Perceived effects of voluntarism on marital life in late adulthood.

The article presents a study dealing with the perceived effects of voluntarism on marital life in late adulthood among a sample of 595 Israelis (336 men and 259 women). These perceptions were examined from three perspectives: benefits, spousal accommodation, and harmful effects. Comparisons focused on different types of families, based on employment status (pre-retired versus retired) and actual volunteer activity (volunteer versus non-volunteer). The findings revealed that among all types of families, the prevailing tendency was to emphasize the beneficial effects of voluntarism on marital life, whereas perceived harmful effects were least prevalent. Synchronous families (both partners pre-retired) and asynchronous families (pre-retired participant / retired spouse) emphasized the need for spousal accommodation to marital life more than the other two types of families. In addition, men were found to emphasize the need for spousal accommodation more than do women. In asynchronous families (one partner pre-retired and the other retired), women showed a greater tendency than men to mention the harmful effects of voluntarism for the marital relationship. Congruent families (where both partners volunteer) showed a greater tendency to perceive voluntarism as having a beneficial effect on marital life than did other types of families.

**********

Voluntarism has become an integral part of leisure culture in modern societies. This trend can be attributed, among other causes, to the awareness that welfare states are no longer able to provide adequate responses to needy populations. Thus, in many Western societies volunteers play an important role in supplementing and improving the services provided by various human service organizations (Hayghe, 1991). In fact, there are those who view voluntarism as a career in itself. In this connection, senior citizens are a potential source of volunteer assistance because they generally have free time and extensive life experience (Fisher, Mueller, Cooper, & Chase, 1989). Given the growing centrality of voluntarism in the lives of senior citizens, there is a need to examine the factors that encourage voluntarism among that population. Although a number of studies have dealt with voluntarism among elderly populations, they have focused on specific issues such as motives for voluntarism in late adulthood (Herzog & House, 1991; Hulbert & Chase, 1991), health and voluntarism (Cohen-Mansfield, 1989) and background variables affecting voluntarism among elderly populations (Fisher & Schaffer, 1993). Thus, there is a lack of research on the family context of voluntarism in general and the perceived impact of voluntarism on marital relations in particular.

The basic assumption of the current study was that volunteer work, like paid employment, affects marital dynamics. On the one hand, it may enrich and vitalize spousal relations in late adulthood. On the other, when one partner volunteers and the other does not, jealousy and tension may ensue, especially considering that social contacts usually diminish after retirement (Kulik, 1999). Although attitudes toward the impact of spousal voluntarism may play an important role in stimulating or restraining volunteer activity among senior citizens, this topic has not received sufficient research attention. In an attempt to fill this gap, the current study examined the perceived impact of voluntarism on marital relations among elderly persons in Israel. These attitudes were examined from three perspectives: Perceived beneficial effects (e.g., the contribution of voluntarism toward enrichment of marital life); perceived harmful effects (e.g., marital tension or feelings of jealousy), and the need for spousal accommodation (e.g., taking the partner's schedule into consideration and accommodating volunteer activity to the partner's needs).

The study also examined whether the perceived impact of voluntarism on marital is affected by gender. Regarding gender differences, various studies have found that women show a greater tendency to volunteer than men, particularly in areas that conform to gender role expectations (Fisher, Mueller, & Cooper, 1991). Moreover, research has found that women are socialized to focus on nurturance and interpersonal relationships (Gilligan, 1982). In addition, it has been argued that women are better able than men to cope with the demands of different social systems such as work and family (Barnett, 1993). Based on this evidence, it was assumed that women are also better able than men to combine family responsibilities with volunteer work. It was therefore hypothesized that women will be less likely to emphasize the harmful effects of voluntarism, and more likely to emphasize its beneficial effects.

Another issue examined in the study was the relationship between employment status (pre-retired versus retired) and the perceived impact of voluntarism on marital life. It is commonly believed that retired couples have more free time, which affects marital dynamics (e.g., division of household tasks and satisfaction with marriage). However, existing studies have yielded inconsistent findings, based on examination of one partner's employment status (see, for example, Szinovacz, Ekerdt, & Vinick, 1992). Therefore, the combined effect of both partners' employment status needs to be considered. Toward this end, recent studies have compared the attitudes of synchronous couples (both partners retired or both partners employed) with those of asynchronous couples (one partner is retired and the other is pre-retired) (Kulik, 2001; Szinovacz, 1996; Szinovacz & Harpster, 1994). Based on this approach, the current study compared perceptions regarding the impact of voluntarism on marital relations among synchronous couples and asynchronous couples.

It was hypothesized that synchronous-retired couples will have the most positive attitudes toward volunteer activity, i.e., they will be less likely to emphasize the harmful effects of voluntarism or the need for spousal accommodation. This hypothesis was based on the rationale that when both partners are retired, there is a decline in marital tensions generated by work stress. Synchronous-pre-retired families, in contrast, are characterized by relatively high role stress, which may exacerbate the harmful effects of voluntarism and highlight the need to accommodate volunteer work to family life.

In this connection, another question arises: How does volunteer activity per se affect perceptions regarding the impact of voluntarism on marital life? The present study also examined this issue in the context of the dyadic unit. A distinction was made between congruent couples (where both partners either volunteer or neither partner volunteers) and incongruent couples (where one partner volunteers and the other does not).

Research Goals

1. To examine which perceptions regarding the impact of voluntarism on marital relations are most prevalent among elderly couples.

2. To examine whether there are differences in perceived impact of volunteer activity on marital relations, among each of the following three groups of participants:

a. Men versus women.

b. Synchronous couples (both spouses are pre-retired or retired) versus asynchronous couples (one partner is employed and the other is retired).

c. Congruent couples (both partners volunteer) and incongruent couples (one partner volunteers and the other does not).

Methods

Sample and Data Collection

The study presented here is part of a larger project that dealt with marital relations in late adulthood. Data were collected from December 1997 to July 1998, using self-administered survey questionnaires. The research sample consisted of 595 Jewish Israeli participants, of whom 341 (57.3%) were pre-retired (up to two years prior to retirement), and 254 (42.7%) were retired (up to six years after retirement). Among the pre-retired sample, interviews were conducted in pre-retirement workshops offered by various workplaces and organizations. Most of the workshops were held about a year to a year-and-a-half before retirement. The organizations chosen to participate in the study represented a broad range of sectors and occupations (e.g., services, manufacturing, and commercial organizations). The sample of retirees was drawn primarily from pensioners' clubs throughout the country (northern, southern, and central regions). Clubs were located in large urban localities and small development towns. The overall response rate was approximately 80%.

Demographic Characteristics of the Sample

Of all the participants, 336 (56.5%) were men and 259 (43.5%) were women. Participants ranged from 54 to 75 years of age. Regarding ethnicity, 186 (31.3%) were Israeli born, 244 (41.4%) were born in Europe, 68 (11.5%) were born in Asia, 60 (10.0%) were born in Africa, and 37 (6.3%) were born in America. As for religiosity, 345 participants (58%) defined themselves as secular, 176 (29.6%) defined themselves as traditional, and 74 (12.5%) defined themselves as Orthodox. Regarding education level, 189 (31.7%) had partial secondary education, 238 (40%) were secondary school graduates, and 168 (28.2%) had post-secondary (academic or other) education. With respect to volunteer activity, 253 of the participants (42.5%) volunteered, and 342 (57.5%) did not. In general, the distribution of participants by background variables reflected the distribution of the population in Israel in that age range (Central Bureau of Statistics, 1998).

Instruments

Perceived Effects of Voluntarism on Marital Life. The questionnaire was constructed for the purpose of this study, and included 10 items relating to the potential impact of voluntarism on marital life. Participants were asked to answer to the following question: What are your views about volunteering (regardless of whether you or your partner actually engage in such activity)? Responses were based on a scale ranging from 1 (strongly disagree) to 5 (strongly agree). Factor analysis by Varimax rotation was conducted in order to identify common content areas for the items that measured attitudes toward volunteering. The analysis revealed three factors which, taken together, explain 56.9% of the variance in this item. The first factor explains 25.4% of the variance, the second explains 18.0% of the variance, and the third explains 13.5% of the variance (see Table 1 for a list of the items in the questionnaire and their loadings).

Table 1 indicates that the first factor focus on the contribution of volunteering toward the marital relationship (henceforth "perceived beneficial effects"). The second factor focuses on the need to accommodate volunteer activities to marital life (henceforth "spousal accommodation"), and the third factor focuses on the harmful effects of volunteering on the dyadic unit (henceforth "perceived harmful effects").

For each of the three factors, one score was derived by the computing the means of the items. The Cronbach's reliability coefficient for the first factor ("perceived beneficial effects") was .82; the reliability coefficient for the second factor ("spousal accommodation") was 78, and the reliability of the third factor ("perceived harmful effects") was .40. Because there were only two items in the factor, the reliability coefficient was relatively low.

Background Variables

Data on several background variables were collected for participants and their spouses: Age, religiosity, education level, and ethnicity (Israeli, Asian-African, and European-American). In addition, participants were asked whether they or their spouses engage in volunteer activity, and how many hours per week they spend volunteering.

Data Analysis

Two-way MANOVA (4 x 2, family type x gender) was conducted in order to examine how family type (by employment status) and gender affect the perceived impact of volunteer activity on marital relations.

Four types of families were identified on the basis of employment status. Two types of families were defined as synchronous (both partners employed and both partners retired); and two as asynchronous (pre-retired participant / retired spouse; and retired participant / pre-retired spouse).

In addition, MANOVA was conducted in order to examine whether family type (by volunteer activity) affects perceptions regarding the impact of volunteer activity on marital relations. Four types of families were identified on the basis of volunteer activity. Two types were defined as congruent (both partners volunteer, and neither partner volunteers); and two as incongruent (volunteer participant / non-volunteer spouse; non-volunteer participant / volunteer spouse).

Results

a. Differences in Perceived Effects of Voluntarism, by Family Type (Employment Status) and Gender

Two-way MANOVA 2 x 4 (family type by employment status x gender) showed a significant effect for family type: F (9, 1302) = 3.02, p < .01, while univariate analysis conducted separately for each factor showed significant differences between the four types of families only for the spousal accommodation factor. Scheffe tests indicated that members of synchronous-preretired and asynchronous families (pre-retired participant / retired spouse) differed from members of the other two groups (synchronous retired, and pre-retired spouse / retired participant). Of the four family types, members of the first two groups showed a greater tendency to emphasize spousal accommodation.

The same MANOVA also revealed a significant effect for participant's gender: F(3,432) = 4.84, p < .01. However, when univariate analysis was carried out separately for each of the three factors, a significant effect was found only for spousal accommodation (see Table 2). In all four types of families, men showed a greater tendency than women to emphasize spousal accommodation. In addition, the MANOVA revealed an interaction between family type and gender for perceived harmful effects of voluntarism: F(2,437) = 4.18, p < .01 (see Table 3).

Simple main effects tests indicated that for both types of asynchronous families (preretired participant / retired spouse, retired participant / pre-retired spouse), the women were more likely to argue that volunteer activity harms marital wife. By contrast, no gender differences were found for either type of synchronous family (i.e., pre-retired and retired) with regard to the perceived harmful effects of voluntarism.

In order to examine differences in ranking of factors, MANOVA with repeated measures for family type was conducted. Significant differences were found between the three factors, across all four family types: F(2,430) = 4.18, p < .001. Simple main effects tests revealed that of these factors, perceived benefits of voluntarism had the most significant effect in all four types of families. By contrast, perceived harmful effects of voluntarism ranked lowest and generally received low scores (below midpoint 3 for all types of families) (see Table 2, "Total" column). The same test found a significant interaction between the three factors and family type with respect to spousal accommodation: F(6, 876) = 4.34, p < .01.

Newman Keuls tests indicated that among synchronous-preretired families and asynchronous (pre-retired participant / retired spouse) families, the participants felt that spousal accommodation is just as important as the beneficial effects of voluntarism. In contrast, members of the other two types of families differentiated between these two factors.

b. Differences in Perceived Effects of Voluntarism, by Family Type (Volunteer Activity)

One-way MANOVA revealed a significant effect for family type (by volunteer activity): F(9, 1584) = 5.58, p < .001. When separate univariate analysis was carried out for each factor, differences were found between the respective types of families for all three of the dimensions examined (beneficial effects, spousal accommodation, and harmful effects) (see Table 4).

Regarding perceived benefits of voluntarism, Scheffe tests indicated that congruent volunteer families differ significantly from congruent non-volunteer families and both types of incongruent families. Specifically, congruent-volunteer families showed a greater tendency to highlight the beneficial effects of voluntarism than did the other types of families. Regarding spousal accommodation, the findings revealed that congruent-volunteer families differed from incongruent (volunteer participant / non-volunteer spouse) families. Specifically, the latter type of families showed less of a tendency to emphasize the need for spousal accommodation. With respect to perceived harmful effects of voluntarism, differences were found between congruent-volunteer families and the other three family types. Specifically, congruent-volunteer families showed less of a tendency to emphasize the harmful effects of voluntarism on the marital relationship. In general, it can be argued that congruent-volunteer families differ from the other types of families (by volunteer activity). Specifically, the congruent-volunteers placed more emphasis on the benefits of voluntarism, were less likely to perceive volunteer activity as harmful, and were more sensitive to the need for spousal accommodation.

c. Relationship between Perceived Effects of Voluntarism and Time Spent Volunteering

Pearson correlations revealed significant relationships between perceived beneficial effects of voluntarism and hours of volunteer activity spent by the participant and spouse. Specifically, the more the participants emphasized the beneficial effects of voluntarism, the more hours both partners devoted to volunteer work (see Table 5). At the same time, the more hours the participants and their partners spent volunteering, the less they emphasized the need for spousal accommodation. No correlation was found between hours of volunteer work per week and perceived harmful effects of voluntarism.

d. Background Variables and Perceived Effects of Voluntarism on Marital Life

Pearson correlations were conducted in order to examine the relationship between background variables and the perceived impact of voluntarism on marital life. The following background variables were examined for participants and their spouses: Age, religiosity, and education level. No correlations were found between perceived effects of voluntarism and any of the background variables, except for a low-significant negative correlation between participant's level of religiosity and perceived harmful effects of voluntarism (r = -.20, p < .001). Consistent with these findings, a negative correlation was found between spouse's religiosity and perceived harmful effects of voluntarism (r = -. 18, p < .01). This result suggests that the more religious the participants and their spouses, the lower the tendency to emphasize the harmful effects of voluntarism. In addition, MANOVA was conducted in order to examine the impact of participant's and spouse's ethnicity (Israeli, Asian-African, European-American) on the perceived effects of voluntarism. These analyses revealed no significant effect for participant's ethnicity F(6, 1074) = .49, p > .05 or spouse's ethnicity F(6, 1074) = .04, p > .05 on any of the attitudes examined.

Discussion

It is important to gain insight into perceptions regarding the impact of voluntarism on marital relations--particularly in late adulthood, which is characterized by a decline in social relations that largely centered around the workplace. At this life stage, the spouse is usually the main confidant and primary source of social support (Anderson & McCulloch, 1990). Thus, emphasis on the contribution of voluntarism toward marital life may encourage such activity. By contrast, new social contacts established in the volunteer setting may arouse feelings of jealousy in the non-volunteer spouse and discourage the volunteer spouse from engaging in such activity. These arguments further underscore the importance of investigating attitudes regarding the impact of voluntarism on marital life.

One of the main conclusions of the current study is that regardless of gender or family type, the beneficial effects of voluntarism in late adulthood were highlighted most. At the same time, spousal accommodation was also prevalent, whereas perceived harmful effects were least prevalent for all family types.

Family type (by employment status) only affected one dimension of the attitudes examined, i.e., spousal accommodation. Specifically, this dimension was highlighted by two types of families: synchronous-preretired, and asynchronous (pre-retired participant / retired spouse) families. In families, high scores were obtained for this dimension (following beneficial effects of voluntarism). When one or both spouses are employed, it can be assumed that volunteer activity may upset the non-volunteer spouse and constitute a potential source of tension in the family. This would explain why members of these types of families emphasized the need for spousal accommodation. Regarding gender differences, men showed a greater tendency than women to emphasize the need for spousal accommodation. This tendency is consistent with earlier gender-specific patterns that prevail during the period of employment. Employed women typically bear the dual burden of work and household responsibilities, and are usually the ones that accommodate their work schedules to the demands of family life (Izraeli, 1994; Lewis, Izraeli, & Hootsman, 1992). This pattern may continue in late adulthood, when retired women who volunteer outside of the home are usually the ones to accommodate their activity to family demands. These gender differences were found to be consistent for all four types of families examined.

Gender differences were also found in both types of asynchronous families with respect to perceived harmful effects of voluntarism. Surprisingly, in these types of families women usually placed more emphasis than men on the harmful effects of volunteer activity. By contrast, no gender differences were found between the two types of synchronous families. In this connection, the study indicates that asynchronous families experience considerable marital tension because the differences in employment status orient partners toward divergent spheres of activity (home versus work) (Kulik, 2001). Thus, because women naturally emphasize harmonious relations (Gilligan, 1982), they are more likely to highlight the potential harmful effects of voluntarism in asynchronous families, out of fear that such activity will generate tension in the home. It should be mentioned that no relationships were found between background variables and perceived effects of voluntarism, except in the area of religiosity. In this case, it was found that the more religious the couple, the less they emphasized the harmful effects of voluntarism on marital relations, perhaps because religious families are influenced by the injunction to help the needy.

With respect to the relationship between actual volunteer activity and the perceived impact of voluntarism on marital relations, the findings support the research hypotheses. In congruent families (where both partners volunteer), the perceived benefits of voluntarism for the marital relationship ranked high, whereas perceived harmful effects ranked low. It should also be noted that in incongruent (volunteer participant / non-volunteer) families, spousal accommodation ranked relatively low. This suggests that people who decide to volunteer outside of the home and leave the spouse behind are not sensitive enough to the need to accommodate such activity to the partner's needs. In the same vein, it is worth mentioning that a relationship was found between the spouse's hours of volunteer activity and the perceived benefit of voluntarism for the marital relationship. By contrast, the number of hours devoted to volunteer work was inversely related to emphasis on spousal accommodation. It should be mentioned that the correlative nature of the study made it difficult to arrive at definitive conclusions and clear explanations of these findings. On the one hand, emphasis on the beneficial effects of voluntarism may encourage spouses to devote more hours to such activity. On the other, the time investment in itself may provide an incentive to develop attitudes that justify voluntarism. In order to clarify these issues, it would be worthwhile to conduct studies based on longitudinal designs.

Finally, some limitations of the study should be mentioned. Since the study focused specifically on elderly Jewish couples in Israel, future research might examine voluntarism among other sectors of the Israeli population such as Arabs and Druze. Such research would shed light on how different cultural norms affect the attitudes of people living in a given social context. In addition, it would be worthwhile to expand the research population and include people who have been retired for over six years in order to explore the impact of length of retirement on attitudes toward voluntarism.

Practical Recommendations

The research findings elicit several practical recommendations for third sector organizations that offer training and preparation programs for senior citizens who wish to volunteer. Because congruent-volunteering couples were most likely to express positive perceptions of volunteering, it would be desirable to encourage both partners to engage in some form of voluntarism. Altogether, the shared experience of volunteer activity may enrich marital relations at this stage of life. Moreover, the need for spousal accommodation was mentioned frequently--especially by members of synchronous-pre-retired families and asynchronous (pre-retired participant / retired spouse) families. In light of this finding, elderly couples might be encouraged to accommodate volunteer activity to the family context in order to prevent potential marital conflict. In this connection, special consideration should be given to "incongruent" families (where one spouse volunteers and the other does not), because these families appear to be less sensitive to the need for spousal accommodation to volunteering. Specifically, programs for volunteers might combine theoretical and practical courses, which could include a unit focusing on spousal accommodation of volunteer activity. This unit might cover topics such as time management or planning a joint schedule of activities. On the whole, proper planning and preparation of elderly couples for volunteer work may enrich their marital life and prevent spousal tensions, while also encouraging them to contribute to their communities and to needy populations.

Table 1
Factor Analysis of Variables Related to Attitudes toward
Voluntarism--Item Content and Loadings

Item Content                            Factor 1   Factor 2   Factor 3

1. Community volunteer work improves       .86        .07        .06
   marital life.

2. Both partners should engage in          .86        .09        .09
   volunteer activity.

3. Volunteer activity enriches             .66        .28        .13
   marital life.

4. When volunteer activity detracts        .01        .67        .04
   from marital life, the spouse that
   volunteers should give it up
   immediately.

5. When the volunteer activity of one      .04        .71        .33
   partner disturbs the other, it
   should be stopped.

6. When one spouse volunteers, it is       .05        .66        .88
   important to share these
   experiences and involve the
   partner.

7. The schedule of volunteer should        .01        .64        .25
   be adapted to marital life.

8. The success of the volunteering         .08        .47        .07
   spouse depends on cooperation from
   the partner.

9. Volunteer activity may make the         .08        .04        .67
   non-volunteering spouse jealous.

10. Volunteer activity may generate        .24        .03        .79
   marital tension.
Table 2
Perceived Impact of Volunteer Activity on Marital Life, by Family
Type (Employment Status) and Gender

                   S-R     S-PR    AS-PR/R     AS-R/PR    Total

Benefits
              M    3.92    3.83     3.73        3.89       3.87
             SD   (.59)   (.45)     (.62)       (.61)     (.55)

Accommodation
              M    3.62    3.80     3.78        3.69       3.72
             SD   (.67)   (.48)     (.44)       (.60)     (.57)

Harmful Effects
              M    2.32    2.26     2.46        2.47       2.36
             SD   (.77)   (.61)     (.58)       (.75)     (.72)

                                      F                       F
                                    Family        F         Inter-
                   Male   Female     Type      Gender       action

Benefits
              M    3.82    3.91     2.00        2.06        .70
             SD   (.52)   (.57)

Accommodation
              M    3.80    3.64     3.67 **     5.96 *      .74
             SD   (.53)   (.61)

Harmful Effects
              M    2.33    2.34     1.94        3.20       3.25 **
             SD   (.65)   (.74)

S-R = Synchronous Retired; S-PR = Synchronous Pre-retired;
AS-PR/R = Asynchronous-Pre-retired Participant/Retired Spouse;
AS-R/PR = Asynchronous-Retired Participant/Pre-retired Spouse

* p < .01

** p < .001
Table 3
Interaction between Gender and Family Type (by Employment Status):
Harmful Effects of Voluntarism

            S-R     S-PR     AS-PR/R     AS-R/PR
Men
     M     2.29     2.39       2.22       2.32
     SD    (.60)    (.73)     (.41)       (.70)

Women
     M     2.23     2.25       2.71       2.62
     SD    (.62)    (.81)     (.65)       (.80)

S-R = Synchronous Retired; S-PR = Synchronous Pre-retired; AS-PR/R
= Asynchronous-Pre-retired Participant/Retired Spouse; AS-R/PR =
Asynchronous-Retired Participant/Pre-retired Spouse

Table 4
Perceived Impact of Voluntarism on Marital Life, By Volunteer
Family Type (Congruent vs. Incongruent)

                  Family
Factor             Type     CV      C-NV   IC-V/NV   IC-NV/V   F Value

Benefits
                    M      4.05     3.83    3.86       3.75    6.33 **
                    SD     (.50)   (.57)    (.54)     (.51)

Accommodation
                    M      3.81     3.77    3.56       3.72    4.49 *
                    SD     (.49)   (.59)    (.63)     (.49)

Harmful effects
                    M      2.15     2.51    2.32       2.29    7.24 **
                    SD     (.62)   (.77)    (.75)     (.54)

CV-Congruent-both spouses volunteer; C-NV-Congruent-neither
spouse volunteers; IC-V/NV-Incongruent-Volunteer participant/
non-volunteer spouse; IC-NV/V-Non-volunteer participant/volunteer
spouse

* p < .05

** p < .01
Table 5
Correlations Between Hours of Volunteer Activity Per Week and
Perceived Impact of Voluntarism

                 HVA/            HVA/              HVA/
               Perceived       Spousal          Perceived
               Benefits     Accommodation    Harmful Effects

Participant     .23 **         -.18 *              .01
Spouse          .29 **         -.21 *              0.6

HAV--Hours of Volunteer Activity

* p < .01

** p < .001

References

Anderson, T. B., & McCulloch, J. B. (1990). Conjugal support: Factor structure among older husbands and wives. Paper presented at the Annual Meeting of the Gerontological Society of America, Boston MA.

Barnett, R. C. (1993). Multiple roles, gender and psychological distress. In L. Goldberg & S. Breznitz (Eds.), Handbook of stress: Theoretical and clinical aspects (pp. 427-448). The Free Press: New York.

Cohen-Mansfield, J. (1989). Employment and volunteering roles for the elderly: Characteristics, attributions and strategies. Journal of Leisure Research, 21, 214-227.

Fisher, L. R., & Mueller, D. P., Cooper, P. W., & Chase, R. A. (1989). Older Minnesotans: What do they need? How do they contribute?. St. Paul, MN: Amherst H. Wilder Foundation.

Fisher, L. R., & Mueller, D. P., & Cooper, P. W. (1991). Older volunteers: A discussion of the Minnesota Senior Study. The Gerontologist, 31, 183-194.

Fisher, L. R., & Schaffer, K. B. (1993). Older volunteers: A guide to research and practice. Newbury Park, CA: Sage.

Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University Press.

Hayghe, H. V. (1991). Volunteers in the US: Who donates time? Monthly Labor Review, 114, 7-23.

Herzog, A. R., & House, J. S. (1991). Productive activities and aging well. Generations. 15, 49-54.

Hulbert, J. R., & Chase, R. A. (1991). Retiree volunteers and the agencies they serve: A national survey. Report prepared for the National Retiree Volunteer Center. St. Paul, MN: Amherst H. Wilder Foundation.

Israel Central Bureau of Statistics (1998). Statistical Abstract of Israel (Tables 2.22, 22.1). Jerusalem, Israel.

Izraeli, D. N. (1994). Money masters: Spousal incomes and family work relations among physician couples in Israel. Sociological Quarterly, 35(1), 69-74.

Kulik, L. (1999). The impact of retirement on spousal relations in late adulthood: Comparative analysis of women. Research Report, School of Social Work, Bar Ilan University (in Hebrew).

Kulik, L. (2001). Marital relationships in late adulthood: Synchronous versus asynchronous couples. Journal of Aging and Human Development, 52(4), 323-336.

Lewis, S., Izraeli, D. N., & Hootsman, H. (Eds.) (1992). Dual earner families: International perspective. Newbury Park, CA: Sage Publications.

Szinovacz, M. (1996). Couples: Employment-retirement patterns and perceptions of marital quality. Research on Aging, 18(2), 243-268.

Szinovacz, M., Ekerdt, D. J., & Vinick, B. H. (1992). Families and retirement: Conceptual and methodological issues. In M. Szinovacz, D. J. Ekerdt, & B. H. Vinick (Eds.), Families and retirement (pp. 1-22). Newbury Park, CA: Sage.

Szinovacz, M., & Harpster, P. (1994). Couples: Employment-retirement status and the division of household tasks. Journal of Gerontology, Social Sciences, 49(3), S125-S136.

Liat Kulik Bar Ilan University School of Social Work