| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
1 Department of Sociology, Georgia State University, Atlanta, Georgia.
2 Institute on Aging, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.
Address correspondence to Dr. Donald C. Reitzes, Department of Sociology, Georgia State University, Atlanta, GA 30303. E-mail: socdcr{at}langate.gsu.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Data were collected in 1997 from 203 older grandmothers and grandfathers living in a metropolitan area. Grandparent and parent identity meanings are measured with an introductory identifier focusing attention on being a grandparent or a parent, followed by a set of 10 adjective pairs to capture identity meanings. Intergenerational family identity combined grandparent and parent identity meanings. Self-esteem and depressive symptoms serve as two indicators of well-being.
Results. We find that there are no significant differences between grandparent and parent identity meanings and that men and women are more positive about their grandparent identities than they are about other adult identity meanings. Further, grandparent identity is significantly related to well-being when it is the only identity in the model but not when parent identity is included in another model. Finally, intergenerational family identity is positively related to well-being.
Discussion. The findings confirm the expectation that grandparent identity meanings may encourage well-being. Further, the intergenerational identity reflects the overlapping meanings and experiences of being a parent and a grandparent.
Grandparenthood is interesting for several reasons. As a possible consequence of the fact that there are relatively few role models available from an earlier generation (Uhlenberg & Kirby, 1998
), the grandparent role is unusually varied in its form and function. Indeed, several studies note that the grandparent role has few explicit norms and describe it as a "tenuous" or "ambiguous" role (Cherlin & Furstenberg, 1986; Silverstein, Giarrusso, & Bengtson, 1998). Further, the grandparent role is often associated with old age norms that are increasingly inappropriate, given the age range and activity levels of most grandparents (Gauthier, 2002). Finally, becoming a grandparent is unusual in that formal role entrance is not dependent upon the actions of the person but on the fertility of one's children (Whitbeck, Hoyt, & Huck, 1993). Grandparenthood is initially a dual process signifying both the transition to parenthood for the adult child and the transition to being a grandparent for his/her parents (Szinovacz, 1998a).
In order to study the grandparent identity, we will use data collected from 203 older men and women, all of whom had been employed full-time in 1992, living in the Raleigh–Durham–Chapel Hill, North Carolina, metropolitan area. Three related issues will be investigated. First, we are interested in exploring the grandparent identity. A new grandparent identity measure allows us to compare grandparent identity meanings with the meanings of the religious, friend, spouse, and parent identities. We are especially interested in whether grandparent identity meanings are similar to parent identity meanings. Second, we will investigate the relationship between the grandparent identity and well-being. As well-being is multidimensional, we include two indicators: self-esteem and depressive symptoms. Third, recognizing the connection between being a grandparent and a parent, we combine grandparent and parent identities to form an intergenerational family identity and will investigate the relationships between grandparent identities, parent identities, and the intergenerational family identity on well-being. Finally, there is some evidence in the literature that there may be gender differences in grandparent identity meanings, so both grandmothers and grandfathers were sampled, and the data analysis explores whether the identities are related to well-being in a similar manner for both men and women.
Review of the Literature and Research Expectations
Background issues
Identity, in general, and grandparent identity, in particular, bridge two theoretical traditions. Erikson (1963) proposed eight psychosocial stages that reflect the lifelong interplay between ego identity and social interaction. Healthy middle-aged and older adults, accordingly, face challenges of aging and the issues of generativity and ego integrity. From this theoretical perspective, grandparenting may enable older adults to work through psychosocial crises and enhance well-being (Roberto, Allen, & Blieszner, 1999). Kivnick (1982) used Erikson's psychosocial stages to suggest that grandparent identity contains meanings of immortality through descendants and the continuity of the family, being a valued elder, and reinvolvement with one's personal past. Symbolic interactionism borrowed identity from Erikson in order to focus on social and cognitive self meanings. Identity from a symbolic interactionist perspective refers to self meanings in roles as opposed to Erikson's more general ego integrity (LaRossa & Reitzes, 1993). As individuals move through the life course, acquiring new roles and exiting other roles, they actively construct identities in their roles. In an earlier analysis, we predicted that positive identities would contribute to increased well-being, and we found that worker and parent identities did influence self-esteem for older working men and women (Reitzes & Mutran, 2002).
Grandparent identity
Of all the studies of grandparenthood, including its norms and values, behaviors and conduct, relatively few have investigated the grandparent meanings and identity. This was true 40 years ago and still remains the case today (Neugarten & Weinstein, 1964; Szinovacz, 1998b). In one of the earliest studies, Neugarten and Weinstein (1964) suggested three dimensions: (a) the degree of comfort with the role, (b) the significance of the role, and (c) the style with which the role is enacted. Robertson (1977) distinguished a social and a personal dimension of the grandmother identity. More recently, Thomas (1990) constructed her own measure of grandparent identity, including symbolic meanings, authority, and satisfaction dimensions. Hayslip, Shore, Henderson, and Lambert (1998), in their investigation of custodial grandparents, created a positive grandparent meaning measure that combined Kivnick's five meaning dimensions (1982) with Thomas's satisfaction dimension (1990), and Silverstein and Marenco (2001) proposed an affective–cognitive dimension of the grandparent role that included perceived emotional closeness, self-assessment of performance as a grandparent, and the importance of being a grandparent.
Grandparent and parent identities
Symbolic interaction theory (Stryker & Burke, 2002) recognized that roles and identities do not exist in isolation but emerge in contrast and comparison with related role meanings. Thus, being a grandparent is expected to overlap and share meanings with being a parent or great-grandparent. However, although it is expected that roles share meanings with related roles, the relationship between grandparent and parent suggests extensive overlapping meanings.
Past research highlights the interesting and complex nature of the ties between grandparents, their adult children, and grandchildren. Kivett (1998) described the grandparent–parent relationship as one of interlocking roles. Several researchers recognized that parents serve as gatekeepers and mediate the amount of contact and the quality of the relationship between grandparents and grandchildren (Hodgson, 1992; Silverstein & Marenco, 2001; Whitbeck et al., 1993). The divorce of parents can have a direct impact on the quality and quantity of grandparent–grandchildren relationships, as can new kinship ties that result from the remarriage of parents (Johnson, 1998). Grandparenting also has been described as an extension of the parent role. Fingerman (1998) noted that grandparents may consider the successes or failures of grandchildren as a reflection of their parenting abilities. Roberto, Allen, and Blieszner (2001) concluded that men accepted responsibility for grandchildren out of a sense of obligation and love for their children. Among older women, Roberto and colleagues (1999) found that relationships with adult children were very important to them, whereas ties to grandchildren were meaningful to these women but much more peripheral to their everyday lives. Others have conceptualized the grandparent role from the perspective of common intergenerational ties, sentiments, and exchanges of support that link grandparents, parents, and grandchildren (Silverstein et al., 1998). Tomlin (1998) identified six direct and mediating paths of influence linking grandparents, parents, and grandchildren.
Research expectations
Although past studies have tended to focus on the distinctive meanings and characteristics of being a grandparent, we are interested in comparing grandparent identity meanings with identity meanings in different roles. Our initial research interest is to explore the extent to which the grandparent identity is similar to or different from other adult identities and especially whether grandparent identities are similar to parent identity meanings. Reitzes and Mutran (1994) propose that three meanings dimensions suggested by Mortimer, Finch, and Kumka (1982) are generally applicable to identity meanings in later life roles. They include items to tap (a) competence, which is similar to the evaluation dimension of Osgood, Succi, and Tannenbaum (1957), or the task-directed dimension of Bales (1951) and covers instrumental identity meanings; (b) confidence, which captures Turner's identity-directed dimension (1968) and reflects emotional or affective identity meanings; and (c) sociability, which indicates interest in others and is similar to Bales's expressive dimension. We found that these dimensions contain a common structure and can be summarized into a single grandparent identity. Thus, our grandparent identity meaning measure shares with that of Silverstein and Marenco (2001) the recognition of an affective–cognitive dimension to the grandparent role and with Thomas (1990) an understanding of the symbolic meanings of being a grandparent, as well as the efforts of Hayslip and colleagues (1998) to construct a positive grandparental meanings measure. However, our grandparent identity measure is different in that it will allow us to compare identity meanings across roles. We are interested in whether men and women hold more positive grandparent identity meanings than they do identity meanings in other roles and expect that as a result of overlapping meanings, there will be no differences in grandparent and parent identities.
The earlier work on the grandparent identity focused on grandmothers, so the question arises of whether there are similarities or differences in the way that men and women perceive themselves as grandparents. On the one hand, traditional gender norms suggest that women are expected to invest more of themselves in family and caregiving roles (Spitze & Ward, 1998), so women are expected to have stronger and more positive grandparent identity meanings than men. On the other, several studies report at least some similarities in the responses of men and women to grandparenthood (Cherlin & Furstenberg, 1986; King & Elder, 1998; Silverstein & Marenco, 2001). One of the advantages of our study is that our sample includes both men and women. We are able to explore gender differences in grandparent identity meanings.
The second direction of our research is to explore the relationship between grandparent identity meanings and well-being. Well-being reflects a wide range of cognitive and affective self-assessments (Rowe & Kahn, 1997), so we selected self-esteem and depressive symptoms to capture two very different dimensions of well-being. Self-esteem highlights cognitive and evaluative assessments of one's self-worth. It is a global self concept that tends to be stable across situations and over time (Rosenberg, Schoenbach, Schooler, & Rosenberg, 1995). Depressive symptoms, in contrast, tap an affective assessment of emotional well-being that may be more sensitive than self-esteem to current experiences and events and less stable over time. Although clearly different, we expect that these indicators of well-being will be related to grandparent identity meanings.
It is our expectation that grandparent identity meanings will be positively related to self-esteem and negatively related to depressive symptoms. Reitzes and Mutran (2002) found that whereas the grandparent role was not as personally important as the parent or spouse roles, it nevertheless was among the most important roles held by middle-aged workers. The grandparent role ranked fourth for both men and women. Positive identity meanings in the grandparent role, as a valued role, should encourage positive self-regard and thwart depressive symptoms. What is unclear is the extent to which grandparent identity meanings will be independently related to well-being when both grandparent and parent identity meanings are included in the same analysis. Third, we are interested in pursuing the implications of the interlocking ties between the grandparent and parent roles. Recognizing the overlapping linkages between grandparents, parents, and grandchildren (Fingerman, 1998; Silverstein et al., 1998; Tomlin, 1998), we created an intergenerational family identity measure that combines parent and grandparent identities. Our expectation is that intergenerational family identity meanings also will be positively related to self-esteem and negatively related to depressive symptoms.
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
. In this study, we focus on the third and final wave of data collection that occurred between March and June 1997, during which 255 people remained in the study either because they were still working or had not completed the last retirement interview. We limit our attention to the 203 men and women who identified themselves as grandparents and answered the 31 grandparent questions that were added for the 1997 data collection. Overall, the grandparent sample contains a diverse set of respondents with a variety of social background characteristics: 54% men, 87% White, 73% married, and 32% retired. This degree of sample diversity allows us to proceed with the primary goals of exploring the relationship between the grandparent identity and well-being.
Variables
Beginning with the two well-being variables, self-esteem is a global measure that focuses on the cognitive and social dimensions of self-concept. We use Rosenberg's self-esteem scale (1965), which has proven to be a durable and useful measure of a person's summary or global assessment of self. The scale includes 10 items such as "I feel that I'm a person of worth, at least on an equal basis with others" and "I wish I could have more respect for myself" (recoded). Responses range from "strongly agree" (4) to "strongly disagree" (1; 
=.89). Depressive symptoms come from Radloff's 20-item Center for Epidemiological Studies—Depression Scale (1977). Sample items include "I feel that I am just as good as other people" (recoded) or "I have trouble keeping my mind on what I was doing," with response categories of "rarely or none of the time" (1), "a moderate amount of time" (2), and "most of the time" (3; =.75).
Mortimer and colleagues (1982) provide the theoretical foundation for our identity meanings variables. To measure grandparent identity meanings, we begin with a stem, "As a grandparent, I am ...," followed by 10 adjective pairs organized in a semantic differential 5-point format (Osgood et al., 1957). The adjective pairs include active–inactive, successful–unsuccessful, competent–not competent, relaxed–tense, happy–sad, confident–not confident, warm–cold, open–closed, interested in others–interested in self, and sociable–solitary ( =.90). For each adjective pair, responses nearer to the positive adjective meaning (e.g., happy, confident, warm) receive higher scores (4 or 5), and responses closer to the negative or less positive adjective meaning (sad, not confident, or cold) receive lower scores (1 or 2, with middle responses of 3). Parent identity meanings contain the stem, "As a parent, I am ..." followed by the same adjective pairs ( =.85) and intergenerational family identity meanings, which combined scores from the 10 grandparent and the 10 parent adjective pairs ( =.92). The religious identity, friend identity, and spouse identity are likewise constructed from a stem identifying the role and the same set of adjective pairs (with scores of.90,.90, and.88, respectively). Thus, all the identities measure positive identity meanings.
Finally, we include in the analysis six social background variables that past research suggests may influence well-being (Reitzes, Mutran, & Fernandez, 1996). Male is a dichotomous variable with males coded "1" and females coded "0." Poor health measures functional limitations and is based on a seven-item scale of difficulties in performing activities such as walking, using stairs, standing, or sitting for long periods of time, bending, lifting weights up to 10 lbs., and reaching above your head, with responses of "never" (1), "sometimes" (2), or "often" (3; =.83). Retired is a dichotomous variable coded in the direction of being retired, and married comes from a dichotomous variable that identified respondents who were currently married, whereas family income reflects total household income with 10 response categories ranging from "$7,500 or less" (1) and "$35,001 to $50,000" (5) to "$200,001 and over" (10), and education records the highest grade completed in school and coded in years.
Table 1 reveals that our respondents are in good mental health. They exhibit high self-esteem and low depressive symptoms as well as generally positive identity meanings. The men and women surveyed are between the ages of 62 and 70 years with generally few functional limitations, one third are retired, and 73% are married, with an average income a little above the $35,001 to $50,000 range, and average years of education of 2 years past high school. We expect that poor health will be negatively related to well-being, and family income and education will be positively related to well-being, as will the social support of being married. Retirement, without the potential stress of the work environment, and controlling for poor health and income, also is expected to be positively related to well-being (Reitzes et al., 1996).
|
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
) tested a model that compared the 10 grandparent identity items for men and women. By progressively freeing the measurement error terms as indicated by the modification indexes and allowing them to vary between men and women, an acceptable fit was achieved (
2 = 106.93, df = 77, p =.014, root mean squared of the error approximation [RMSEA] = 063). Both the factor loadings and the error variance in the construct were found to be equal. Therefore, we conclude that there are not gender differences in the structure of grandparent identity meanings.
Grandparent Identity Meanings Compared with Other Identity Meanings
Now we are ready to compare grandparent identity meanings with identity meanings in other adult roles. The first panel in Table 2 focuses on the entire sample and presents means and standard deviations for the grandparent, religious, friend, spouse, and parent identities. Two findings are interesting. First, grandparent identity meanings are significantly different from—indeed are higher than—the means for the religious, friend, and spouse identity meanings. Second, there are no statistically significant differences between grandparent and parent identity meanings. The findings are consistent with theoretical descriptions that highlight the linkage and interlocking character of the grandparent and parent identities. The second panel allows us to compare the identity meanings of women and men. We find that women are not only more positive about their grandparent identity than are men, but that women are more positive about all five of their identities than are men. The findings also reveal that when we separately compared identities by gender, the findings of the first panel were confirmed. For women and for men, when analyzed separately, grandparent identity meanings were higher than the religious, friend, and spouse identity meanings and similar to the parent identity meanings.
|
). The results suggest that there are no systematic differences by gender, so we proceed to pool men and women and to enter gender as an independent variable with the other identity and social background factors. The results are presented in Tables 3 and 4.
|
|
Social background factors are also related to well-being. Poor health and family income are related to both indicators of well-being. Poor health is negatively related to self-esteem and positively related to depressive symptoms, whereas the reverse is true of family income. It is positively related to self-esteem and negatively related to depressive symptoms. Other social background factors are related to only one of the well-being indicators. Gender is related only to self-esteem. Women are more likely than men to have higher self-esteem. Education is positively related to self-esteem once parent identity and intergenerational family identity are in the analysis, as they are in Models 2 and 3. Finally, being retired is not related to self-esteem but is negatively related to depressive symptoms, and being married is not related to either indicator of well-being.
A second approach to assessing the relationship between grandparent identity meanings and well-being is to use linear structural equations. We include grandparent and parent identity as separate indicators of an underlying construct, intergenerational family identity. Both measures are strong predictors of the construct, and the model fits the data. The results for self-esteem produced a 2 of 9.14, with 7 df, probability of.24, and RMSEA of.04. The model for depressive symptoms yields a 2 of 7.81, with 7 df, probability of.35, and RMSEA of.02. Whereas differences in measures of estimating the equations produce somewhat different significant effects among the social background variables, the general patterns of positive and negative relationships are similar to the regression analyses. Of particular interest is the finding that intergenerational family identity meanings are significantly related to both indicators of well-being.
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
; Kivett, 1998; Roberto et al., 1999; Tomlin, 1998); including the dual transition into the roles (Szinovacz, 1998a) and that parents often mediate the quantity and quality of grandparent–grandchildren interaction (Hodgson, 1992; Silverstein & Marenco, 2001;Whitbeck et al., 1993). With our measure, we are able to test whether there are overlapping grandparent and parent identity meanings. We find that grandparent identity meanings are similar to parent identity meanings, while being notably higher than other adult role-related identities.
We also find that whereas the structure of the grandparent identity measure fits both men and women equally well, grandmothers have higher, more positive grandparent identities than do grandfathers. Further, women have higher mean identity scores on religious, friend, spouse, and parent identity meaning measures as well. We suspect that as a result of traditional gender norms, women may be expected to be more positive about themselves than men in their family, friendship, and caregiving roles (Spitze & Ward, 1998). The finding is consistent with our earlier one that while the role of worker is important for both men and women, women tend to attribute greater importance to other family roles and the role of friend than do men (Reitzes & Mutran, 2002).
Turning to the relationship between grandparent identity and well-being, we found grandparent identity meanings, when considered alone, are positively related to self-esteem and negatively related to depressive symptoms. Following symbolic interaction theory (Stryker & Burke, 2002), positive grandparent identity meanings encourage individuals to seek confirming behaviors and attitudes. More positive grandparent identity meanings may encourage a heightened sense of self-regard and lower depressive symptoms by providing a sense of authenticity, a sense of meaning and purpose, which enhances well-being (Gecas & Burke, 1995).
Finally, the experiences of being a grandparent are connected with the experiences of being a parent. Therefore, it is not surprising that when both grandparent and parent identity meanings are included in the same regression analyses, there are no statistically significant relationships between grandparent identity meanings and well-being indicators. However, combining grandparent and parent identity meanings into an intergenerational family identity measure produces a factor related to both self-esteem and depressive symptoms. Intergenerational family identity meanings are positively related to self-esteem and negatively related to depressive symptoms. Beyond its demonstrated relationship with well-being, intergenerational family identity further highlights the linkage between grandparent and parent roles. The parent and grandparent roles are interrelated, and it seems valid and appropriate to consider a nested, combined role, the intergenerational family role, that reflects the overlapping meanings and experiences of being a parent and a grandparent. The intergenerational family role may be particularly appropriate for the experiences of older adults. Among younger adults, there may be greater discord between being a grandparent and a parent than for older adults (Burton, 1996).
This research has at least three limitations. First, our sample is local. All respondents are from the Raleigh–Durham–Chapel Hill metropolitan area. Although the sample provides a diverse set of working and retired grandmothers and grandfathers, a nationally representative sample would allow us to generalize our descriptive findings. Second, a larger sample with more Blacks also would enable us to explore the distinctive grandparenting experiences of African Americans. Past research has noted the centrality of the grandparent role in African American culture (Burton, 1996). In an exploratory analysis of differences of means scores, we find that there are no differences between Blacks and Whites in their grandparent and parent identity meanings or in their mean well-being scores. However, a larger sample is necessary for more definitive results.
Third, in this study, we are limited to data gathered at a single point in time and so have only a snapshot of the grandparent identity. It would be fascinating to explore the transition into grandparenthood. Recognizing that identity meanings may change over time, it is possible that grandparent identity meanings upon entry into the role may be more positive than grandparent identity meanings later in the role career. As an initial exploration, we separated grandparents with grandchildren 3 years old and younger from grandparents with older grandchildren and compared their grandparent identity scores. We found that there were no statistically significant differences in mean scores, but these results are only tentative.
Further, we expect that the grandparent identity, as well as parent and the intergenerational family identity, would be more strongly related to grandparents' well-being when grandchildren are younger. Not only would grandparents be younger, but also we expect that the novelty of being a grandparent would increase the impact of the identity on well-being. So, in an exploratory probe, we separated grandparents with grandchildren 3 years old and younger from grandparents with older grandchildren and tested the same regression model presented in Tables 3 and 4. The results suggest that grandparent identity meanings and the intergenerational family identity meanings are more strongly positively related to self-esteem and negatively related to depressive symptoms for grandparents with younger grandchildren. The grandparent identity is dynamic and may vary in its significance to older adults and its relationship to well-being.
In conclusion, we believe that future research on grandparenthood would be well served to include grandparent identity in investigations of factors that influence well-being. We recommend the adoption of a grandparent identity measure that allows comparison with other adult identities. It would be interesting to explore further why, as suggested in Table 2, men and women may view themselves more positively in the grandparent identity than in the friend or spouse identity. Finally, it is very important to consider being a grandparent in conjunction with other family roles. A combined intergenerational family identity encourages the recognition of the overlapping experiences among older adults of being both a parent and a grandparent. It also suggests an interesting strategy for improving the well-being and mental health of older adults. Activities and programs that bring both adult children and grandchildren together with grandparents may have a strong effect on improving the well-being and mental health of grandparents.
|
Footnotes |
|---|
Received for publication September 12, 2003. Accepted for publication March 30, 2004.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. M. Musil, C. B. Warner, J. A. Zauszniewski, A. B. Jeanblanc, and K. Kercher Grandmothers, caregiving, and family functioning. J. Gerontol. B. Psychol. Sci. Soc. Sci., March 1, 2006; 61(2): S89 - S98. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
