| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
Department of Masters in Gerontology, Center for Research and Study of Aging, Faculty of Welfare and Health Studies, The University of Haifa, Israel.
Address correspondence to Ariela Lowenstein, Head, Center for Research and Study of Aging, Faculty of Welfare and Health Studies, The University of Haifa, Haifa, Israel 31905. E-mail: ariela{at}research.haifa.ac.il
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Data were from a sample of 2,064 elders (aged 75 and older) from the five-country OASIS study (Old Age and Autonomy: The Role of Service Systems and Intergenerational Family Solidarity; Norway, England, Germany, Spain, and Israel). Multivariate and block-recursive regression models estimated the predictivity of the two conceptualizations of family dynamics on quality of life controlling for country, personal characteristics, and activity of daily living functioning.
Results. Descriptive analyses indicated that family solidarity, especially the affective/cognitive component (called Solidarity A), was high in all five countries, whereas conflict and ambivalence were low. When I entered all three constructs into the regression Solidarity A, reciprocal intergenerational support and ambivalence predicted quality of life. Controlling for activity of daily living functioning, socioeconomics status, and country, intergenerational relations had only a weak explanatory power, and personal resources explained most of the variance.
Discussion. The data suggest that the three constructs exist simultaneously but in varying combinations, confirming that in cross-cultural contexts family cohesion predominates, albeit with low degrees of conflict and ambivalence. The solidarity construct evidenced relatively robust measurement. More work is required to enhance the ambivalence measurement.
Social gerontology has witnessed few conceptual and theoretical conflicts since the debate over disengagement theory more than 40 years ago. Recently, however, a controversy has developed over two competing paradigms of parent–child relations in later life: the solidarity–conflict model versus the intergenerational ambivalence model. These offer different conceptual lenses for understanding complex family relationships in societies undergoing social change. They provide different ways to understand microlevel interpersonal relations and macrolevel structural forces and the interactions between them. The clash is between social psychologists, who developed and tested the longstanding solidarity–conflict paradigm (Bengtson & Roberts, 1991
; Parrott & Bengtson, 1999); and critical theorists, who advocate applying the concept of ambivalence to intergenerational relationships (Connidis & McMullin, 2002b; Luescher & Pillemer, 1998).
The premise of the solidarity–conflict model is that levels of cohesion and conflict predict parent–child relations and their consequences in later life; the ambivalence model states that adult intergenerational relations revolve around sociological and psychological contradictions. Further analysis of the two paradigms can enrich researchers' understanding of the complex social phenomena involved in family relations in later life. This is important because of the profound increase in average life expectancy, which means that more people spend more years within family structures while these structures are constantly changing. Moreover, population aging and globalization have increased the diversity and complexity of family lives and intergenerational bonds (Lowenstein & Bengtson, 2003).
This article has two goals: (a) to explore the controversy behind the ambivalence versus solidarity–conflict models and to empirically examine which is the more useful or accurate model for explaining parent–child relations in adulthood, and (b) to compare the predictive adequacy of each model for quality of life (QOL) of older people in five societies that differ in family culture characteristics and public welfare policies.
Three Conceptualizations of Parent–Child Relations in Adulthood
Intergenerational solidarity in later life
The paradigm of intergenerational solidarity represents an effort to conceptualize family relations in adulthood and to develop a theory about differences between parent–child dyads in such relations. The solidarity model, first proposed in the 1970s, is a taxonomy for describing sentiments, behaviors, and attitudes in family relationships (Roberts & Bengtson, 1990). The first attempt to use this model was 30 years ago (i.e, Bengtson, 1975); other researchers have subsequently critiqued, modified, and expanded upon it (e.g., Atkinson, Kivett, & Campbell, 1986; Rossi & Rossi, 1990). The paradigm reflects several theoretical traditions, including (a) early theories of social organization (e.g., Parsons, 1973), (b) social psychology of group dynamics (Homans, 1961), and (c) the family development perspective (e.g., Hill, Foote, Aldous, Carlson, & MacDonald, 1970). Bengtson and colleagues eventually demonstrated six dimensions of parent–child solidarity: association, affect, consensus, function, normative, and family structure (Bengtson & Schrader, 1982). Each of these dimensions was empirically proven as distinctive (orthogonal) and as representing a dialectic (high vs low). Further analyses suggested that the six dimensions reflect two underlying dimensions: (a) structural–behavioral (comprising association, function, and structure), and (b) affective–cognitive (comprising affect, consensus, and normative solidarity; Roberts & Bengtson, 1990).
Subsequent research has demonstrated several advantages of the model. It focuses on family cohesion as an important component of family relations, particularly for enhancing psychological well-being in old age (Silverstein & Bengtson, 1994) and even for longevity (Silverstein & Bengtson, 1991). It emphasizes that intergenerational relations are multidimensional (Silverstein & Bengtson, 1997). Researchers have widely used it to study variations in parent–adult child relations in various ethnic groups (e.g., Kauh, 1997) and cross-national contexts (e.g., Lowenstein & Ogg, 2003).
The solidarity framework remained a dominant paradigm in social gerontology for two decades. However, some scholars raised concerns about the model being normative (i.e., that it points to how family relationships should be rather than how they are). The very term solidarity implies consensus, although there are obviously nonconsensual aspects of family relationships. Critics argued, therefore, that the solidarity model contains normative implications that easily lend themselves to idealization (Marshall, Matthews, & Rosenthal, 1993). Some also asserted that the model does not take into account conflict, nor does it provide insight into conflictual intergenerational relationships (Luescher & Pillemer, 1998).
Intergenerational conflict in later life
The solidarity model proved adaptable to innovations in methods and to challenges to its dominance and universality. Researchers modified the paradigm in the 1980s to become the family solidarity–conflict model, which incorporates conflict and considers the possible negative effects of too much solidarity (Silverstein, Chen, & Heller, 1996).
In developing the intergenerational conflict model, Bengtson and others (Clarke, Preston, Raskin, & Bengtson, 1999; Parrott & Bengtson, 1999) argued that conflict is a normal aspect of family relations, that it affects the way family members perceive one another, and that consequently it affects their willingness to assist one another. Conflict can mean that some difficult issues never are resolved, but that others are, over time, and that the overall quality of relationships improves rather than deteriorates. Solidarity and conflict do not represent a single continuum from high solidarity to high conflict. Rather, family relations can exhibit both high solidarity and high conflict, or low solidarity and low conflict, depending on family dynamics and situations.
Bengtson and colleagues saw conflict as a natural part of human life (the basic assumption of conflict theory) and as representing a separate dimension of family intergenerational relations (Bengtson, Giarrusso, Mabry, & Silverstein, 2002). In formulating the solidarity–conflict model, Bengtson and Silverstein represented a group of contemporary theorists of aging who viewed conflictual relations as an important element in understanding aging as part of a system of age stratification, wherein relations between different age groups are based not just on norms of reciprocity or equality of exchange. These revisions of the solidarity model, which was developed as an inductive approach, exemplify the scientific process of theory building that aspires to build cumulative knowledge and uses empirical testing as a means of assessing the utility of a model or theory (Katz, Lowenstein, Phillips, & Daatland, 2005). Recently, Giarrusso, Silverstein, Gans, and Bengtson (2005) advocated multidimensional typologies based on solidarity and conflict dimensions.
Intergenerational ambivalence in later life
Luescher and Pillemer (1998) introduced the term ambivalence, which reflects contradictions and ambiguities in relationships, as a valuable revived conceptual perspective for studying parent–child relations in later life. They noted that the term ambivalence has had a relatively long history in the field of psychology—both in psychotherapy and in research on attitudes in close relationships—and that in the field of sociology it reflects postmodern approaches to the family. They proposed intergenerational ambivalence to "designate contradictions in relationships between parents and adult offspring that cannot be reconciled" (Luescher & Pillemer, 1998, p. 416). The concept of ambivalence, they argued, should be the primary topic of study of intergenerational relations, because "societies and the individuals within them are characteristically ambivalent about relationships between parents and children in adulthood" (Pillemer & Luescher, 2004, p. 6).
Several years later, Connidis and McMullin (2002b) submitted an article to the Journal of Marriage and Family proposing a reconceptualization of ambivalence tied to critical theory. They emphasized "socially structured" ambivalence, which they described as "both a variable feature of structured sets of social relationships and a catalyst for social action" (p. 559). They based their approach to ambivalence on the connection between individual experiences, social relationships, social institutions, and societal change (Connidis & McMullin, 2002a).
The editor of Journal of Marriage and Family, Alexis Walker, sensed a controversy that could stimulate discussion about theory in family sociology. She invited three other scholars (including Luescher and Bengtson, whose conceptualizations Connidis and McMullin had criticized) to provide comments and rebuttals. This was, after all, the rationale for critical theory, which Connidis and McMullin espoused. And rebuttals there were. For example, Connidis and McMullin (2002b) had argued that Luescher "conflate[d] institutions and social structures...while they themselves specif[ied] social structure" (p. 600). Luescher (2004) responded that this was inaccurate: "Ambivalence is based on attributions and as an interpretation of modes of behaviour, cognitions, and emotions which can be conditioned by social structures or located within them" (p. 58). In their response, Bengtson and colleagues (2002) questioned the utility of the ambivalence construct. Ambivalence does not necessarily tie individual agency and social structure together, and it may be a motivator to do nothing at all. They asked how ambivalence differs from the classic symbolic interactionist depiction of role conflict. They wondered how ambivalence could be operationalized as a variable to predict or explain differences in intergenerational family dynamics. They concluded that the ambivalence concept complemented rather than competed with the solidarity–conflict framework, which was conceptually adequate for exploring mixed feelings: "From the intersection of solidarity and conflict comes ambivalence, both psychological and structural" (p. 575). They argued that both the solidarity–conflict and ambivalence models could be regarded as lenses "through which one can look at family relationships—complementary instead of competing" (p. 575).
Pillemer and Suitor (2002) and Luescher (2004) have provided empirical support for ambivalence. Other scholars have attempted to measure ambivalence in parent–child relations in adulthood to provide an empirical assessment, with mixed results (Fingerman & Hay, 2004; Wilson, Shuey, & Elder, 2003). These studies demonstrate the need for further conceptual and empirical development.
Intergenerational Relationships and QOL
In all societies, the family holds a crucial position at the intersection of generational and gender lines. Because individuals live longer and therefore share more years and experiences with members of other generations, intergenerational bonds among adult family members may be more important today than they were in earlier decades. In addition, researchers may best understand the needs of older people and their QOL within the context of the family (Bengtson, 2001). Intergenerational relationships are one of the elements that affect subjective QOL and are important components in family relations, especially for successful coping and social integration in old age. The presence or absence of positive intergenerational relations affects an individual's self-esteem and psychological well-being (e.g., Silverstein & Bengtson, 1991).
Testing the Models: Accounting for Variation in Elders' Family Relationships
Although both the solidarity–conflict and ambivalence models have strong advocates, no study to date has directly compared them. Using a common set of data from the cross-national OASIS study (Old Age and Autonomy: The Role of Service Systems and Intergenerational Family Solidarity), this analysis aimed to examine evidence regarding each model in a highly diverse sample of elders and to assess the models' utility in accounting for variations in QOL.
Research Question 1
The first research question was the following: How different are the observed patterns of intergenerational solidarity, conflict, and ambivalence across several societies that differ in welfare provision and family traditions? On the one hand, I hypothesized that the different dimensions of solidarity would be much stronger in countries with more familistic family cultures (i.e., Spain or Israel, where there is, for example, a legal obligation for children to support aging parents) compared to the other three countries (Hypothesis 1a). On the other hand, I hypothesized that conflict and ambivalence would be much lower (Hypothesis 1b). For support exchanges, conflict would be lower in Israel and Norway (as the more developed welfare states with a broad service network for elders) compared to the other three countries.
Research Question 2
The second research question was the following: Are there differences between the effects of these concepts as predictors of individuals' QOL, and, if so, do they affect the QOL of older members controlling for personal, activity of daily living (ADL) functioning, and country variables? Specifically, I hypothesized that solidarity, conflict, and ambivalence would all contribute to QOL as complementing constructs, as suggested by Bengtson and colleagues (2002), and that solidarity would have a significant positive impact and conflict and ambivalence a negative one (Hypothesis 2a); and ambivalence would be a stronger negative predictor of QOL than would solidarity or conflict, as suggested by Luescher and Pillemer (1988; Hypothesis 2b).
The OASIS study represents a comparative perspective and draws on data from five countries: Norway, England, Germany, Spain, and Israel. These countries reflect a diverse range of welfare regimes (institutional, conservative, residual) and familial cultures (family oriented, individualistic), differences that may reflect in intergenerational family relationships and may impact elders' QOL.
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Study researchers based the decision to restrict samples to urban areas on the premise that potential differences between countries depend in part upon stages of urbanization. Urban areas, defined as cities with more than 100,000 inhabitants, were the primary sample units. In Norway, Spain, and Israel, study researchers included all such urban units whereas in England and Germany they made a selection of urban areas. The sample strategies on the individual level differed slightly between the countries according to national conditions and availability of registries. Study researchers chose the different strategies because they represented the best research practice in each country based on its legal system and the ability to obtain addresses from registries. Subcontracted survey research organizations performed the field work.
About 12% of the gross sample consisted of natural dropouts, a category that included faulty addresses, people afraid to open doors, and elders who were not independent. The analysis of systematic drop out shows a distribution typical for surveys (Motel-Klingebiel, von Kondratowitz, & Tesch-Roemer, 2004). Table 1 presents the comparative distribution of the background variables.
|
Measures
The team of researchers compiled the OASIS questionnaire with the cooperation of all country teams and based its design on scales that have been frequently used and validated. A basic master version and an operational manual in English was compiled. Using the standard back translation method, each country conducted seven pretests and introduced revisions along the way. The research team accepted the eighth version as the final version.
Intergenerational solidarity
Researchers selected solidarity items for the OASIS study from the Longitudinal Study of Generations. The LSOG instrument contains 54 items relating to the respondent's children or parents as follows: (a) proximity, or geographic distance that might constrain or facilitate interaction, on a six-point scale, ranging from 1 = living 3 hr or more traveling distance, to 6 = living together; (b) association, or frequency of face-to-face contact, coded as 1 = several times a year, to 6 = daily or more often; (c) affect, or feelings of emotional intimacy between family members, determined by three questions like "How close do you feel to (this child)?" The questions were coded from 1 = not at all, to 6 = extremely; (d) functional, or instrumental assistance operationalized as receiving from or providing help to at least one child and/or parent in the following areas: shopping and transportation, household chores, house repair and gardening, personal care or child care, financial assistance, and emotional support. The questions were asked about all children and a mean score was used; (e) consensus, or degree of similarity in opinions and values, coded as 1 = not at all similar, to 6 = extremely similar.
Factor analysis performed for all countries (pooled samples) revealed a two-factor structure. The first (labeled Solidarity S) reflected the structural–behavioral dimension (proximity and contact); factor loadings = 0.9015 and 0.8715, respectively, 
=.84. The second (labeled Solidarity A) reflected the affective–cognitive (affect and consensus) dimension; factor loadings = 0.7526 and 0.9833, respectively, =.69. This dual structure is somewhat similar to that noted by Bengtson and Roberts (1991) and Silverstein and Bengtson (1994). Receiving or providing help (labeled Solidarity H-1 and H-2, respectively) did not emerge in the factor structure, and I thus used it separately in the analyses.
Conflict
Based on the Longitudinal Study of Generations, researchers measured conflict with three items relating to the degree of conflict or tension, criticism, and arguments between the generations, coded as 1 = none at all, to 6 = a great deal. I used a mean score.
Ambivalence
Researchers originally measured ambivalence with three items, based on those designed by Luescher and Pillemer (1998). For this analysis I used only the following item: "Sometimes family members can have mixed feelings in their relationships. Thinking about your relationships with your parent/child, how often do you have such mixed feelings?" The response rates were from 1 = very often, to 5 = never.
QOL
The researchers measured QOL with the World Health Organization Quality of Life-BREF inventory (World Health Organization Quality of Life Group, 1998). The instrument was designed for use in cross-cultural and cross-societal research. The scale is multidimensional and covers 24 facets of QOL with indicators relating to physical health, psychological well-being, and satisfaction with social relationships and with living conditions. Items are rated on a five-point Likert scale, from 1 = very dissatisfied, to 5 = very satisfied. Scale reliability was 0.65 to 0.87. Factor analysis of the 24 items revealed a one-factor structure, which I used in the analysis (factor loadings = 0.75–0.84; =.72). I computed a mean score.
Personal characteristics
I also measured gender (1 = male, 0 = female), marital status (1 = married, 0 = not married), number of living adult children older than 21, highest level of education attained (1 = primary, 2 = secondary, 3 = higher), and financial adequacy (1 = comfortable, 0 = not comfortable). The short version of the Short Form-36, which contains 12 items (Ware & Sherbourne, 1992), measured ADL functioning on a scale of 0 to 100 (higher score indicating better functioning). I selected these attributes because several studies (e.g., Fernandez-Ballesteros, Zamarron, & Ruiz, 2001) have found them to affect family relations and well-being.
Data Analysis
I analyzed the data in two phases. First, I calculated descriptive statistics for solidarity, conflict, ambivalence, and QOL comparing the five countries, using analysis of variance (ANOVA) and Duncan Multiple Range tests for differences between countries. Second, I calculated a block-recursive regression to examine the effects of the three concepts on QOL, and their differential impact on it, controlling for country, personal characteristics, and ADL functioning.
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
Levels of conflict appeared to be low in all countries (Ms = 1.3–1.7), with a score of 6 indicating high conflict. The Israeli sample reported the highest level of conflict, but the differences were minor, F(4, 1456) = 11.63, p <.0001. Duncan tests showed that three groups were formed: Israel by itself, Germany and Spain, and Norway and England.
Ambivalence also appeared low in all countries (Ms = 1.6–1.7), with Norway, Germany, and Spain forming the first group; followed by Israel; and England as a third group, F(4, 1456) = 12.36, p <.0001.
To answer the first part of Research Question 2 (Are there differences between the three concepts as predictors of elders' QOL?), the second part (Is there a difference between the effect of the three constructs on QOL controlling for country, personal, and ADL functioning?), and the hypotheses related to it, I performed a block-recursive regression with four models. The first contained only country variables. In the second, I added family variables. In the third, I entered personal variables and ADL functioning. The fourth included interactions between countries and the solidarity–conflict and ambivalence dimensions. However, results indicated that this fourth model added barely 2% to the variance, and the number of significant interactions was small. Thus, I present only these significant interactions. Table 3 shows these data.
|
Model 1, which contained only country variables, showed that Norway, England, and Germany differed from Spain and Israel, confirming partially Hypothesis 1a. Country variables, though, contributed 9%. Model 2 indicated that three of the solidarity dimensions—Solidarity A (affect and consensus), Solidarity H-1 (help received), and Solidarity H-2 (help provided)—impacted QOL. The exception was Solidarity S (proximity and contact), which was surprising. Ambivalence was negatively associated with OQL, whereas conflict was not. However, the explained variance for all the intergenerational variables was rather low at 10%. Those who indicated higher ambivalence rated QOL lower. Thus, Hypothesis 2a was partially confirmed. A similar pattern emerged when I performed regressions for each country separately.
Help received was negatively associated with QOL, meaning that those who received more help (apparently being more limited physically) perceived their QOL as lower than did the group who received less help. These findings suggest that dimensions of solidarity (except Solidarity S) are somewhat stronger predictors of QOL than ambivalence.
The regression data indicated that the overall explained variance of QOL by country, intergenerational relations, and personal characteristics was 47%. The most powerful predictors of QOL were ADL functioning and financial adequacy; with the addition of education, they contributed 27% to the explained variance. Solidarity S and conflict had no significant impact on QOL in any of the models. In the fourth model, however, when I added interactions, ambivalence had no effect and Solidarity H-1 and Solidarity H-2, minimal ones. The only significant interactions found were for Solidarity H-1 in Norway and Germany and for conflict and ambivalence in Germany.
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Solidarity–Conflict Versus Ambivalence in Different Cultures
The majority of respondents in all five countries reported strong and positive emotional solidarity (Solidarity A), whereas negative intergenerational emotions (conflict and ambivalence) were rather low. These findings support the assertion that in cross-cultural contexts, extended families today have maintained considerable cross-generational cohesion with some conflict (Bengtson, 2001) albeit with some ambivalent feelings (Luescher, 2004; Pillemer & Luescher, 2004). The data thus support the perspective of the solidarity–conflict model, but further study of the balance between solidarity and conflict is needed (Clarke et al., 1999), as is further exploration of ambivalence as it is "yet an understudied predictor of parents' well-being" (Pillemer & Suitor, 2002, p. 611).
Similarities and differences found between the countries on the various dimensions of intergenerational relations reflect variations in family norms and behavior patterns, as well as traditions of social policy in the participating countries. This heterogeneity can, as Silverstein and Bengtson (1997) suggested "be attributed to historical trends over the last century, such as geographic and economic mobility of generations or increasing numbers of later-life families" (p. 454).
In linking the testing of solidarity–conflict and ambivalence on the microlevel of individuals and families to the macro perspective of the cross-national study, one must consider historical and familial developments in the context of the countries involved. The higher rates of close parent–child relationships found in Israel may be closely related to the country's recent history and geopolitical situation. However, the higher rates of conflict might reflect a culture that encourages very open and frank communication between generations (Katz & Lavee, 2005). Similarly, the apparent generation gap between current cohorts of older parents and their adult children in Germany may be related to the polarization along generational lines of traditional/radical attitudes that occurred in the 1960s. In Spain, findings of relatively low rates of close parent–child relationships, which were contrary to expectations, may be due to rapid modernization (reflected, for example, in low fertility rates). Younger generations are more exposed to this process and are better educated and better off than their parents. This could have resulted in the emergence of a significant generation gap.
The OASIS countries also represent different contexts and opportunity structures for family life and elder care. Although confronted by similar challenges, like the growing numbers of elders, they have taken different strategies toward solutions. Of particular interest is the fact that Germany and Spain have welfare policies that favor family responsibility, with welfare provisions playing a secondary (Germany) or even residual role (Spain). In both countries there are legal obligations between generations but they have relatively low levels of social care services for elders, although Germany provides high levels of medical services. By comparison, England and Norway have individualistic social policies, no legal obligations between generations, and higher levels of social care services. Younger generations there find it more possible to combine work with family obligations than do their counterparts in Germany and Spain. The mixed Israeli model is illustrated by legal family obligations (as in Spain and Germany) and high service levels (as in Norway).
Solidarity–Conflict Versus Ambivalence as Predictors of QOL
I have also examined the validity of the three concepts under review as predictors of individuals' outcomes (QOL). The data indicate that, of the family relations variables, when entered separately into the regression, the affective–cognitive factor (Solidarity A) and the reciprocal exchange of support had the greatest predictive value, followed by ambivalence. Conflict had no effect on QOL (Model 2, Table 3). The conclusion, therefore, is that the solidarity dimensions have a somewhat better predictive validity for QOL, even though ambivalence also contributed. This is basically congruent with other studies that have shown that affectual solidarity, in particular, both was associated with greater longevity of older parents who experienced losses (e.g., Silverstein & Bengtson, 1991) and contributed to well-being both directly and indirectly (Venkatraman, 1995).
Regarding the reciprocal exchange of support, studies have shown the relative importance of positive and negative exchange (e.g., Newsom, Rook, Nishishiba, Sorkin, & Mahan, 2005), in that help received was related to help needed but with an adverse effect on the relationships. Other studies have revealed that if the support exchanged was reciprocal, elders reported a higher QOL (e.g., Kim & Kim, 2003), thereby supporting one of the main ideas underlying exchange theory, which forms the basis for the solidarity paradigm—the norm of reciprocity between generations.
The structural–behavioral factor (Solidarity S) was not statistically related to QOL, which was somewhat surprising. Yet help provided and received, especially if a reciprocal exchange, did contribute to elder QOL. Because the exchange of support entails contact, conceivably the Solidarity S factor was confounded with these dimensions of exchange.
The Contribution of Personal Resources to QOL
Even though researchers have generally found that family ties affect the psychological well-being of the individual throughout the life course (e.g., Rossi & Rossi, 1990), some studies have emphasized the importance of personal resources over family relations (Fernandez-Ballesteros et al., 2001). QOL in this study was associated with level of physical functioning and with education and financial adequacy. Apparently, people older than 75 perceive personal resources that relate to basic needs as the first priority for the QOL. These findings correspond to previous research showing that social integration, good health, and high socioeconomic status are the central predictors of subjective QOL (Pinquart & Sorensen, 2000).
Conclusions and Future Research Directions
The present analysis represents a first attempt to empirically examine the accuracy of the clashing theoretical claims of the solidarity–conflict versus ambivalence models of older parent–adult child relations from a cross-national perspective. The solidarity–conflict model was especially useful in evaluating the strength of family relationships in the different societies, as indicated by Bengtson and Roberts (1991). However, the model does not claim to capture the entire complex and diverse picture of late-life family relations, as noted by Bengtson and colleagues (2002). This is especially true at points of transition along the life course, such as the failing health of older parents or the changing needs of working caregivers, when more negative and/or ambivalent feelings may surface (Wilson et al., 2003). Nevertheless, the OASIS study demonstrates the validity and utility of the solidarity dimensions, especially the emotional–cognitive factor (Solidarity A), for expanding knowledge of the key dilemmas identified in the intergenerational relations literature. Researchers have examined solidarity–conflict primarily by using quantitative measures and, as Giarrusso and colleagues (2005) indicated, "Continuing efforts at refining the measurement properties of solidarity and conflict items have made this protocol the ‘gold standard’ in assessing intergenerational relations" (p. 415).
The OASIS design allowed for testing the positivist model of solidarity–conflict. Conflict did not have any effect on QOL, and ambivalence had little effect. It may be harder to capture some of the key components of ambivalence with survey measures compared to in-depth interviews, as may be the case with conflict, wherein a multiple domain approach might be more fruitful (Clarke et al., 1999). Also, the operationalization of ambivalence that was in its infancy when the OASIS study started is currently more developed (Fingerman, Chen, Hay, Cichy, & Lefkowitz, 2006), which might provide further insight for its understanding.
Conceivably, the conflict and ambivalence concepts are useful heuristically but difficult to measure empirically. Moreover, they may be prone to social desirability when measured quantitatively. Thus, testing these two concepts with less normative samples, such as in cases of elder abuse or estranged families, might yield new insights. I would recommend using both quantitative and qualitative methods of data collection and the triangulation of databases in order to further address and examine these different concepts.
A word of caution is in order about the limitations of this study, which in turn suggest directions for future research in this area. First, the analysis reports one side only of the parent–child relationship: the parent's viewpoint. In order to fully understand the complexity of family relations, researchers should also examine the adult child's point of view and that of other family members. Understanding dyadic relations within the total context of family networks and roles may further help to test the utility of the two paradigms. Second, the OASIS data are cross-sectional and show a static family relations situation. Replication and extension of the analysis using a longitudinal design would provide a more dynamic picture. Third, although the research design was comparative, testing these paradigms empirically in less developed countries would be beneficial (Wenger, 2005).
The findings indicate that solidarity is a robust concept and that high levels of solidarity are reported in diverse countries, although it may take other forms when circumstances change. The data underscore the process, in an intergenerational context, of individuals actively negotiating and renegotiating solutions and management strategies in response to change and transitions over the life course (Katz et al., 2005). Scholars should consider the possible paradigmatic changes in the social fabric of families and in societal networks that might impact family intergenerational relationships in the future.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication May 8, 2005. Accepted for publication June 7, 2006.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. L. Fingerman, L. Pitzer, E. S. Lefkowitz, K. S. Birditt, and D. Mroczek Ambivalent Relationship Qualities Between Adults and Their Parents: Implications for the Well-Being of Both Parties J. Gerontol. B. Psychol. Sci. Soc. Sci., November 1, 2008; 63(6): P362 - P371. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Ward Multiple Parent-Adult Child Relations and Well-Being in Middle and Later Life J. Gerontol. B. Psychol. Sci. Soc. Sci., July 1, 2008; 63(4): S239 - S247. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
