The influence of intergenerational relationships on marital quality following the death of a parent in adulthood

Abstract

Bereavement not only is an individual psychological experience but also has meaningful social and familial impacts. This article examines how relationship quality with both a deceased and a surviving parent influence adult children’s marital quality over time and whether this differs by gender. Data were drawn from the Longitudinal Study of Generations, a study of three- and four-generation families from Southern California. The sample included married adult children who experienced the death of a parent between survey waves (N = 304). A series of multilevel random effects models were estimated using a before/after loss framework. Analyses revealed that improvements and declines in relationship quality with a surviving parent were related with improvements and declines in marital quality following the death of a first parent, regardless of gender. High pre-loss relationship quality with a deceased parent resulted in improved marital quality only for sons who lost mothers. Findings support the linked lives framework and offer some evidence for the “greedy marriage” thesis.

Keywords

Bereavement intergenerational relations marriage marital quality quantitative analysis

The death of a parent is the most common bereavement experience for adults in the West today (Umberson, 2003). It presents the loss of a significant intergenerational bond that is an important influence throughout adults’ lives (Bengtson, 2001; Marks, Jun, & Song, 2007; Umberson, 1992). The death of a parent is a normative yet profound life course transition (Barner & Rosenblatt, 2008; Marks et al., 2007; Umberson, 1995, 2003). In addition to affecting psychological well-being, drinking behaviors, and physical health (Marks et al., 2007), the death of a parent also affects an individual’s close relationships with siblings, surviving parents, and spouse (Fuller-Thomson, 2000; Rosenblatt & Barner, 2006; Umberson, 1995).

Marriage typically is the closest relationship in adults’ lives (Sarkisian & Gerstel, 2008), and the quality of one’s marriage influences physical and mental health (Carr & Springer, 2010). Marriage is therefore a key relationship to study following parental loss. The limited body of research on parental death and marital outcomes illustrates declines in marital harmony and support as well as increases in marital strain, conflict, and negative behaviors (Rosenblatt & Barner, 2006; Umberson, 1995, 2003). These effects may occur because a parent’s death alters the dynamics of a marital relationship, including desired closeness/distance and the extent to which partners want and provide one another with support (Rosenblatt & Barner, 2006).

However, factors contributing to marital decline—or improvement—following the death of a parent have not been thoroughly examined. In particular, parent–child relationship quality may be an important factor. Parents and children lead “linked lives” across the life course, and the quality of their relationships impacts the quality of adult children’s marriages (Elder, 1999; Elder & Johnson, 2003; Reczek, Liu, & Umberson, 2010). Thus, emotional need may be particularly high after the death of a close parent, putting pressure on a spouse to provide compensatory empathic support (Umberson, 2003). At the same time, marriage may be a “greedy” institution (Sarkisian & Gerstel, 2008), and the death of a close parent may offer the opportunity for adults’ marriages to strengthen further over time (Rosenblatt & Barner, 2006). This article examines whether relationship quality with both deceased and surviving parents influences adults’ marital quality after the death of a parent, using data from 304 married adults who experienced the death of a parent between 1988 and 2005.

Linked lives and the greediness of marriage

Parents and children lead intertwined, mutually influential lives over time; that is, they lead linked lives (Elder, 1999; Elder & Johnson, 2003; Greenfield & Marks, 2006). Intergenerational relationships are becoming increasingly important as multiple generations co-survive for longer periods of time (Bengtson, 2001). Thus, when a parent dies, a bereaved adult child may seek to compensate for losses that can include the friendship, stability and sense of family, and even personal validation the parent provided (Ferraro, 1984; Umberson, 2003). Not all attempts at compensation are effective, however, and the bereaved may end up disappointed in a perceived lack of support from one’s spouse and/or surviving parent (Umberson, 2003; Zettel & Rook, 2004).

Adults’ successes and problems impact their parents’ well-being into old age (Fingerman, Cheng, Birditt, & Zarit, 2012; Greenfield & Marks, 2006; Suitor, Pillemer, & Sechrist, 2006), and the extent to which parents and adult children worry about one another affects their relationship quality (Hay, Fingerman, & Lefkowitz, 2007). Good relationships with parents can improve adults’ marital quality, and widowed parents both receive support from and give support to bereaved adult children (Ha, Carr, Utz, & Nesse, 2006; Reczek et al., 2010). Therefore, it is possible that improved relationship quality with a surviving parent following the loss of a first parent may have a beneficial influence on adult children’s marital quality. The death of a second parent, though, may pose a shift in the eldest surviving generation and represent the passing of “generational time” (Bengtson & Allen, 1993) and could have a more profound impact on adult children’s marriages, as there is no surviving parent to provide support. This article examines the influence of relationship quality with a surviving parent in order to assess linked lives in the context of bereavement.

A tension exists, however, concerning the respective influence of intergenerational and spousal relationships. While parents and children are linked across the life course, marriage is also a greedy institution that lowers closeness and contact in adults’ intergenerational ties (Sarkisian & Gerstel, 2008). Furthermore, spouses in more intimate and secure marriages also report greater marital satisfaction (Feeney, 2002; Greeff & Malherbe, 2001), implying that greedier marriages may be happier marriages. While widowed parents can become more dependent on their adult children after loss and benefit from their support (Ha, 2010; Ha et al., 2006), bereavement also impacts the marital relationship (Umberson, 1995). The death of a first parent may potentially offer an opportunity for bereaved adult children to turn primarily to their marital relationship, rather than to their surviving parent, for support. That is, bereaved adult children may seek out spouses in order to compensate for the loss of a parent; this may result in stronger, closer marriages or in disappointment in a spouse’s inability to meet the bereaved’s expectations for support (Ferraro, 1984; Ferraro, Mutran, & Barresi, 1984; Umberson, 2003).

These frameworks both suggest that marriages can—and even should—improve following parental loss. Linked lives posits that surviving parents will act as a source of support and compensation following the death of a first parent, with improvements (or declines) in surviving parent–child relationships being related with improvements (or declines) in marital quality. After the death of a second parent, the effect of loss on marital quality may also be more profound since there is no surviving parent available for support. The greediness of marriage suggests that spouses will be the primary source of support and compensation after the death of a parent and that the loss of particularly close parents will be related with the greatest increases in marital quality. This study addresses the tension between linked lives and greedy marriage by assessing whether surviving parents and spouses compensate for bereaved adult children’s loss.

Gender in family relationships

The death of a parent may be the loss of either a mother or a father, and these loss experiences might differ because the roles mothers and fathers fulfill in the extended family can differ. The impact of a loss depends in part upon the salience of that relationship to the bereaved (Umberson & Chen, 1994). While men’s roles in the family are changing (e.g., Gerstel & Gallagher, 2001), the literature on gender and intergenerational relationships emphasizes the historical importance of ties to women; that is, mothers have typically been closer to their children than have fathers, including after widowhood and divorce, and they have been the primary attachment figure for both daughters and sons (Chodorow, 1978; Ha et al., 2006; Lin, 2008; Umberson, 1992). In fact, after taking into account relationships with mothers, Umberson (1992) found that relationships with fathers had no influence on adult children’s psychological functioning. More recent research has shown that fathers do contribute independently to children’s psychological well-being, though perhaps not as critically or stably as mothers do (e.g., Videon, 2005). Additionally, daughters are closer to their parents than are sons, especially to mothers; thus the mother–daughter dyad is viewed as especially close and salient (Chodorow, 1978; Fingerman, 2001; Suitor et al., 2006; Umberson, 1992).

The implications of these phenomena are twofold. First, given the primacy of mothers over fathers, I expect the loss of a mother–adult child relationship to be more impactful than the loss of a father–adult child relationship. As Umberson (1992, p. 672) notes, “fathers’ lesser involvement with children throughout the life course,” along with mothers’ “unique and strong attachment” with children, may result in the quality of the relationship with one’s mother being more important for adult children’s lives than the quality of the relationship with one’s father. The loss of that relationship, then, may also be a more important life event than the loss of one’s relationship with a father.

As concerns the gender of the adult child, the implications are less clear. On the one hand, the mother–daughter relationship is considered especially important, in that daughters not only regard mothers as an attachment figure, but are also socialized to identify with and emulate mothers (Chodorow, 1978). Thus, this relationship may be particularly salient, suggesting that daughters who lose mothers may be the most impacted by the bereavement experience (Umberson & Chen, 1994).

On the other hand, parent’s gender may be the most crucial factor in determining parent–adult child relationship salience for both sons and daughters (Umberson & Chen, 1994). Moreover, bereaved adult daughters’ identification with—and emulation of—mothers may also better equip them with the emotional skills and social supports needed to cope with loss. In brief, adult daughters—like their own mothers—foster more and stronger family ties than do adult sons and tend to engage more in kin keeping, that is, the “efforts expended on behalf of keeping family members in touch with one another” (Gerstel & Gallagher, 1993; Kalmijn, 2007; Rosenthal, 1985, p. 965). Thus, bereaved adult sons may have fewer resources for social support after loss than bereaved adult daughters do and therefore fewer alternative sources of support if spousal support is insufficient.

Bereaved men and women have different emotional and coping skills as well as different resources for support. Umberson (2003, p. 22) found that, after a parent’s death, women were far more likely than men to describe their partner as “uncomfortable” with their emotional reactions and to feel “let … down emotionally” by their spouse. Wives may be better than husbands at providing support to a bereaved spouse as well as maintaining social and family ties for the marital dyad (Umberson, 2003). However, if a husband cannot adequately compensate for the loss of a salient mother–daughter relationship, a bereaved daughter may seek to strengthen relationships with siblings or other family members instead (Ferraro, 1984; Umberson, 1995). Sons’ lack of alternative family and social supports may make them especially dependent upon a spouse following the death of a parent, particularly a mother. This implies that sons who lose mothers may be especially dependent on wives but may also be more likely to have their emotional needs met than bereaved daughters.

The literature on the death of a parent has exhibited varied results concerning gender; for example, the death of a mother may result in worse psychological well-being and a greater mortality risk than the death of a father (Lawrence, Jeglic, Matthews, & Pepper, 2006; Rostila & Saarela, 2011); yet the loss of a same-gender parent may have a stronger negative effect on adults’ well-being than the loss of an opposite-gender parent (Marks et al., 2007). Additionally, there is some foundation for the hypothesis that sons who lose mothers are in a unique position (Rostila & Saarela, 2011). They maintain fewer and weaker family ties than daughters and have closer ties with mothers than with fathers. This study incorporates subjective parent–child relationship quality in order to better analyze the role of relationship salience and gender on bereaved adult children.

Summary of research aims

This study analyzes the influence of intergenerational relationships on adults’ marriages following the death of a parent. The loss of a parent is an extremely common yet understudied experience and one that can impact adults’ well-being and close relationships. This study investigates this important life transition and assesses how family relationships affect the bereavement experience, for better or worse. This not only will contribute to the literature on linked lives, greedy marriage, and the role of gender in intergenerational relationships but may also inform policy and practice by addressing bereavement as a familial rather than a purely individual, psychological experience. Specifically, my analyses examine the following research questions:

Research Question 1: Does relationship quality with a deceased parent influence marital quality after loss?

Research Question 2: Does relationship quality with a surviving parent influence marital quality after the death of a first parent?

Research Question 3: Is the influence of relationship quality with a deceased parent on marital quality most pronounced for the most vulnerable bereaved adults, that is, sons who lose mothers?

Methods

Data and sample

I used data from the Longitudinal Study of Generations (LSOG). Data were first collected in 1971 from 300 three-generation Southern California families, selected randomly from an 840,000-member California Health Maintenance Organization (Silverstein, Parrott, & Bengtson, 1995). Study investigators selected the parent–child relationships that participants reported on. Generation 1 parents (both mother and father) were assigned focal “study children” (Generation 2) and “study grandchildren” (Generation 3) to report upon, with Generation 3 study grandchildren being the study children of the Generation 2 study children. Moreover, these assignments were dyadic; the focal parents (Generation 1), children (Generation 2), and grandchildren (Generation 3) reported on one another. In each case, younger generations reported on each of their parents/grandparents, while older generations reported on their direct descendant children/grandchildren. Participants’ spouses were included in the survey for applicable relationships (e.g., a Generation 2 son’s wife would be included in the survey as the Generation 3 grandchild’s mother but not as the Generation 1 parents’ daughter).

In Wave 1, a total of 2,044 individuals from 328 families participated. Follow-up surveys were conducted in 1985, 1988, 1991, 1994, 1997, 2000, and 2005. Nonmortality attrition patterns indicate that older participants, males, minorities, and the less educated were somewhat more likely to drop out of the study. The response rate for Wave 2 (1985) was 73%, and longitudinal response rates averaged 80% from 1985 onward (Gans & Silverstein, 2006). The LSOG sample represents White, economically stable middle- and working-class families (Silverstein et al., 1995).

In this study, Waves 1 and 2 were excluded due to the absence of survey questions crucial to this project. The analytic subsample consisted of adult children (Generations 2 and 3) who reported (a) experiencing the death of a parent between waves and (b) being continuously married in the survey waves immediately prior to and following the death of a parent. Generation 2 participants reported on their Generation 1 parents, while Generation 3 participants reported on their Generation 2 parents. Analyses conducted with continuously cohabiting adults (n = 12) included yielded the same conclusions.¹

Adult children who separated or divorced (n = 17), were widowed (n = 10), or who married (n = 13) between the same waves as a parent’s death were excluded. In each case, the order of the marital transition and parent’s death could not be confidently determined. I excluded 12 cases whose loss experience could not be determined as first or second as well as 6 cases due to missing data for marital quality.

The analyses were performed using a before/after design, using the death of a parent as the focal event. Regardless of the parent’s year of death, measures were included from the wave immediately prior to and immediately following the adult child’s loss. This allowed for all parent deaths between Waves 3 and 8 to be included in analyses. Models were estimated using dichotomous indicators for the date of the wave before loss, with no meaningful differences found, so these were excluded from final analyses. The final analytic subsample consisted of 304 parent deaths, nested within 269 adult children, nested within 156 families (i.e., some adult children experienced the death of both parents during the course of the study, and in some families, multiple generations/individuals experienced parental loss during the course of the study).

Dependent variable

Change in marital quality

Change in marital quality following the death of a parent was measured using a 10-item marital satisfaction scale (Gilford & Bengtson, 1979) generated as a mean score scale (α = .85). It was designed as a two-dimensional measure, including elements of both positive interaction and negative sentiment (Gilford & Bengtson, 1979), and has been consistently used in marital research (e.g., Feng, Giarrusso, Bengtson, & Frye, 1999; Orth, Robins, & Widaman, 2012). Responses ranged from 1 (hardly ever) to 5 (almost always) on items such as “When you are with your spouse or partner: You calmly discuss something together,” “One of you is sarcastic,” “You work together on something,” and “One of you becomes critical and belittling.” Responses were coded so that higher values indicate greater marital quality. Participants who answered fewer than half of the items were coded as missing.

Change in marital quality was measured as a difference score of before and after the loss of a parent, which allows for intuitively correct inferences (Allison, 1990; Johnson, 2005; Roest, Dubas, Gerris, & Engels, 2009). Difference score models control for any unmeasured influences on the dependent variable that are stable over time and often produce less bias than lagged dependent variable (LDV) models (Allison, 1990; Johnson, 2005; Roest et al., 2009). Results were substantively similar using LDV models.

Independent variables

Child–deceased parent relationship quality

A 6-item Affectual Solidarity Scale (Bengtson & Roberts, 1991) was recorded at each wave of the LSOG and was used to determine relationship quality between the adult child and the deceased parent at the wave prior to the parent’s death (α = .86). This scale is commonly used to measure affinity between family members across generations (e.g., Moorman & Stokes, 2014; Silverstein, Bengtson, & Lawton, 1997). Questions were asked about one’s mother and father independently, with responses ranging from 1 (not at all) to 6 (extremely). Sample items include “Taking everything into consideration, how close do you feel is the relationship between you and your parent at this point in your life?” and “How well do you feel your parent understands you?” The solidarity variable was constructed using a mean score scale and was transformed by squaring the distribution to account for significant negative skewness. Participants who answered fewer than half of the items were coded as missing.

Change in child–surviving parent relationship quality

The same Affectual Solidarity Scale was used to determine relationship quality between the adult child and the surviving parent both prior to and following the other parent’s death (α = .86). This variable was also a mean score scale that was squared to account for significant negative skewness. Change in child–surviving parent relationship quality was measured as a difference score of before and after the loss of the other parent. Since this measure applies only to first loss experiences, dummy variable adjustment was used for adult children who experience a second loss (Allison, 2001; Cohen & Cohen, 1985). For these cases, I imputed the mean value of the change score, and this adjustment was signaled by the inclusion of a dichotomous variable indicating first versus second loss. Dummy variable adjustment allows for participants with nonapplicable missingness on an item—here, adults who lost a second parent were necessarily missing data on change in relationship quality with a surviving parent—to be included in the analyses for the estimation of all applicable variables’ coefficients. Since the signal variable (second loss) is a perfect predictor of the manually imputed nonapplicable variable (change in relationship quality with a surviving parent), nonapplicable participants have no impact on the coefficient for the nonapplicable variable. Thus, information from adult children who lost a second parent is included for the estimation of all effects except change in relationship quality with a surviving parent. Listwise deletion would remove adult children who lost a second parent from the analyses entirely due to missing data, whereas multiple imputation would improperly treat nonapplicable missingness as random.

Second loss

A dichotomous indicator for second parent death (1 = second loss and 0 = first loss) was used to analyze any differences between first and second bereavement experiences. It also served as a signal variable for dummy variable adjustment. Since the mean value of change in surviving parent relationship quality was manually imputed for those who experienced a second parent death, the coefficient for second loss can be interpreted for those who lost a second parent compared with a reference group of those who experienced a first parent death and had average change in relationship quality with a surviving parent.

Parent and child gender pairings

Gender was modeled using four mutually exclusive and exhaustive parent–child gender pairings. These are dichotomous cross-level interaction terms. Sons and daughters who lost both parents between the same two waves (n = 7) were coded as having experienced the loss of a mother. Coding them as having lost a father did not significantly alter the results.

Interaction terms

Child–Deceased Parent Relationship Quality × Parent–Child Gender Pairings

Three-way multiplicative interaction terms between child–deceased parent relationship quality and each of the parent–child gender pairings were estimated in order to determine whether the influence of child–deceased parent relationship quality on adult children’s marital quality differed by gender of both parent and child.

Analytic strategy and missing data

I used multilevel random effects models to address my research questions. The data were clustered in a three-level model of loss experiences nested within individuals nested within families. In this case, ordinary least squares (OLS) regression would incorrectly calculate standard errors, due to nonindependence in the data resulting from multiple respondents from the same families and multiple responses from the same individuals at different points in time. Such underestimation of standard errors increases the risk of Type 1 error or falsely significant findings. In this way, multilevel modeling is a more conservative and appropriate strategy.

Random effects models are appropriate in this case because I am interested both in within- and between-person variables and effects. That is, I am interested in both change across time for individuals and differences between individuals. Intercepts at each level were modeled as random, with slopes for Level-1 and Level-2 predictors modeled as fixed. Random slopes could not be estimated since group sizes of at least three are needed for this, and participants could only experience up to two parent deaths. Therefore, the effect of any predictor on change in marital quality was the same for all participants, but random intercepts addressed variation across persons and families.

Model 1 addresses Research Questions 1 and 2, concerning the associations between relationship quality with a deceased parent and marital quality, as well as change in relationship quality with a surviving parent and marital quality. It includes the key independent variables of interest—child–deceased parent relationship quality and change in surviving parent relationship quality—as well as all control measures. Model 2 addresses Research Question 3, concerning a possible unique effect of parent–child relationship quality for sons who lost mothers. The difference between Model 1 and Model 2 is that Model 2 includes three-way multiplicative interaction terms between child–deceased parent relationship quality and the parent–child gender pairings. Sons who lost mothers served as the reference group.

Controls for age, time (in years) between loss and survey response, income, and number of persons in the multigenerational family were explored but excluded from the final analyses, as they were not significant and had no substantial impact on results. Change in depression was also tested as a control measure. It had a significant influence on the outcome variable but did not significantly alter the effects of any other predictors. It has been excluded in the interests of parsimony. Models were also estimated that examined cohort effects (i.e., any differences between Generations 2 and 3) as well as period effects (i.e., any differences between Waves 3 through 8) but were excluded as there were no significant impacts on results. Interaction terms between change in surviving parent relationship quality and parent–child gender pairings were tested, with no significant results found. Gender and gender interactions were also modeled using separate indicators for gender of the parent and gender of the child, with no significant results.

Missing data were addressed using multiple imputation by chained equations (Royston, 2005). A total of 10 imputed data sets were produced. Of the 322 potential cases in the analytic subsample, 86% had complete data for all measures. The greatest level of missing data was for the child–deceased parent relationship quality measure, which had 7.4% of cases missing data. The outcome was included in the imputation command but analyses did not use imputed marital quality (von Hippel, 2007). Imputation enhanced final sample size and protected against potentially biased coefficients resulting from listwise deletion but did not substantially affect any findings.

Results

Descriptive statistics for all variables are shown in Table 1. Overall, there was no average trend concerning change in marital quality. Mean change in marital quality was not significantly different from zero. Yet more than 25% of respondents displayed positive or negative changes of .5 or more on the 5-point scale, and approximately 64% reported change of less than .5, indicating variability in the measure rather than uniform stability.

Table 1.

Descriptive statistics, adult children who experienced the death of a parent, Longitudinal Study of Generations, 1988–2005.

	Mean or n	SD or %	Observed minimum	Observed maximum
Dependent variable
Change in marital quality^a	−.02	.43	−1.51	1.70
Key independent variables
Child–deceased parent relationship quality^b,c	3.96	1.08	1.00	6.00
Change in relationship quality with a surviving parent^b,d	−.10	.71	−2.67	2.00
Second loss experience	144	47.4%	—	—
Daughters who lost mothers	85	28.0%	—	—
Daughters who lost fathers	79	26.0%	—	—
Sons who lost mothers	82	27.0%	—	—
Sons who lost fathers	58	19.1%	—	—

Note. N = 304.

^aRaw scale ranges from 1 = lowest quality to 5 = highest quality; change scores can range from −4 = greatest decline in quality to 4 = greatest increase in quality.

^bDescriptive statistics displayed for raw variable; transformed variable used in analyses.

^cRaw scale ranges from 1 = worst relationship quality to 6 = best relationship quality.

^dRaw scale ranges from 1 = worst relationship quality to 6 = best relationship quality; change scores can range from −5 = greatest decline in relationship quality to 5 = greatest improvement in relationship quality.

Table 2 displays the results of the multilevel random effects regression models. As concerns Research Question 1, Model 1 showed no association overall between child–deceased parent relationship quality and change in the adult child’s marital quality after loss. Regarding Research Question 2, change in relationship quality with a surviving parent was significantly positively related with marital quality (B = .02, p < .01). Improvement in relationship quality with a surviving parent was associated with improved marital quality overall. There were no differences in marital quality change according to gender of the parent and child.

Table 2.

Change in marital quality following the death of a parent.

	Model 1	Model 2
Fixed effects	B (SE)	B (SE)
Child–deceased parent relationship quality^a,b	0.00	0.02*
	(0.00)	(0.01)
Change in relationship quality with a surviving parent^a	0.02**	0.02*
	(0.01)	(0.01)
Second loss experience^c	0.02	0.00
	(0.06)	(0.06)
Daughter who lost father^d	−0.02	−0.04
	(0.07)	(0.07)
Daughter who lost mother^d	−0.10	−0.11
	(0.07)	(0.06)
Son who lost father^d	−0.00	−0.01
	(0.08)	(0.08)
Relationship Quality × Daughter/Father pairing	—	−0.02*
		(0.01)
Relationship Quality × Daughter/Mother pairing	—	−0.02*
		(0.01)
Relationship Quality × Son/Father pairing	—	−0.02* (0.01)
Constant	0.02	0.03
	(0.06)	(0.06)
Random components	σ (SE)	σ (SE)
Level 1: Loss experience	0.42***	0.41***
	(0.02)	(0.02)
Level 2: Individual	0.00	0.00
	(0.00)	(0.00)
Level 3: Family	0.00	0.00
	(0.00)	(0.00)
R ²	4.00%	6.11%
F; df	2.02; 6	2.05*; 9

Note. N = 304. Analyses using each other pairing as reference group reveal that child–deceased parent relationship quality is significant only for the son–mother pairing. df = degrees of freedom.

^aTransformed variable.

^bMean-centered variable.

^cReference group is first loss experience.

^dReference group is son who lost mother pairing.

*p < .05; **p < .01; ***p < .001.

Model 2 revealed further significant effects. For sons who lost mothers, better pre-loss relationship quality was associated with improvement in marital quality (B = .02, p < .05). For each of the other three parent–child gender pairings, pre-loss relationship quality with a deceased parent was not related with any change in marital quality. The three interaction terms between relationship quality with a deceased parent and parent–child gender pairings were all negative and significant (B = −.02, p < .05).

Supplemental analyses (not shown) that examined each parent–child gender pairing as the reference group confirmed that child–deceased parent relationship quality did not have a significant influence on change in marital quality for any group other than sons who lost mothers. In short, the closer sons were to deceased mothers, the more their marriages improved after loss; the poorer their relationship quality with deceased mothers, the worse their marriages fared after loss. This association holds only for sons who lost mothers, as Figure 1 illustrates.

It is worth noting that despite the significant effects revealed in Models 1 and 2, variance explained is relatively low. The R²values indicate that Model 1 explained approximately 4.00% of the variance in change in marital quality, while Model 2 explained 6.11%. This means that the significant interaction effect for sons who lost mothers only explained an additional 2.11% of variance in the outcome measure overall. A supplemental analysis revealed that change in child–surviving parent relationship quality alone explained 2.94% of variance in change in marital quality. This is partly due to the use of a change score as an outcome variable, as all stable between-person differences are parceled out (Allison, 1990). Moreover, in this case pre-loss marital quality explained over 50% of the variance in post-loss marital quality. However, parent–child relationships appear to explain a relatively small amount of the variation in marital quality.

Discussion

The present analyses demonstrate new information concerning adult children’s marriages following the death of a parent. This information extends theory on linked lives and the greediness of marriage and underscores the importance of considering parent–child relationship quality when determining the role and influence of intergenerational relationships. Improvements in relationship quality with a surviving parent were related with improvements in marital quality following the death of a first parent; moreover, the loss of a second parent did not have a larger effect on adult children’s marriages than the loss of a first parent. Findings also highlight the uniqueness of the mother–son gender pairing. Better relationship quality with a deceased mother before her death was related with improvements in adult sons’ marriages following loss. This effect was not generalizable to other parent–child gender pairings, revealing the complexity of gender in family relationships. Lastly, the overall stability of marital quality implies resilience in the marital relationship, especially as time passes following bereavement. The remaining sections situate these findings in the literature, note the limitations of this study and point to future avenues of research, and discuss the implications of the present findings for the literature as well as for policy and practice.

Lives remain linked

This study provided clear support for the linked lives perspective, as change in relationship quality with a surviving parent was significantly related with change in marital quality following the death of a first parent. That is, when bereaved adult children experience improvements in their relationships with widowed parents, they also experience better marriages; similarly, declines in relationship quality with widowed parents were related with declines in marital quality as well. This effect did not differ by gender of the parent or the child but rather held true for adult children generally who had a surviving parent. This finding supports the linked lives framework, as the parent–child relationship remains influential for adult children’s marriages following the death of a parent.

It is noteworthy, also, that the analytical results showed no difference in marital quality between bereaved adult children who had lost a first parent and those who lost a second parent. This suggests that the quality of the surviving parent–child relationship matters more for adult children’s marital quality than the mere presence of a surviving parent does. Surviving parents may be resources or stressors for adult children, whereas the presence—or lack—of a surviving parent is neither beneficial nor detrimental in itself for adult children’s marriages.

How greedy is marriage?

For sons who lost mothers, closeness with the deceased parent was related with improvement in marital quality following loss. Marriage thus appears to act as a greedy institution (Sarkisian & Gerstel, 2008) for sons who lost close mothers, “filling the void” left by the loss of a salient intergenerational relationship (Umberson, 1995). However, the greediness of marriage was not apparent for any other gender pairs. Moreover, the loss of a second parent had no significantly greater an effect on adult children’s marriages than the loss of a first parent. This suggests that marriage does not get progressively greedier over time but is instead a stable factor in adults’ relationships.

The unique results for sons who lost mothers may be due to the salience of the loss combined with the effective emotional and social support provided by wives (Chodorow, 1978; Umberson, 2003). However, the finding may also be due to sons’ closeness with their mothers negatively impacting marital quality prior to mothers’ deaths, implying that the loss of that close relationship does in fact offer an opportunity for the marital relationship to further assert its prominence in adult men’s lives. Unusual mother–son closeness at Time 1 may negatively affect marital quality, so the cessation of that marital stressor may improve marital quality at Time 2. This finding would be in keeping with greedy marriage (Sarkisian & Gerstel, 2008).

The finding that better relationship quality with a deceased mother was related with improved marital quality for bereaved sons also means that poorer relationship quality with a deceased mother was related with declines in marital satisfaction for bereaved sons. Following the loss of more distant or difficult mothers, sons may transfer those negative emotions or poor relationship skills to their marriages. This may also be due in part to low quality in the mother-in-law/daughter-in-law relationships, since spouses’ perceptions of deceased in-laws play an important role in their ability and willingness to provide empathic support (Umberson, 2003).

Relationship quality, gender, and marital quality

This study serves to extend the literature on gender in intergenerational relationships as well. Sons who lost close and distant mothers had very different outcomes from other gender pairs. These differential marital consequences may be explained, in part, by the compensation model of role loss (Ferraro, 1984; Ferraro et al., 1984). A large literature highlights the compensatory social behaviors of widowed older adults (Donnelly & Hinterlong, 2010; Ferraro et al., 1984; Ha, 2008). Following loss experiences, the bereaved may seek to “fill the void” left by the loss of a particular relationship (Umberson, 1995). However, not all attempts at compensation are successful (Zettel & Rook, 2004). As Umberson (1995, p. 2003) notes, declines in marital quality among recently bereaved adult children may be due primarily to their failed attempts to compensate for their loss and their disappointment in their spouses’ failure to provide adequate support.

By incorporating parent–child relationship quality in the analyses, this paper may have identified a group for whom compensatory behaviors successfully reap benefits: sons who lose close mothers. The case of sons who lose mothers with whom they had poor relationship quality is less fortunate as they experience a significant loss, depend heavily on their spouse for support, but may be especially likely to encounter disappointment. Spouses are unlikely to be empathic and supportive following the death of a difficult or troublesome parent (Umberson, 1995, 2003).

This framework may also partially explain the lack of significant findings for daughters who lost mothers. It is not that this loss is less salient for daughters than for sons but rather that (a) daughters’ compensatory behaviors may be met with unsupportive, even disappointing husbands and (b) daughters have more and stronger close relationships outside marriage that they can turn to, such as siblings or other family members (Umberson, 1995, 2003). The lack of significant results for daughters who lost mothers may also be related to caregiving as daughters are more likely than sons to engage in caregiving for aging parents, particularly mothers (Gerstel & Gallagher, 2001; Lin, 2008; Silverstein, Gans, & Yang, 2006; Umberson, 2003). In this event, caregiving adult children can engage in “anticipatory coping,” meaning that their grieving process begins while the dying parent is still alive (Stroebe & Stroebe, 1993; Umberson, 2003). This would depress Time 1 marital quality scores and reduce apparent “change” in marital quality following the loss experience itself. Unfortunately, the present data do not include information on caregiving needs or provision over time, leaving this alternative hypothesis for future inquiry.

Overall, though numerous possible gender interactions were hypothesized, only sons who lost mothers revealed unique effects. The lack of other significant interactions may be due to a combination of factors. First, fathers tend to play a less significant role in adult children’s lives, and their loss may be less influential than the loss of a mother (Chodorow, 1978; Ha et al., 2006; Lin, 2008; Umberson, 1992). Further, adult daughters tend to have greater recourse to social support following loss than adult sons do, potentially insulating them and their marriages from the impacts of loss (Umberson, 1995, 2003). Lastly, the present sample may not be sufficiently large to detect smaller effects on marital quality (Umberson, 1995).

Stability of marital quality

While all three research questions garnered some significant results, the relatively small proportion of variance explained raises questions about the practical importance of intergenerational relationships for adults’ marital quality. This finding may imply that adults’ marriages are secure over time and resilient in the face of bereavement, particularly over a period of years rather than weeks or months. For instance, marital quality prior to loss accounts for more than 50% of the variation in marital quality after loss. Thus, while the death of a parent is a major life transition, its impact on adults’ marital quality may be minimal, especially as time goes on following the bereavement (e.g., Bonanno et al., 2002). Moreover, the small proportion of variance explained by relationship quality factors implies that spousal reports of marital quality may depend more heavily on intra-relational factors than on external or intergenerational influences. This fits the greedy marriage thesis that marriage weakens adults’ other ties, including intergenerational relationships, and may weaken their influence as well (Gerstel & Sarkisian, 2006; Sarkisian & Gerstel, 2008).

The loss of a parent is a major transition that necessitates a renegotiation of the spousal relationship in order to find a new equilibrium (Barner & Rosenblatt, 2008; Umberson, 2003). For some couples, no new equilibrium is to be found, with the marriage suffering or ending in divorce (Umberson, 1995, 2003). For others, however, an initial period of disruption in marital dynamics may lead effectively to a restored or even an improved marriage (Barner & Rosenblatt, 2008; Rosenblatt & Barner, 2006). Initial declines in marital quality after a loss may indicate spouses engaging in necessary relationship work, rather than evidence of the stress of bereavement undermining the spousal relationship itself (Barner & Rosenblatt, 2008; Rosenblatt & Barner, 2006; Umberson, 1995). While the significant findings in the present analyses offer support for linked lives and greedy marriage, the general stability of marital quality in the present sample underlines the steadiness of the marital relationship even in the face of a dramatic life transition.

Limitations

The current study has several limitations. First, the sample was not nationally representative and overrepresented White, middle-class persons. Therefore, the potential influence(s) of race/ethnicity and socioeconomic status could not be examined. Research has indicated that family relationships can vary across race and class, as can their influence on marital happiness (Fingerman, Vanderdrift, Dotterer, Birditt, & Zarit, 2011; Sarkisian, Gerena, & Gerstel, 2007). For instance, greater extended family integration and higher rates of financial, emotional, and other support may shape how adult children experience family relationships and bereavement. Future research should test the findings of the current study in other populations.

Second, divorce could not be examined as an additional hazard because the present data did not indicate whether the divorce preceded or followed the loss experience. The relatively small number of divorced participants in this sample (n = 17) implies that divorce was not a major hazard here, but parental death can be a precursor to divorce (Umberson, 1995, 2003) and should be examined more closely in future research.

Additionally, the most recent wave of data came from 2005 and may not represent contemporary conditions. However, the lack of empirical differences by wave and the scarcity of comparably rich sources of data justify use of the LSOG in the present case.

Lastly, the intergenerational family was limited to two generations and does not include sibling or in-law relationships. Siblings and in-laws are central influences in family life and could serve as additional sources of support or strain following loss (McHale, Updegraff, & Whiteman, 2012; Willson, Shuey, & Elder, 2003). Additionally, parental status was not included because the vast majority of the sample (approximately 95%) reported having children; this is due to the structure of the data set, as it specifically concerns multigenerational families. Future research should incorporate relationships with additional members of the extended, intergenerational family as well as examine parent loss in the context of families that are not multigenerational.

Conclusion

The death of a parent is a major life transition, which impacts not only individuals’ psychological well-being but also their families and relationships (Barner & Rosenblatt, 2008; Rosenblatt & Barner, 2006; Stroebe, Schut, & Boerner, 2013; Umberson, 1995, 2003). This study illustrates some of the intergenerational family influences on adults’ marital quality following the death of a parent, building upon prior research (Marks et al., 2007; Umberson, 1995, 2003), and extending the theoretical frameworks of linked lives and greedy marriage while also contributing to the literature on gender and relationship salience in multigenerational families (Chodorow, 1978; Elder, 1999; Sarkisian & Gerstel, 2008; Umberson & Chen, 1994).

In all, the results of this study illustrate that bereavement is not solely an individual psychological experience but a familial experience as well. Moreover, incorporating affectual relationship measures reveals that the quality of one’s relationships is of primary importance following bereavement. These insights can inform both policy and practice, as extended family relationships, such as those with surviving parents, can be sources of positive support or negative strain. Further, the death of a second parent was no worse for adult children’s marriages than the death of a first parent; thus, one cannot assume that larger, multigenerational families are more protective than smaller families or those without a surviving matriarch or patriarch. The quality of one’s relationships appears crucial, and programs that involve entire families—and the relationships therein—may be best equipped to address bereavement experiences. The results of this study reinforce the insight that family relationships remain important across the life course and indeed influence one another both during life and after death.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Data Training and Users Workshop for the Longitudinal Study of Generations, Summer 2012.

Note

References

Allison

P. D.

(1990). Change scores as dependent variables in regression analysis. Sociological Methodology, 20, 93–114. doi:10.2307/271083

Allison

P. D.

(2001). Missing data. Thousand Oaks, CA: Sage Publications.

Barner

Rosenblatt

(2008). Giving at a loss: Couple exchange after the death of a parent. Mortality, 13, 318–334. doi:10.1080/13576270802383899

Bengtson

V. L.

(2001). Beyond the nuclear family: The increasing importance of multigenerational bonds. Journal of Marriage and Family, 63, 1–16. doi:10.1111/j.1741-3737.2001.00001.x

Bengtson

V. L.

Allen

K. R.

(1993). The life course perspective applied to families over time. In Boss

P. G.

Doherty

W. J.

LaRossa

Schumm

W. R.

Steinmetz

S. K.

(Eds.), Sourcebook of family theories and methods: A contextual approach (pp. 469–504). New York, NY: Plenum.

Bengtson

V. L.

Roberts

R. E.

(1991). Intergenerational solidarity in aging families: An example of formal theory construction. Journal of Marriage and Family, 53, 856–870. doi:10.2307/352993

Bonanno

G. A.

Wortman

C. B.

Lehman

D. R.

Tweed

R. G.

Haring

Sonnega

… Nesse

R. M.

(2002). Resilience to loss and chronic grief: A prospective study from preloss to 18-months postloss. Journal of Personality and Social Psychology, 83, 1150–1164. doi:10.1037//0022-3514.83.5.1150

Carr

Springer

K. W.

(2010). Advances in families and health research in the 21st century. Journal of Marriage and Family, 72, 743–761. doi:10.1111/j.1741-3737.2010.00728.x

Chodorow

(1978). The reproduction of mothering: Psychoanalysis and the sociology of gender. Berkeley, CA: University of California Press.

10.

Cohen

(1985). Applied multiple regression and correlation analysis for the behavioral sciences, 2nd ed. Hillsdale, NJ: Erlbaum.

11.

Donnelly

E. A.

Hinterlong

J. E.

(2010). Changes in social participation and volunteer activity among recently widowed older adults. The Gerontologist, 50, 158–169. doi:10.1093/geront/gnp103

12.

Elder

G. H.

(1999). Children of the great depression: Social change in life experience. Boulder, CO: Westview Press.

13.

Elder

G. H.

Johnson

M. K.

(2003). The life course and aging: Challenges, lessons, and new directions. In Settersen

Jr . (Ed.), Invitation to the life course: Toward new understandings of later life (pp. 49–81). Amityville, NY: Baywood.

14.

Feeney

J. A.

(2002). Attachment, marital interaction, and relationship satisfaction: A diary study. Personal Relationships, 9, 39–55. doi:10.1111/1475-6811.00003

15.

Feng

Giarrusso

Bengtson

V. L.

Frye

(1999). Intergenerational transmission of marital quality and marital instability. Journal of Marriage and Family, 61, 451–463. doi:10.2307/353761

16.

Ferraro

K. F.

(1984). Widowhood and social participation in later life: Isolation or compensation? Research on Aging, 6, 451–468. doi:10.1177/0164027584006004001

17.

Ferraro

K. F.

Mutran

Barresi

C. M.

(1984). Widowhood, health, and friendship support in later life. Journal of Health and Social Behavior, 25, 246–259. doi:10.2307/2136423

18.

Fingerman

K. L.

(2001). Aging mothers and their adult daughters: A study in mixed emotions. New York, NY: Springer.

19.

Fingerman

K. L.

Cheng

Y.-P.

Birditt

Zarit

(2012). Only as happy as the least happy child: Multiple grown children’s problems and successes and middle-aged parents’ well-being. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 67, 184–193. doi:10.1093/geronb/gbr086

20.

Fingerman

K. L.

Vanderdrift

L. E.

Dotterer

Birditt

K. S.

Zarit

(2011). Support to aging parents and grown children in Black and White families. The Gerontologist, 51, 441–452. doi: 10.1093/geront/gnq114

21.

Fuller-Thomson

(2000). Loss of the kin-keeper? Sibling conflict following parental death. OMEGA Journal of Death and Dying, 40, 547–559. doi: 10.2190/J357-QF31-F0FC-PV4U

22.

Gans

Silverstein

(2006). Norms of filial responsibility for aging parents across time and generations. Journal of Marriage and Family, 68, 961–976. doi:10.1111/j.1741-3737.2006.00307.x

23.

Gerstel

Gallagher

S. K.

(1993). Kinkeeping and distress: Gender, recipients of care, and work-family conflict. Journal of Marriage and the Family, 55, 598–608. doi:10.2307/353341

24.

Gerstel

Gallagher

S. K.

(2001). Men’s caregiving: Gender and the contingent character of care. Gender and Society,15, 197–217. doi:10.1177/089124301015002003

25.

Gerstel

Sarkisian

(2006). Marriage: The good, the bad, and the greedy. Contexts, 5(16), 16–21. doi: 10.1525/ctx.2006.5.4.16

26.

Gilford

Bengtson

(1979). Measuring marital satisfaction in three generations: Positive and negative dimensions. Journal of Marriage and Family, 41, 387–398. doi:10.2307/351705

27.

Greeff

A. P.

Malherbe

H. L.

(2001). Intimacy and marital satisfaction in spouses. Journal of Sex & Marital Therapy, 27, 247–257. doi:10.1080/009262301750257100

28.

Greenfield

E. A.

Marks

N. F.

(2006). Linked lives: Adult children’s problems and their parents’ psychological and relational well-being. Journal of Marriage and Family, 68, 442–454. doi:10.1111/j.1741-3737.2006.00263.x

29.

J.-H.

(2008). Changes in support from confidants, children, and friends following widowhood. Journal of Marriage and Family, 70, 306–318. doi:10.1111/j.1741-3737.2008.00483.x

30.

J.-H.

(2010). The effects of positive and negative support from children on widowed older adults’ psychological adjustment: A longitudinal analysis. The Gerontologist, 50, 471–481. doi: 10.1093/geront/gnp163

31.

J.-H.

Carr

Utz

R. L.

Nesse

(2006). Older adults’ perceptions of intergenerational support after widowhood: How do men and women differ? Journal of Family Issues, 27, 3–30. doi: 10.1177/0192513X05277810

32.

Hay

E. L.

Fingerman

K. L.

Lefkowitz

E. S.

(2007). The experience of worry in parent-adult child relationships. Personal Relationships, 14, 605–622. doi: 10.1111/j.1475-6811.2007.00174.x

33.

Johnson

(2005). Two-wave panel analysis: Comparing statistical methods for studying the effects of transitions. Journal of Marriage and Family, 67, 1061–1075. doi:10.1111/j.1741-3737.2005.00194.x

34.

Kalmijn

(2007). Gender differences in the effects of divorce, widowhood and remarriage on intergenerational support: Does marriage protect fathers? Social Forces, 85, 1079–1104. doi:10.1353/sof.2007.0043

35.

Lawrence

Jeglic

E. L.

Matthews

L. T.

Pepper

C. M.

(2006). Gender differences in grief reactions following the death of a parent. OMEGA Journal of Death and Dying, 52, 323–337. doi:10.2190/55WN-1VUF-TQ3W-GD53

36.

Lin

I.-F.

(2008). Consequences of parental divorce for adult children’s support of their frail parents. Journal of Marriage and Family, 70, 113–128. doi:10.1111/j.1741-3737.2007.00465.x

37.

Marks

N. F.

Jun

Song

(2007). Death of parents and adult psychological and physical well-being: A prospective U.S. national study. Journal of Family Issues, 28, 1611–1638. doi:10.1177/0192513X07302728

38.

McHale

S. M.

Updegraff

K. A.

Whiteman

S. D.

(2012). Sibling relationships and influences in childhood and adolescence. Journal of Marriage and Family, 74, 913–930. doi:10.1111/j.1741-3737.2012.01011.x

39.

Moorman

S. M.

Stokes

J. E.

(2014). Solidarity in the grandparent–adult grandchild relationship and trajectories of depressive symptoms. The Gerontologist. doi:10.1093/geront/gnu056

40.

Orth

Robins

R. W.

Widaman

K. F.

(2012). Life-span development of self-esteem and its effects on important life outcomes. Journal of Personality and Social Psychology, 102, 1271–1288. doi:10.1037/a0025558

41.

Reczek

Liu

Umberson

(2010). Just the two of us? How parents influence adult children’s marital quality. Journal of Marriage and Family, 72, 1205–1219. doi:10.1111/j.1741-3737.2010.00759.x

42.

Roest

A. M.

Dubas

J. S.

Gerris

J. R. M.

Engels

R. C. M. E.

(2009). Value similarities among fathers, mothers, and adolescents and the role of a cultural stereotype: Different measurement strategies reconsidered. Journal of Research on Adolescence, 19, 812–833. doi:10.1111/j.1532-7795.2009.00621.x

43.

Rosenblatt

P. C.

Barner

J. R.

(2006). The dance of closeness-distance in couple relationships after the death of a parent. OMEGA Journal of Death and Dying, 53, 277–293. doi:10.2190/X1W4-M210-6522-1034

44.

Rosenthal

C. J.

(1985). Kinkeeping in the familial division of labor. Journal of Marriage and Family, 47, 965–974. doi:10.2307/352340

45.

Rostila

Saarela

(2011). Time does not heal all wounds: Mortality following the death of a parent. Journal of Marriage and Family, 73, 236–249. doi:10.1111/j.1741-3737.2010.00801.x

46.

Royston

(2005). Multiple imputation of missing values: Update of ice. Stata Journal, 5, 527–536.

47.

Sarkisian

Gerena

Gerstel

(2007). Extended family integration among Euro and Mexican Americans: Ethnicity, gender, and class. Journal of Marriage and Family, 69, 40–54. doi:10.1111/j.1741-3737.2006.00342.x

48.

Sarkisian

Gerstel

(2008). Till marriage do us part: Adult children’s relationships with their parents. Journal of Marriage and Family, 70, 360–376. doi:10.1111/j.1741-3737.2008.00487.x

49.

Silverstein

Bengtson

V. L.

Lawton

(1997). Intergenerational solidarity and the structure of adult child-parent relationships in American families. American Journal of Sociology, 103, 429–460. doi:10.1086/231213

50.

Silverstein

Gans

Yang

F. M.

(2006). Intergenerational support to aging parents: The role of norms and needs. Journal of Family Issues, 27, 1068–1084. doi:10.1177/0192513X06288120

51.

Silverstein

Parrott

T. M.

Bengtson

V. L.

(1995). Factors that predispose middle-aged sons and daughters to provide social support to older parents. Journal of Marriage and Family, 57, 465–475.

52.

Stroebe

Schut

Boerner

(2013). Goals and overview of the special issue. Bereavement: A family affair. Family Science, 4, 1–3. doi:10.1080/19424620.2013.819223

53.

Stroebe

M. S.

(1993). Determinants of adjustment to bereavement in younger widows and widowers. In Stroebe

M. S.

Stroebe

Hanson

R. O.

(Eds.), Handbook of bereavement: Theory, research, and intervention (pp. 208–226). Cambridge: Cambridge University Press.

54.

Suitor

J. J.

Pillemer

Sechrist

(2006). Within-family differences in mothers’ support to adult children. Journals of Gerontology Series B: Psychological and Social Sciences, 61, S10–S17. doi:10.1177/0164027507303636

55.

Umberson

(1992). Relationships between adult children and their parents: Psychological consequences for both generations. Journal of Marriage and the Family, 54, 664–674.

56.

Umberson

(1995). Marriage as support or strain? Marital quality following the death of a parent. Journal of Marriage and the Family, 57, 709–723.

57.

Umberson

(2003). Death of a parent: Transition to a new adult identity. New York, NY: Cambridge University Press.

58.

Umberson

Chen

M. D.

(1994). Effects of a parent’s death on adult children: Relationship salience and reaction to loss. American Sociological Review, 59, 152–168. doi:10.2307/2096138

59.

Videon

T. M.

(2005). Parent-child relations and children’s psychological well-being: Do dads matter? Journal of Family Issues, 26, 55–78. doi:10.1177/0192513X04270262

60.

Von Hippel

P. T.

(2007). Regression with missing Ys: An improved strategy for analyzing multiply imputed data. Sociological Methodology, 37, 83–117. doi:10.1111/j.1467-9531.2007.00180.x

61.

Willson

A. E.

Shuey

K. M.

Elder

G. H.

(2003). Ambivalence in the relationship of adult children to aging parents and in-laws. Journal of Marriage and Family, 65, 1055–1072. doi:10.1111/j.1741-3737.2003.01055.x

62.

Zettel

L. A.

Rook

K. S.

(2004). Substitution and compensation in the social networks of older widowed women. Psychology and Aging, 19, 433–443. doi:10.1037/0882-7974.19.3.433