Gender ideologies, relative resources, and the division of housework in intimate relationships: a test of Hyman Rodman's theory of resources in cultural context **.
Sociologists have studied the division of housework and marital power in intimate relationships since the 1960s. Little has changed since Betty Friedan (1963) described gender inequality (i.e., the dissatisfaction and the problems of American women with housewifery) in her famous book, The Feminine Mystique. It is a well-documented finding in family research that women performed, and continue to perform, the lion's share of housework (Hochschild and Machung 1989; Lennon and Rosenfield 1994; Rubin 1976; Shelton 1992; Shelton and John 1996). A notable explanation of the division of domestic tasks is Robert O. Blood and Donald M. Wolfe's (1960) theory of relative resources. The basic assumption of this theory is that "the balance of power will be on the side of that partner who contributes the greater resources to the marriage" (Blood and Wolfe 1960:12). Since husbands typically are specialized in income generation, while wives work in part-time jobs or are homemakers, husbands have greater bargaining, which ena bles them to withdraw from unpleasant and monotonous housework (Nakhaie 1995). From this view, the division of labour in the home results from a rational (or at least a reasonable) decision. (1) While the findings from several empirical studies support the theory of relative resources (Ferree 1991; Kamo 1988; Rodman 1970), there is also contradicting evidence. It has been found, for instance, that husbands' authority does not increase linearly with education, occupational status, and income (Safilios-Rothschild 1967; Fox 1973). Also, Wives who earn more than their husbands, do far more housework than their spouse (Atkinson and Boles 1984). This still holds true when husbands are not employed (Brayfield 1992). The gender ideology perspective on the division of labour in the home can explain these findings: The division of housework does not result from the relative resources of the partners, but from how each partner identifies himself or herself with regard to marital, familial, and occupational roles that ar e traditionally linked to gender (Greenstein 1996; Kamo 1988; Sanchez 1994). Thus, traditional gender ideology that ascribes domestic labour to women results in wives performing the lion's share of housework, even if they earn more income than their husbands. Also, husbands may avoid housework even when they are economically dependent on their wives, because this provides an opportunity to express an inviolable identification with the male role (by refusing to adopt characteristics of the female role). Thus, they are "doing gender" despite their economic dependency (Haddad 1996).
Relative Resources in Cultural context--Hyman Rodman Revisited
In general, the findings indicate that both explanations of the division of labour in the home--resource theory and the gender ideology approach--are not mutually exclusive. Instead, it seems that the relationship between the relative resources of spouses and their marital power is mediated by cultural and subcultural norms concerning gender roles or--as one might prefer to say--gender ideology. As early as 1967, Hyman Rodman (1967, 1970) described the interaction between cultural norms and relative resources. Rodman argued that in traditional cultural settings characterized by patriarchy (such as India), the cultural norm ascribing all authority to the husband is so strong that the relative resources of spouses do not influence the marital power structure. In cultures with strong egalitarian norms concerning gender roles (such as Scandinavian countries) there should be no effect of resources. It is in cultural settings that fall between traditional and egalitarian groups where the relative resources of spous es impact the structure of marital power. In such cultures there are few strong norms concerning gender roles, thus, leaving room for the influence of relative resources. Rodman distinguishes two types of transitional cultural contexts: modified patriarchy and transitional state toward egalitarianism. In a modified patriarchy, traditional gender norms predominate, but male authority is decreases as level of social class increases. In the transitional state, resources play a decisive role in the absence of clear norms concerning gender-roles. It is in this state where there is a struggle for power between husbands and wives. Since the role of provider continues to be ascribe to men in the transitional state, and given that men in the lower strata have fewer opportunities than men in higher strata, Rodman expects an association between male authority and social stratum in transitional cultural contexts (in the narrow sense) that is opposite to that in modified patriarchy. In all, this means that the resource th eory is limited to specific cultural (or historical) settings in which the population is at odds with regard to gender roles, for instance, in transitional cultural contexts.
Rodman advanced his theory of resources in cultural context in order to investigate the empirical findings produced by studies that replicated the work of Blood and Wolfe (1960). As such, theory was designed to explain ex-post-facto. Until now, there are no comparative, empirical tests of Rodman's theory. This paper takes a closer look at the interaction between prevailing gender ideology and the relative resources of spouses when it comes to explaining the division of household tasks in cross-cultural contexts.
Objectives and Hypotheses
Blood and Wolfe's (1960) resource theory and Rodman's (1967, 1970) reformulation of the theory are tailored to explain marital power structure. But it is not clear whether the notion of marital power structure is meant to represent the process of bargaining, the result of a bargaining process or the availability of means to get one's way. Usually, it is measured by asking respondents about who makes the decisions in their relationship, for instance, about the organization of family income. It is difficult to construct a good measure of marital power structure. Making a decision for the couple or the family is not marital power as such, since decision-making is not identical to power in decision-making. Thus, the division of household tasks is chosen as the explanandum in this analysis because marital power is better conceptualized by what spouses are actually doing than in who is deciding what to do, especially when it seems reasonable to suggest that few people like to do household chores.
With the division of household tasks being the explanandum (or the dependent variable in the statistical analysis) the main objective of this study is to test whether there is variation between the cultural contexts that Rodman distinguishes (i.e., traditional, transitional or egalitarian contexts). If the division of household chores is simply a function of prevailing gender ideology, one would suggest that such tasks are almost exclusively dealt with by wives in traditional cultures. In egalitarian cultural contexts, chores would be equally divided between the spouses. But it is also possible that an egalitarian gender ideology does not result in equal sharing because arrangements often are made to fit the needs and wishes of both spouses (Hunt and Hunt 1982). Here, much depends on what actually is regarded as egalitarian, homogamy or complementarity. Nevertheless, it is expected that couples in egalitarian cultural contexts share in household tasks more equally than couples in transitional or traditional cultural frameworks. Greater variation in the division of housework is also expected in transitional cultural environments than in traditional or egalitarian cultures because of the greater flexibility of norms concerning gender roles in transitional contexts. Since Rodman suggested that transitional cultures (with regard to gender ideology) might be modified patriarchies, it is expected that the division of household tasks between husbands and wives is more "equal" in the higher social strata of transitional contexts than in lower strata.
If the division of labour in the home is a function of the spouses' relative resources alone, then one would predict no systematic variation in the allotment of tasks in different cultural contexts. Gainful employment of the husband, wife or both, and the relative contribution each makes to the family income should then, explain the division of chores. According to Rodman, in transitional contexts there are no clear norms concerning gender-roles, thus, there is the most room for the spouses' relative resources to have an impact on the division of tasks. Therefore, it is suggested that the effect of relative resources is stronger in transitional cultures than in traditional or egalitarian ones. there To summarize, the main questions to be answered in this paper are: Is systematic variation in the division of household tasks according to the type of cultural context? Is there a more "equal" sharing in household tasks between husbands and wives in the higher social strata of transitional cultures than in lower strata? How well do the relative resources of spouses explain the division of household tasks, and how do norms concerning gender roles and relative resources interact?
Analysis for this study is based on data from the 1994 International Social Survey Programme (ISSP) concerning "Family and Changing Gender Roles." It covers 24 countries on different continents and 33,590 individuals. The number of respondents in each country varies between 647 in Northern Ireland to 2,494 in Spain. Respondents are aged between 15 and 96. Interviewers asked 15,230 men and 18,323 women (and 37 respondents for which the information about gender is lacking) to give their opinion on items concerning gender roles, marriage, divorce, premarital sexual relationships, the perceived effects of mothers being gainfully employed on their children, and the role that children play in marriage and divorce, among other things. Respondents were also asked to indicate their marital status, household composition, division of household tasks, the management of the family income, and various demographic characteristics such as education, occupation, subjective social class, and income. There are only a few questi ons about the respondents' partners, but among them is the question about his/her employment. Thus, the 1994 ISSP is suitable for testing Rodman's theory of resources in various cultural contexts (at least approximately). The ISSP provides information on gender roles, some measures of resources of both spouses, and the division of household tasks in 24 countries.
The Affiliation of Countries to Specific Cultural Contexts
The main problem with empirically testing Rodman's theory is how to decide what are cultural contexts in which traditional or egalitarian norms concerning gender roles prevail, and what are cultural contexts that can be described as being in transition from traditional to egalitarian norms. In order not to have to assign countries to a specific cultural setting on the basis of some arbitrary criterion, this problem is solved empirically. Agreement or disagreement with the item: "A man's job is to earn money; a woman's job is to look after the home and family," is taken to reflect gender ideology. At the national level, the 24 countries can be divided into three groups according to the prevailing ideology (see Table 1).
The first group of countries comprises the egalitarian context: Canada, East Germany, Sweden, Norway, The Netherlands, United States, New Zealand, and Israel. These nations have the highest means indicating (strong) disagreement with traditional norms concerning gender roles. The second group: Great Britain, Australia, Northern Ireland, Ireland, Italy, Spain, West Germany, and Japan, represent countries in which gender ideology is in transition. The third group (Slovenia, Austria, the Czech Republic, Hungary, Poland, Bulgaria, Russia, and the Philippines) represents traditional cultures. Hence, the assignment procedure results in three groups of eight countries. By comparing the standard deviations shown in Table 1, it becomes clear that concord is greatest in the Philippines, while dissension is greatest in Japan, Austria, and Bulgaria. The countries in which concord is greatest (s.d. [less than or equal] 1.10) are also those in which respondents disagree most with traditional gender roles (Canada, East Ger many, Sweden, Norway, The Netherlands) or those in which respondents agree most with traditional roles (Philippines, Poland, Russia). There is a tendency for the standard deviation to be greater when the means are closer to 3, indicating that respondents are undecided, that is, neither agree nor disagree. One might say then, that nations in the central part of Table 1 can be regarded as being in role-transition in two respects: More respondents in these countries are undecided (neither agree nor disagree) and dissension is greater than in the other countries.
The Division of Household Tasks
The division of household task is the dependent variable in the analysis. It is an index built on the basis of five items indicating who does (1) the washing and ironing, (2) small repairs around the house, (3) looks after sick family members, (4) the shopping for groceries, and (5) who decides what to have for dinner. Answers can be "always the woman," "usually the woman," "about equal or both together," "usually the man," "always the man" or "is done by a third person." For the creation of the index the last category was excluded, since the index is meant to represent the division of household tasks between spouses. The index ranges from 5 (indicating that wives always do all the work) and 25 (indicating that the husband does all the work). (2)
In the ISSP, respondents are asked to ascribe themselves to a social class: lower class, working class, upper working or lower-middle class, middle class, upper-middle class, or upper class. This variable was recoded into three categories designating (1) lower or working class, (2) upper working/lower middle or middle class, and (3) upper middle or upper class.
The Relative Resources of Spouses
Because resource theory is about relative resources (according to Blood and Wolfe and Rodman), it does not seem appropriate to use variables that represent individual characteristics of the respondent when the corresponding data for the spouse is not available. Therefore, the explanatory variables in this analysis are based on ratios, indicating whether it is the respondent or his/her spouse who has the most resources. The first measure of relative resources represents the couple's occupational arrangement: a combination of two variables that indicate the current employment status of the respondent and his/her spouse. This synthetic variable has eight categories: (1) both spouses are working full-time or both spouses are working part-time, (2) the husband is working full-time or part-time, while the wife is also working, but less than her husband, (3) (3) the husband is working fulltime or part-time, while the wife is a housewife or unemployed, (4) the husband is working full-time or part-time, while the wife has not finished her education yet, is in training, retired, disabled or not in the labour force for some other reason, (5) the wife is working full-time or part-time, while the husband is working, but less than his wife, (6) the wife is working full-time or part-time, while the husband is a houseman or unemployed, (7) the wife is working full-time or part-time, while the husband has not finished his education yet, is in training, retired, disabled or not in the labour force for some other reason, and (8) both spouses have not finished their education yet, are retired, disabled or not in the labour force for some other reason.
The second measure of relative resources is "who earns more money, the husband or the wife." This variable has three categories: (1) husband earns more than the wife, (2) they both earn about the same, and (3) the wife earns more than the husband.
The third measure indicates how income is organized in the relationship. This variable has three categories: (1) husband manages all the money, (2) money is pooled and each spouse takes out what is needed or the money is kept separate, and (3) the wife manages all the money.
Apart from descriptive statistical analysis, several one-factorial and multi-factorial analyses of variance are computed in order to find statistically significant differences between the means of various subgroups with regard to the division of household tasks, and to estimate the variation explained by the independent variables. In other words, it is possible to find out about the differential impact of factor variables on the dependent variable. It is also possible to check for the main effects of the factors or explanatory variables, as well as for interaction-effects.
Included in the analysis are only those respondents who are married or live as married (N = 21,589 couples). They are referred to as "spouses" or "husbands and wives," though, strictly speaking, this term is suitable for married couples only. In the analyses of variance, cases with missing values are excluded pair-wise. Thus, the number of cases included in the analyses vanes.
Does husband-wife division of tasks vary between the three cultural contexts? On the horizontal axis of Figure 1, nations are arranged in groups representing the different cultural contexts. Egalitarian countries are shown on the left side, traditional ones are on the right, and transitional nations are found in the middle. The axis representing the index for the division of household tasks ranges from 9 to 14 and reproduces a section of the total scale ranging from 5 to 25. Because the national means are not lower than 9 or higher than 14, the categories below 9 and above 14 have been omitted in Figure 1.
Figure 1 indicates that there is systematic variation in the division of household tasks according to the type of cultural context, but these variations are only partly in the expected direction. The share of chores done by husbands is greatest in the eight countries characterized by an egalitarian gender ideology (i.e., Canada, East Germany, Sweden, Norway, The Netherlands, United States, New Zealand, and Israel). It becomes clear from Figure 1, that even in cultural contexts where egalitarian ideology prevails, household tasks are not equally shared (a mean of 15 on the index indicates that housework is shared equally). In countries where traditional ideology prevails, it is not the case that wives are doing all or the lion's share of the housework (as would be indicated by a mean between 5 and 10), but it is consistent with the expectation that the means of these countries are lower than that of countries representing the egalitarian context. Contrary to our expectation, wives in some transitional countri es do more housework than wives in traditional contexts. In line with our expectations, Figure 1 clearly shows that in the transitional context, variation between the countries is greatest with Japan having a mean of 9.45 and West Germany having a mean of 12.24.
Analysis of variance including pairwise, post-hoc tests shows that the observed differences between the three cultural contexts with regard to the division of housework is statistically significant (p [less than or equal to] .05). This indicates that there is a relationship between the prevailing type of gender ideology and the division of housework. The mean difference with regard to the division of household tasks in the egalitarian context as compared to the transitional context is 1.56. In the egalitarian context as compared to the traditional context, the mean difference is 0.91, and in the traditional context as compared to the transitional context, it is 0.65.
The next test determines whether there are differences is sharing chores between social strata, as Rodman suggested would be the case for the transitional context. Figure 2 illustrates the association between the share of housework in different cultural contexts and the ascribed social strata.
In the egalitarian context, there is no big difference between the social strata with regard to the sharing in housework, though in the upper social stratum, husbands do slightly more housework (M = 13.3) than in the lower (M = 13.1) and middle strata (M = 12.9). Within the egalitarian context, regardless of social strata, housework is not shared equally (i.e., means are lower than 15). Husbands in the traditional context do slightly more housework when they ascribe themselves or are ascribed by their wives to the middle class (M = 12.1) as compared to the lower social stratum (M = 11.9), but in the upper social stratum (M = 11.8) they do slightly less than in the lower one. Again, there are no big differences between the social classes. In the transitional context, husbands in the middle class (M = 11.4) do less than husbands in the lower class (M = 11.8), and husbands in the upper class do the least housework (M = 10.8). This is contrary to what Rodman suggested with regard to modified patriarchies. In the transitional context, the biggest difference is the one between husbands in the lower class and husbands in the upper class (amounting to one point on the scale). In the other cultural contexts, the biggest differences amount to 0.4 or 0.3 points on the scale. While upper-class husbands in the egalitarian context are those who perform the greatest share of household tasks, upper-class husbands in transitional enivronments are the ones who do the least housework.
In the next step of the analysis, the relative resources of spouses in the three contexts are considered. Table 2 presents the descriptive results from the bivariate analysis. Opposed to what one might expect, it is not in the traditional context that wives have the least resources; the transitional group has less. We see that husbands earn more than wives more often and wives earn more than husbands less often in the transitional context than in the other contexts. Table 2 also shows that money is managed jointly or kept separate less often in the transitional context, while the husband manages all the money more often in the transitional context than in the other contexts. Finally, an occupational arrangement in which the husband is working full-time or part-time while the wife is unemployed or a housewife, is found much more often in the transitional context than in the other contexts: 28.8 percent of all the couples in the transitional group have such an occupational arrangement compared to only 15.3 per cent of the couples in the egalitarian cluster and 11.7 percent of the couples in the traditional context. At the same time, an occupational arrangement in which the husband and the wife are both working full-time or part-time is less often chosen by couples in the transitional context (24.6 percent) than by couples in the other groups (37.3 percent in the egalitarian context and 40.6 percent in the traditional culture). The overall impression from Table 2 is that the practice of the couples in the transitional category is what one would have expected according to a gender ideology that ascribes the role of the provider to the man while the woman is the housekeeper, for instance, in the traditional context. Obviously, the expression of gender ideology and the relevant practices are divergent.
Table 3 shows the results of a bivariate analysis of the associations between the division of tasks and the cultural context on the one hand, and between the division of tasks and each of the variables indicating the relative resources of spouses on the other hand. Cultural context has the strongest association with the division of chores, followed by the organization of income and the occupational arrangement. Which spouse earns more money is only weakly associated with the division of household tasks.
It was hypothesized that the effect of variables representing the spouses' relative resources is strongest in the transitional culture because of the greater flexibility of gender ideology in this context as compared to the egalitarian or traditional groups. Table 4 shows the differential impact that relative resources have on the allocation of chores in each of the three cultures.
Two of the three indicators of relative resources confirm the hypothesis that the impact of who earns more money and who organizes the income in the relationship on the division of tasks is strongest in the transitional context. Concerning the occupational arrangement, the expectation is refuted: the impact of occupational arrangement on sharing chores is strongest in the egalitarian context. Further, from Table 4 it is clear that all three variables of relative resources have the smallest impact on the allocation of chores in the traditional context, which confirms our expectations.
The bivariate analysis shows that cultural context and relative resources have an impact on the division of tasks. The question to be answered now is whether the effects prove to be statistically significant. In order to test this, a multi-factorial analysis of variance was conducted using context and resources as independent variables. Since information on who earns more money is only collected when both partners are gainfully employed, this variable is excluded from the mufti-factorial analysis in order to minimize the loss of cases. To keep the multivariate analysis simple, "occupational arrangement" was recoded into a variable with three categories: (1) both partners are working full-time or part-time, or they are both not in the labour force; (2) the husband is working more than the wife (with "working more" meaning that he is working for more hours per week in gainful employment); and (3) the wife is working more than the husband. The age of the respondent is included into the analysis as a covariate though it is a non-relational variable, because for most countries included in the ISSP it is plausible to expect that age differences between partners are not very large. The hypothesis is that younger couples will share housework more equally than older ones. In addition to the main effects of the factors and the covariate, interactions are allowed for.
Figure 3 shows the factors and interaction effects in the model of division of tasks. We see that 24.2 percent of the variance is explained by factors and interactions that are not statistically significant in the model. Among the statistically significant factors in the model, the organization of income explains 19.4 percent of the variance and the cultural context explains 16.4 percent. The age of the respondent explains 24.2 percent of the variance, thus, having the highest explanatory power of all the main effects in the model. From the interactions, only the 2-way interaction between the occupational arrangement and the cultural context (10.3 percent of the variance explained) and the 3-way interaction between the occupational arrangement, the cultural context, and age of respondent (5.5 percent of the variance explained) are significant. Overall, the model's [R.sup.2] is only .119, indicating that the variables included in the model explain the allocation of choresonly poorly.
Figures 4 and 5 show how much of the variance of the division of housework is explained by factors and interaction effects in the egalitarian and transitional context. In the transitional context, the variables included in the model explain the division of housework better ([R.sup.2] = .123) than in the egalitarian context ([R.sup.2] = .065), though the explanatory power of the model is poor in both cultural contexts.
In the traditional context, none of the factor variables and none of the interactions are statistically significant. Accordingly, the model of the traditional category fits the data very poorly ([R.sup.2] = .023).
Thus, the variables representing relative resources have most explanatory power in the transitional context and no explanatory power in the traditional environment, which supports our initial prediction. Moreover, it is striking that in the egalitarian context (Figure 4), all statistically significant effects are main effects, while in the transitional context (Figure 5), in addition to the main effects of occupational arrangement and organization of income, all 2-way interaction effects and the 3-way interaction effect are significant. This indicates that in the transitional context, relative resources influences task-division only indirectly via specific constellations.
This paper began by questioning whether Rodman's (1967, 1970) explanation for differential power in intimate relationships is a culture-specific theory in the sense that it applies only to Western cultures (i.e., North America and most parts of Europe). According to Rodman, it is in these cultures where norms concerning gender roles are flexible or in transition from traditional to egalitarian values. Though the findings from 24 nations do not support the view that relative resources are only effective in Western cultures or transitional cultural contexts, the results show that they actually have a greater impact on the division of tasks in transitional contexts than in the egalitarian or traditional settings (see Table 4 and Figures 4 and 5). Moreover, interaction effects between the different measures of relative resources and the distribution of tasks are only significant in the transitional context. It can be argued that perhaps a bargaining process in which the lack of resources is decisive for the asses sment of other resources (or the lack of other resources). In all, the effects of different kinds of resources and different constellations of resources are much less patterned or predictable in the transitional context than in the egalitarian or the traditional settings. In general, the data supports Rodman's resource theory in cultural context. The analysis did not substantiate Rodman's notion of modified patriarchies: in the transitional context, upper-class husbands do less housework than lower-class husbands and middle-class husbands (see Figure 2). Of course, this does not exclude the possibility that there might be specific transitional cultural environments that support Rodman's notion of a modified patriarchy.
Apart from the findings that are of direct significance to Rodman's theory, the analysis produced some striking results that are worth being discussed here. First, couples in the transitional context seem to follow the patterns (i.e., with regard to earning money, organization of income, and occupational arrangement) that are associated most with the traditional model--in the sense that the man is the provider while the woman is the housewife (see Table 2). Also, the arrangements found in the traditional context cannot be understood against the backdrop of the prevailing gender ideology. It happens to be the case that in the traditional context wives are likely to earn more than the husbands and that wives work full-time or part-time, while the husbands are unemployed or a houseman. Of course, this does not conform to the view that a man should be the provider and the woman the caregiver. However, in the traditional context, the overall economic situation simply may not allow for an adjustment of practices t o the prevailing gender ideology. This line of argument suggests that the overall political or economic situation of a country is relevant to understanding living arrangements, gender ideology, and relative resources of spouses. In other words, factors outside of individual preferences can be significant.
Furthermore, it became clear from this analysis that even in the egalitarian context, there is unequal sharing of labour in the home (and outside the home) between husbands and wives (see Figure 1). As was mentioned before, it is not clear whether an egalitarian gender ideology implies that men and women do the same things (inside and outside the home), or that men and women are free to do the things they know best or want most.
As the status of women continues to improve, household arrangements will become more egalitarian. As shown in Figure 4, this argument is supported by the finding that in the egalitarian context, age and occupational arrangement are the most significant variables (see Figure 4).
It is clear that the debate about what determines the differential power of spouses, gender ideology or relative resources must reckon with interaction effects. Moreover, it is not true that different kinds of resources necessarily accompany each other (see Table 2). For example, in all cultural contexts, wives are more likely than husbands to manage the money, but this does not mean that wives are more likely than husbands to earn more money or work more. Here, we also have to expect interaction effects.
In connection with the organization of income, resource theory fails because it is not clear whether martial power stems from having resources or from allocating resources. For instance, the organization of income can be regarded as a specific qualification needed for housekeeping, which in turn suggests greater involvement in housework. In addition, in some situations the organization of income might reflect the sole decision of one spouse.
It should also be noted that the resources considered in this analysis could not explain the division of housework as shown by the poor fit of the models. Thus, further research is required to account for the division of household tasks. While this study is a specific test for Rodman's theory of resources in cultural context, it is certainly plausible to assume that there are variables that would have a significant effect if they were introduced into the multivariate model, for example, the number and age of children.
In testing Rodman's theory, this paper provides an empirically based measure of the three cultural constructs. Though this procedure might draw criticism, it is less arbitrary than previous attempts.
Finally, the measures of relative resources are drawn from Western perspectives and exclude such variables as the age-difference between the spouses. This raises the question of what resources mean in specific cultures. Since our test of resource theory is inconclusive, it is difficult to determine whether the findings simply refute resource theory or whether the specification of the resources is flawed. Future research needs to explore these methodological and theoretical issues in order to further our understanding of the impact of relative resources on differential power in intimate relationships.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Table 1 Gender Role Orientations Across 24 Nations (Ranked Means) * Extreme Category (1) Country Mean Std.Dev. Cases n % Canada 4.02 1.03 1426 40 2.8 East-Germany 3.96 1.01 1079 32 2.9 Sweden 3.91 1.06 1244 31 2.4 Norway 3.82 1.09 2062 63 3.0 Netherlands 3.61 1.10 1946 87 4.4 USA 3.54 1.15 1414 76 5.3 New Zealand 3.53 1.14 1020 40 3.8 Israel 3.51 1.25 1271 110 8.5 Great Britain 3.44 1.15 960 50 5.1 Australia 3.44 1.19 1767 91 5.1 Northern Ireland 3.33 1.16 633 37 5.7 Ireland 3.28 1.29 920 88 9.4 Italy 3.26 1.28 1015 105 10.3 Spain 3.21 1.25 2426 216 8.7 West-Germany 3.15 1.32 2263 290 12.5 Japan 3.07 1.49 1285 261 20.0 Slovenia 3.02 1.21 1017 111 10.8 Austria 2.80 1.45 953 241 24.7 Czech Republic 2.55 1.22 1015 236 23.0 Hungary 2.31 1.28 1494 568 37.9 Poland 2.30 1.09 1554 386 24.2 Bulgaria 2.29 1.38 1082 431 38.3 Russia 2.20 1.10 1921 590 29.5 Philippines 2.16 0.81 1200 169 14.1 Total 3.21 1.32 32,967 4,349 12.9 Extreme Category (5) Country n % Canada 541 37.6 East-Germany 338 30.8 Sweden 439 34.5 Norway 635 30.4 Netherlands 392 19.9 USA 296 20.5 New Zealand 215 20.5 Israel 294 22.8 Great Britain 167 17.0 Australia 375 21.1 Northern Ireland 93 14.4 Ireland 174 18.6 Italy 196 19.3 Spain 341 13.7 West-Germany 402 17.3 Japan 353 27.0 Slovenia 103 10.0 Austria 166 17.0 Czech Republic 70 6.8 Hungary 105 7.0 Poland 35 2.2 Bulgaria 123 10.9 Russia 32 1.6 Philippines 12 1.0 Total 5,897 17.6 * Gender role orientation was measured with the item, "A man's job is to earn money; a woman's job is to look after the home and family." The answer categories were: 1 = strongly agree, 2 = agree, 3 = neither agree nor disagree, 4 = disagree, 5 = strongly disagree. Table 2 The Distribution of Relative Resources in Different Cultural Contexts (Column Percentages) Cultural Context Egal- Trans- Trad- itarian itional itional Occupational arrangement Both are working full-time or part-time 37.3 24.6 40.6 Husband is working full- or part-time; Wife is working less than husband 16.1 16.0 5.8 Husband is working full- or part-time; Wife is unemployed or housewife 15.3 28.8 11.7 Husband is working full- or part-time; Wife is not working for other reasons (education, retired, disabled) 4.9 4.0 7.8 Wife is working full- or part-time: Husband is working less than wife 1.3 0.9 1.7 Wife is working full- or part-time; Husband is unemployed or houseman 1.9 1.2 2.4 Wife is working full- or part-time; Husband is not working for other reasons (education, retired, disabled) 5.0 2.8 4.7 Both are not in the labour-force (education, retired, disabled) 18.2 21.7 25.4 Who earns more money? Husband 75.9 79.3 68.9 Both the same 13.9 14.3 18.7 Wife 10.3 6.4 12.4 Organization of income in the relationship Husbands manages all the money 5.9 11.6 6.6 Money is managed jointly and each spouse takes out what s/he needs or money is kept separate 86.7 65.7 70.1 Wife manages all the money 7.4 22.6 23.2 Table 3 The Impact of Cultural Context and Spouses' Relative Resources on the Division of Household Tasks Dependent Variable: Division of Household Tasks (index) Eta Number of Cases Cultural context .227 18,925 Occupational arrangement .178 14,884 Who earns more money? .126 12,733 Organization of the .213 18,757 income Table 4 The Differential Impact of Relative Resources Variables on the Division of Household Tasks in Different Cultural Contexts (eta) Dependent Variable: Division of Household Tasks (index) Egalitarian Transitional Traditional Occupational arrangement .222 .202 .070 (n=5271) (n=4745) (n=4868) Who earns more money? .139 .174 .086 (n=4402) (n=4016) (n=4315) Organization of the .111 .277 .110 income in the (n=6349) (n=6331) (n=6077) relationship Figure 3 Variance of Division of Household Tasks Explained by Factor Variables and Interaction Effects Main effects: Cultural context 16,4% Organization of income 19,4% Age 24,2% 2-way interaction effects: Occupational arrangement with 10,3% cultural context 3-way interaction effects: Occupational arrangement with 5,5% cultural context with age Not significant 24,2% Note: Table made from pie chart Figure 4 Variance of Division of Household Tasks Explained by Factor Variables and Interaction Effects in the Egalitarian Context Main effects: Occupational arrangement 23,8% Organization of income 18,4% Age 37,5% Not significant 20,3% Note: Table made from pie chart Figure 5 Variance of Division of Household Tasks Explained by Factor Variables and Interaction Effects in the Transitional Context Main effects: Occupational arrangement 8,2% Organization of income 25,9% 2-way interaction effects: Occupational arrangement with organization of income 10,3% Occupational arrangement with age 30,6% Organization of income with age 14,3% 3-way interaction effects: Occupational arrangement with organization of income with age 10,6% Note: Table made from pie chart.
** I would like to thank Helena Flam for her many helpful editions and comments on an earlier draft.
(1.) Similarly, Gary Becker suggested that women have biologically-based comparative advantages over men in the household sector and that "households with only men or only women are less efficient because they are unable to profit from the sexual difference in comparative advantages" (1993:38-39). Again, the genderspecific division of labour is regarded as being rational for the simple reason that gains from (increased) specialization are greater than the gains from involvement in the same activity.
(2.) Note that the index comprises one item that, unfortunately, does not measure an activity, as the other items do, but a decision (i.e., item 5 "who is deciding what to have for dinner").
(3.) If the husband is working full-time, the wife is working part-time or she is a helping family member. If the husband is working part-time, the wife is working for a few hours per week or she is a helping family member.
ATKINSON, M P. and J. Boles.
1984. "WASP (Wives as Senior Partners)." Journal of Marriage and the Family 46:861-870.
1993. A Treatise on the Family. Cambridge: Harvard University Press.
BLOOD, R.O. and D.M. Wolfe.
1960. Husbands and Wives. The Dynamics of Married Living. New York: Free Press.
1992. "Employment Resources and Housework in Canada." Journal of Marriage and the Family 54:19-30.
1980. "Conflict Resolution in Marriage." Journal of Family Issues 1:397-421.
1991. "The Gender Division of Labor in Two-Earner Marriages: Dimensions of Variability and Change." Journal of Family Issues 12:158-180.
1973. "Another Look at the Comparative Resources Model: Assessing the Balance of Power in Turkish Marriages." Journal of Marriage and the Family 35:718-730.
1963. The Feminine Mystique. New York: Dell Publishing.
1996 "Husbands' Participation in Domestic Work: Interactive Effects of Husbands' and Wives' Gender Ideologies." Journal of Marriage and the Family 58:585-595.
1996. The Sexual Division of Household Labour: Pragmatic Strategies or Patriarchal Dynamics? An Analysis of Two Case Studies. North York: York University.
HOCHSCHILD, A.R. and A. Machung.
1989. The Second Shift: Working Parents and the Revolution at Home. New York: Viking/Penguin.
HUNT, J.G. and L.L. Hunt.
1982. "The Dualities of Careers and Families: New Integrations or Polarizations?" Social Problems 29:499-510.
1988. "Determinants of Household Division of Labor: Resources, Power, and Ideology." Journal of Family Issues 9:177-200.
LENNON, M.C. and S. Rosenfield.
1994. "Relative Fairness and the Division of Housework: The Importance of Options." American Journal of Sociology 100:506-531.
1995. "Housework in Canada: The National Picture." Journal of Comparative Family Studies 26(3):409-425.
1967. "Marital Power in France, Greece, Yugoslavia, and the United States: A Cross-National Discussion." Journal of Marriage and the Family 29:320-324.
1970. "Eheliche Macht und der Austausch von Ressourcen im kulturellen Kontext." Pp. 121-143 in Soziologie der Familie, edited by G Luschen and E. Lupri. Opladen: Westdeutscher Verlag.
1976. Worlds of Pain. New York: Basic Books.
1967. "A Comparison of Power Structure and Marital Satisfaction in Urban Greek and French Families." Journal of Marriage and the Family 29:345-352.
1994. "Gender, Labor Allocations, and the Psychology of Entitlement in the Home." Social Forces 73:533-553.
1992. Women, Men, and Time: Gender Differences in Paid Work, Housework, and Leisure. Westport: Greenwood Press.
SHELTON, B.A. and D. John.
1996. "The Division of Household Labor." Annual Review of Sociology 22:299-322.
Heike Diefenbach *
* Department of Sociology, Leipzig University, Burgstrasse 21, D- 04109 Leipzig, Germany.
|Printer friendly Cite/link Email Feedback|
|Publication:||International Journal of Comparative Sociology|
|Article Type:||Statistical Data Included|
|Date:||Mar 1, 2002|
|Previous Article:||The logic of nazification: the case of the International Criminal Police Commission ("Interpol") **.|
|Next Article:||The globalization of environmental protection: the case of Environmental Impact Assessment (1).|