. β = standardized path coefficients. R ² = total variance explained by all independent variables.

*p < .05.

Table 9.

Multilevel regression predicting relationship evaluations.

	Community	Student
Relationship satisfaction	β	β
Recipient level
Received support	−.01	−.06
Support availability	.66*	.60*
Insufficient support	−.25*	−.12*
Excessive support	.14*	−.01
R ²	.54	.40
Provider level
Received support	.01	.09*
Support availability	.63*	.51*
Insufficient support	−.10*	−.09*
Excessive support	−.09*	.00
R ²	.46	.33
Relationship dissatisfaction	β	β
Recipient level
Received support	.25*	.09
Support availability	−.35*	−.40*
Insufficient support	.06	.02
Excessive support	.13	.20*
R ²	.17	.21
Provider level
Received support	.11*	−.01
Support availability	−.36*	−.27*
Insufficient support	.17*	.10*
Excessive support	.29*	.19*
R ²	.26	.14

Note. β = standardized path coefficients. R ² = total variance explained by all independent variables.

*p < .05.

Discussion

Summary and interpretation of the findings

The pattern of associations that emerged from this study paints a complex picture of the relationship between affect and social experiences. In interpreting these findings, it is worth reemphasizing that positive and negative feelings are only moderately correlated (Diener & Emmons, 1984). Many people experience a fairly high frequency and intensity of both positive and negative emotions, and individuals can also have both strong positive and strong negative feelings about any given relationship (Fincham & Linfield, 1997).

Many studies have found that simply receiving more support is not associated with less negative affect. Others, like ours, have found that received support and negative affect are even positively correlated (Cruza-Guet, Spokane, Caskie, Szapocznik, & Brown, 2008; Gleason, Iida, Shrout, & Bolger, 2008). One explanation for this finding is that individuals experiencing depressive symptoms may be more likely to seek out or elicit support from their providers (Collins & Feeney, 2000; Revenson & Majerovitz, 1990). That would account for the recipient-level association between received support and negative affect, but does not explain our findings at the provider level. Unlike Lakey et al. (2010), we found that recipients experienced the most negative affect around providers who gave them the most support. The fact that this association held after controlling for available support contradicts any suggestion that received support and negative affect are correlated because people express more negative affect around providers whom they see as more likely to be supportive. The association also held after controlling for excess support, meaning that higher received support is probably not related to higher negative affect simply because some of the support is unwanted. It is possible that received support partly reflects the frequency of the recipient’s interactions with the provider. A greater frequency of interactions could include a greater frequency of negative interactions. Additionally, research on support visibility suggests that receiving support may sometimes diminish the recipient’s sense of self-efficacy and create unwanted feelings of indebtedness toward the provider (Gleason et al., 2008). Future research could clarify whether received support is correlated with frequency of negative interactions, self-esteem, and feelings of obligation at the provider level.

Although received support was associated with increased negative affect, it was also associated with increased positive affect, even after controlling for available support and support gaps. This finding builds on the research done by Lakey et al. (2010), who examined only bivariate associations between received support and affect, and does much to underscore the importance of receiving support. Even when receiving more support does not increase perceptions of support availability, even though the support may be insufficient or excessive, even though support may not directly reduce negative feelings, receiving more support is still related to greater overall well-being, and more positive feelings for one’s support providers. The potential for receiving support to promote positive feelings is not well-captured by stress and coping theory. This feature of received support may be better explained through theories on collective emotions that describe how social interaction can spread and amplify positive emotional states (von Scheve & Ismer, 2013).

Consistent with a large body of evidence supporting the social cognitive perspective on social support (Uchino et al., 2012), perceived support availability was consistently associated with both increased positive feelings and fewer negative feelings. As was the case for received support, available support was more strongly related to relationship satisfaction and positive emotions than to relationship dissatisfaction and negative emotions. That pattern contrasted the findings for support gaps, particularly excess support which was more strongly related to negative outcomes. Receiving support that is insufficient or excessive seems more likely to lead the recipient to feel irritated or resentful toward the provider than to diminish the positive feelings the recipient has toward that person. As with received support, it is possible that people who are prone to negative mood states tend to attract excess support. However, the fact that we demonstrated this pattern at the provider, as well as the recipient, level indicates that the covariance between excess support and negative mood cannot be fully accounted for by basic personality traits. It is worth noting that the pattern of findings for excess support is similar to the pattern observed for negative social interactions. Negative interactions such as being rejected or insulted by another person appear more strongly related to outcomes such as depression than is available support, whereas available support appears to be more strongly related to positive outcomes such as relationship closeness (Helgeson et al., 2014). Excess support may be experienced as a negative interaction.

Like other studies that have used the SIRRS-R to measure support gaps (e.g., Brock & Lawrence, 2009), we observed that both insufficient support and excess support were associated with undesirable psychological and relational outcomes. These findings are in contrast to other studies that have found some benefits to excess support (Faw et al., 2018; High & Crowley, 2018; McLaren & High, 2015; Siewert et al., 2011). The discrepancy in these findings may be attributable to differences in how excess support has been operationalized. The studies that found benefits for excess support measured that construct by asking participants to rate the amount of support they desired and the amount of support they received. Receiving more support than desired was operationalized as excess support. The SIRRS-R operationalizes excess support as the degree to which participants would have liked to have received less of a given type of support. It appears that many people who receive more support than they desire experience this as extra support, more of a good thing. Only support that is actually unwanted (which appears to be better isolated by the SIRRS-R) seems to be associated with undesirable outcomes.

Improving support transactions

Our findings have a number of implications for improving support transactions. We found significant recipient-level variability in both insufficient and excessive support. This result indicates that there are some people who tend to receive less or more support than they want from all of their providers. Recipients who find that they often fail to get the amounts and types of support they desire may benefit from strategies developed in the literature on couples and interpersonal psychotherapy (e.g., Brock & Lawrence, 2010; Dimidjian, Martell, & Christensen, 2008; Rogge, Cobb, Johnson, Lawrence, & Bradbury, 2002; Stuart & Robertson, 2012). Such strategies include learning about one’s preferences for support, expressing one’s desires explicitly, and providing informative reinforcement when providers attempting to help are successful.

The significant provider-level variability we observed in support gaps indicates that some providers gave support that better matched the recipients’ desires than others. Providers should heed the requests of support recipients, but also know that most people do not ask for support every time they need it (High & Crowley, 2018). It is therefore incumbent upon providers to ask if support is desired and what kind of support the recipient would prefer. Of course, recipients may not always be able to explicitly state their desires. Even recipients who are able to articulate their support needs are likely to appreciate it when those needs are intuited by the provider. Providers, therefore, must also learn to adapt the support they give based on the recipient’s past responses.

Strengths, limitations, and areas for future research

This study was correlational and cross-sectional. Consequently, conclusions about the effects of any support construct on relationship or psychological outcomes must be drawn tentatively. It may be, for instance, that people who are dissatisfied with their relationship for reasons other than the support they receive (e.g., arguments, betrayal of a confidence, infidelity of a romantic partner) are motivated to find fault with the support from that relationship—a kind of negative halo effect (Nisbett & Wilson, 1977). People who are depressed may similarly perceive the support they receive from others as insufficient or excessive due to negative biases in social perception that sometimes accompany depressive episodes (e.g., Park et al., 2016). Longitudinal research has already shown that fewer support gaps improve romantic relationship satisfaction over time, even after accounting for depression and marital stress (Brock & Lawrence, 2008). Future longitudinal studies could examine these processes in non-romantic relationships and focus on depression and psychological well-being as outcomes.

One strength of this study was that we collected two moderately large samples. Although there were occasional discrepancies in statistical significance, the overall pattern of findings replicated quite consistently across the two samples. It is worth noting, however, that both samples—people signed up on ResearchMatch.org and psychology students—might disproportionately contain people high in socially oriented personality traits such as extroversion and agreeableness that are associated with greater levels of received and available support (Williamson & O’Hara, 2017).

Another limitation stems from the fact that this study focused on support from the recipients’ closest providers. Expanding the sample of providers to include more peripheral support figures would almost certainly increase the provider-level variability in received and available support. Provider-level variability in other constructs could increase as well. It is also reasonable to expect that gaps in support from such providers would have less impact on the recipient’s depressive symptoms and overall well-being.

Conclusion

This study demonstrated several pieces of evidence that the support gap concept provides a valuable contribution to prior theory on the benefits of social support. First, both insufficient support and excess support were associated with unwanted outcomes. Second, support gaps were generally more strongly associated with increased negative outcomes than decreased positive outcomes as opposed to received and available support, which were more strongly related to increased positive than decreased negative outcomes. Third, support gaps were associated with most of the outcomes we studied, even after controlling for received and available support–strong evidence for the incremental validity of the support gap concept. All three social processes must be considered in any attempt to fully understand the effects of social interactions on psychological and relationship functioning.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was financially supported by the Society for Clinical and Social Psychology and the University of Iowa Department of Psychological and Brain Sciences.

ORCID iD

J Austin Williamson

Open research statement

As part of IARR’s encouragement of open research practices, the authors have provided the following information: This research was not preregistered. The data used in the research are not publicly available but can be provided by the authors: jaustinwilliamson@augustana.edu. The materials used in the research are not publicly available but can be provided by the authors: jaustinwilliamson@augustana.edu.

References

Aquino

J. A.

Russell

Cutrona

C. E.

Altmaier

E. M.

(1996). Employment status, social support, and life satisfaction among the elderly. Journal of Counseling Psychology, 43, 480–489. doi:10.1037//0022-0167.43.4.480

Bar-Kalifa

Rafaeli

(2013). Disappointment’s sting is greater than help’s balm: Quasi-signal detection of daily support matching. Journal of Family Psychology, 27, 956–967. doi:10.1037/a0034905

Barry

R. A.

Bunde

Brock

R. L.

Lawrence

(2009). Validity and utility of a multidimensional model of received support in intimate relationships. Journal of Family Psychology, 23, 48–57. doi:10.1037/a0014174

Bates

Maechler

Bolker

Walker

(2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67, 1–48. doi:10.18637/jss.v067.i01

Brock

R. L.

Lawrence

(2008). A longitudinal investigation of stress spillover in marriage: Does spousal support adequacy buffer the effects? Journal of Family Psychology, 22, 11–20. doi:10.1037/0893-3200.22.1.11

Brock

R. L.

Lawrence

(2009). Too much of a good thing: Underprovision versus overprovision of partner support. Journal of Family Psychology, 23, 181–192. doi:10.1037/a0015402

Brock

R. L.

Lawrence

(2010). Support adequacy in marriage: Observing the platinum rule. In Sullivan

K. T.

Davila

(Eds.), Support processes in intimate relationships (pp. 3–25). New York, NY: Oxford.

Collins

N. L.

Feeney

B. C.

(2000). A safe haven: An attachment theory perspective on support seeking and caregiving in intimate relationships. Journal of Personality and Social Psychology, 78, 1053–1073. doi:10.1037//0022-3514.78.6.1053

Cruza-Guet

M. C.

Spokane

A. R.

Caskie

G. I. L.

Szapocznik

Brown

S. C.

(2008). The relationship between social support and psychological distress among Hispanic elders. Journal of Counseling Psychology, 55, 427–441. doi:10.1037/a0013501

10.

Davis

S. M.

High

A. C

. (2019). Widening the gap: Support gaps in same race versus different race female friendship dyads. Journal of Social and Personal Relationships, 36, 187–213. doi:10.1177/0265407517722245

11.

Dehle

Larsen

Landers

J. E.

(2001). Social support in marriage. The American Journal of Family Therapy, 29, 307–324. doi:10.1080/01926180152588725

12.

Diener

Emmons

R. A.

(1984). The independence of positive and negative affect. Journal of Personality and Social Psychology, 47, 1105–1117. doi:10.1037/0022-3514.47.5.1105

13.

Dimidjian

Martell

C. R.

Christensen

(2008). Integrative behavioral couple therapy. In Gurman

A. S.

(Ed.), Clinical handbook of couple therapy (pp. 73–103). New York, NY: Guilford Press.

14.

Downey

R. G.

King

C. V.

(1998). Missing data in Likert ratings: A comparison of replacement methods. The Journal of General Psychology, 125, 175–191.

15.

Faw

M. H.

Harvey

Feng

(2018). Replication of Xu & Burleson (2001) “Effects of sex, culture, and support type on perceptions of spousal social support: An assessment of the ‘support gap’ hypothesis in early marriage”. Communication Studies, 69, 314–325. doi:10.1080/10510974.2018.1464043

16.

Finch

J. F.

Barrera

Okun

M. A.

Bryant

W. H. M.

Pool

G. J.

Snow-Turek

A. L.

(1997). The factor structure of received social support: Dimensionality and the prediction of depression and life satisfaction. Journal of Social and Clinical Psychology, 16, 323–342. doi:10.1521/jscp.1997.16.3.323

17.

Fincham

F. D.

Linfield

K. J.

(1997). A new look at marital quality: Can spouses feel positive and negative about their marriage? Journal of Family Psychology, 11, 489–502. doi:10.1037/0893-3200.11.4.489-502

18.

Fredrickson

B. L.

(1998). What good are positive emotions? Review of General Psychology, 2, 300–319. doi:10.1037//1089-2680.2.3.300

19.

Gleason

M. E. J.

Iida

Shrout

P. E.

Bolger

(2008). Receiving support as a mixed blessing: Evidence for dual effects of support on psychological outcomes. Journal of Personality and Social Psychology, 94, 824–838. doi:10.1037/0022-3514.94.5.824

20.

Graham

J. W.

Cumsille

P. E.

Elek-Fisk

(2003). Methods for handling missing data. In Schinka

J. A.

Velicer

W. F.

Weiner

I. B.

(Eds.), Handbook of psychology: Vol. 2. Research methods in psychology (pp. 87–114). Hoboken, NJ: John Wiley.

21.

Helgeson

V. S.

Palladino

D. K.

Reynolds

K. A.

Becker

D. J.

Escobar

Siminerio

(2014). Relationships and health among emerging adults with and without type 1 diabetes. Health Psychology, 3, 1125–1133. doi:10.1037/a0033511

22.

High

A. C.

Crowley

J. L.

(2018). Gaps among desired, sought, and received support: Deficits and surpluses in support when coping with taboo marital stressors. Communication Research, 45, 319–338. doi:10.1177/0093650215626975

23.

High

Steuber

(2014). An examination of support (in)adequacy: Types, sources, and consequences of social support among infertile women. Communication Monographs, 81, 157–188. doi:10.1080/03637751.2013.878868

24.

Jenkins

S. R.

Belanger

Connally

M. L.

Boals

Durón

K. M.

(2013). First-generation undergraduate students’ social support, depression, and life satisfaction. Journal of College Counseling, 16, 129–142. doi:10.1002/j.2161-1882.2013.00032.x

25.

Kerpelman

Eryigit

Stephens

(2008). African American adolescents’ future education orientation: Associations with self-efficacy, ethnic identity, and perceived parental support. Journal of Youth and Adolescence, 37, 997–1008. doi:10.1007/s10964-007-9201-7

26.

Krause

(2007). Longitudinal study of social support and meaning in life. Psychology and Aging, 22, 456–469. doi:10.1037/0882-7974.22.3.456

27.

Lakey

Cohen

(2000). Social support theory and selecting measures of social support. In Cohen

Gordon

L. U.

Gottlieb

B. H.

(Eds.), Social support measurement and interventions: A guide for health and social scientists (pp. 29–52). New York, NY: Oxford.

28.

Lakey

Drew

J. B.

(1997). A social-cognitive perspective on social support. In Pierce

G. R.

Lakey

Sarason

I. G.

Sarason

B. R.

(Eds.), Sourcebook of social support and personality (pp. 107–140). New York: Plenum Press.

29.

Lakey

Orehek

Hain

K. L.

VanVleet

(2010). Enacted support’s links to negative affect and perceived support are more consistent with theory when social influences are isolated from trait influences. Personality and Social Psychology Bulletin, 36, 132–142. doi:10.1177/0146167209349375

30.

Lakey

Scoboria

(2005). The relative contribution of trait and social influences to the links among perceived social support, affect, and self-esteem. Journal of Personality, 73, 361–388. doi:10.1111/j.1467-6494.2005.00312.x

31.

Lawrence

Bunde

Barry

Brock

R. L.

Sullivan

K. T.

Pasch

L. A.

… Adams

E. E

. (2008). Partner support and marital satisfaction: Support amount, adequacy, provision and solicitation. Personal Relationships, 15, 445–463. doi:10.1111/j.1475-6811.2008.00209.x

32.

Luszczynska

Mohamed

Schwarzer

(2005). Self-efficacy and social support predict benefit finding 12 months after cancer surgery: The mediating role of coping strategies. Psychology, Health & Medicine, 10, 365–375. doi:10.1080/13548500500093738

33.

Matsunaga

(2011). Underlying circuits of social support for bullied victims: An appraisal-based perspective on supportive communication and postbullying adjustment. Human Communication Research, 37, 174–206. doi:10.1111/j.1468-2958.2010.01398.x

34.

Mattson

R. E.

Rogge

R. D.

Johnson

M. D.

Davidson

E. K.

Fincham

F. D.

(2013). The positive and negative semantic dimensions of relationship satisfaction. Personal Relationships, 20, 328–355. Retrieved from http://dx.doi.org.proxy.lib.uiowa.edu/10.1111/j.1475-6811.2012.01412.x

35.

McLaren

R. M.

High

A. C.

(2015). The effect of under- and over-benefited support gaps on hurt feelings, esteem, and relationships. Communication Research. Advanced online publication. doi:10.1177/0093650215605155

36.

Muthén

L. K.

Muthén

B. O.

(2017). Mplus user’s guide (8th ed.). Los Angeles, CA: Muthén & Muthén.

37.

Nisbett

R. E.

Wilson

T. D.

(1977). Halo effect: Evidence for unconscious alternation of judgments. Journal of Personality and Social Psychology, 35, 250–256. doi:10.1037/0022-3514.35.4.250

38.

Nurullah

A. S.

(2012). Received and provided support: A review of current evidence and future directions. American Journal of Health Studies, 27, 173–188.

39.

Nylen

K. J.

O’Hara

M. W.

Engeldinger

(2013). Perceived social support interacts with prenatal depression to predict birth outcomes. Journal of Behavioral Medicine, 36, 427–440. doi:10.1007/s10865-012-9436-y

40.

Park

Lee

D. S.

Shablack

Verbuyn

Deldin

Ybarra

… Kross

. (2016). When perceptions defy reality: The relationships between depression and actual and perceived Facebook social support. Journal of Affective Disorders, 200, 37–44. doi:10.1016/j.jad.2016.01.048

41.

Pierce

G. R.

Sarason

I. G.

Sarason

B. R.

(1991). General and relationship-based perceptions of social support: Are two constructs better than one? Journal of Personality and Social Psychology, 61, 1028–1039. doi:10.1037/0022-3514.61.6.1028

42.

Priem

Solomon

D. H.

(2015). Emotional support and physiological stress recovery: The role of support matching, adequacy, and invisibility. Communication Monographs, 82, 88–112. doi:10.1080/03637751.2014.971416

43.

Revenson

Majerovitz

S. D.

(1990). Spouses’ support provision to chronically ill patients. Journal of Social and Personal Relationships, 7, 575–586. doi:10.1177/0265407590074013

44.

Rogge

R. D.

Cobb

R. M.

Johnson

Lawrence

Bradbury

T. N.

(2002). The CARE program: A preventive approach to marital intervention. In Gurman

A. S.

Jacobson

N. S.

(Eds.), Clinical handbook of couple therapy (pp. 420–435). New York: Guilford Press.

45.

Rogge

R. D.

Fincham

F. D.

Crasta

Maniaci

M. R.

(2017). Positive and negative evaluation of relationships: Development and validation of the Positive-Negative Relationship Quality (PN-RQ) scale. Psychological Assessment, 29, 1028–1043. doi:10.1037/pas0000392

46.

Siewert

Antoniw

Kubiak

Weber

(2011). The more the better? The relationship between mismatches in social support and subjective well-being in daily life. Journal of Health Psychology, 16, 621–631. doi:10.1177/1359105310385366

47.

Stålnacke

B. M.

(2007). Community integration, social support and life satisfaction in relation to symptoms 3 years after mild traumatic brain injury. Brain Injury, 21, 933–942. doi:10.1080/02699050701553189

48.

Stuart

Robertson

(2012). Interpersonal psychotherapy: A clinician’s guide. London, UK: Oxford University Press.

49.

Uchino

B. N.

Bowen

Carlisle

Birmingham

(2012). Psychological pathways linking social support to health outcomes: A visit with the “ghosts” of research past, present, and future. Social Science & Medicine, 74, 949–957. doi:10.1016/j.socscimed.2011.11.023

50.

Vaux

(1987). Appraisals of social support: Love, respect, and involvement. Journal of Community Psychology, 15, 493–502. doi:10.1002/1520-6629(198710)15:4<493:: AID-JCOP2290150407>3.0.CO;2-4

51.

von Scheve

Ismer

(2013). Towards a theory of collective emotions. Emotion Review, 5, 406–413. doi:10.1177/1754073913484170

52.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. doi:10.1037/0022-3514.54.6.1063

53.

Watson

O’Hara

M. W.

Simms

L. J.

Kotov

Chmielewski

McDade-Montez

E. A.

… Stuart

. (2007). Development and validation of the Inventory of Depression and Anxiety Symptoms (IDAS). Psychological Assessment, 19, 253–268. doi:10.1037/1040-3590.19.3.253

54.

Williamson

J. A.

O’Hara

M. W.

(2017). Who gets social support, who gives it, and how it’s related to recipient’s mood. Personality and Social Psychology Bulletin, 43, 1355–1377. doi:10.1177/0146167217711936