Introduction

Religious traditions have long been identified as one of the major social forces promoting gender hierarchy and traditional gender roles around the world (e.g., early critiques by Daly 1968 and Ruether 1974; and global cross-cultural studies by Seguino 2011, and Stover and Hope 1984). Both Christian and Jewish religiosity have been found to be correlated with greater emphasis on values of conformity and tradition in general (e.g. Schwartz and Huismans 1995, in Spain, NL, Greece and Israel), and with traditional attitudes toward gender in particular (e.g., Fitzpatrick Bettencourt et al. 2011 (US); Pearce and Thornton 2007 (US); Seguino 2011, cross-culturally). However, different religious orientations vary in the specific gender ideologies they inculcate, and religion can affect attitudes toward women in complex ways (Hunsberger and Jackson 2005). Moreover, because traditional gender ideologies involve considerable ambivalence (Glick and Fiske 1996, 2001), the differential relationships between religiosity and various components of sexist attitudes may substantially vary across religions.

Very few studies have been conducted that specifically address the relationships between religiosity and ambivalent sexist attitudes. Most of these studies have examined Christian denominations (Burn and Busso 2005 (U.S.); Glick et al. 2002a (Spain); Maltby et al. 2010 (U.S.)) and focused solely on attitudes toward women (Burn and Busso 2005 (U.S.); Maltby et al. 2010 (U.S.); Tasdemir and Sakalli-Ugurlu 2010 (Turkey)).

To further shed light on the role of religiosity in ambivalent sexism, the present study explored the relationships between religiosity and sexist attitudes within the context of Jewish tradition in Israel. The hypotheses were derived from ambivalent sexism theory (Glick and Fiske 1996, 2001) and were based on Judaism’s views of gender relations. To test the hypotheses, ambivalent sexist beliefs and levels of Jewish religiosity were measured in a convenience community sample of Jews in Israel. Controlling for socio-demographic variables, this study further examined sexist attitudes toward men as well as toward women. Revealing the nature of relationships between Jewish religiosity and sexist attitudes may uncover one of the major sources of traditional gender beliefs in Israel in particular, and further establish the role of religiosity in promoting sexist ideologies and gender inequalities in general.

Ambivalent Sexism Theory

Ambivalent sexism theory (Glick and Fiske 1996, 1999, 2001) posits that the basic structure of gender relationships is characterized by the coexistence of power differences and intimate interdependence. Because of this coexistence, attitudes toward men and women encompass considerable ambivalence, consisting of both hostile and benevolent components (Glick and Fiske 1996, 2001).

Glick and Fiske (1996, 2001) suggested three domains of male–female relations that underpin the content of gender attitudes: paternalism—the ideological justification of male dominance; gender differentiation—the belief that men and women differ fundamentally or inhabit separate spheres; and heterosexual intimacy—the norm of having a heterosexual romantic relationship. All three domains give rise to both hostile and benevolent attitudes.

Hostile sexism (HS) represents antipathy and resentment toward women who are viewed as challenging male power or rejecting conventional gender roles. In contrast, benevolent sexism (BS) is a subjectively positive but patronizing attitude, which idealizes women in traditional roles and views them as weak beings who need to be protected and provided for by men (Glick and Fiske 1996, 1997). Both forms of sexism support a gender hierarchy that limits and disadvantages women who are seen as weaker and less competent.

Similar ambivalence also characterizes attitudes toward men (Glick and Fiske 1999, 2001). Hostility toward men (HM) reflects resentment of paternalism and men’s power and aggressiveness. Although reflecting antagonism to men’s higher status, these hostile attitudes characterize men as inherently powerful and aggressive, thus portraying male dominance as natural and inevitable (Glick et al. 2004). In contrast, benevolence toward men (BM) acknowledges and admires men’s roles of protectors and providers. Hostility and benevolence toward men are complementary beliefs that characterize men as powerful and arrogant while admiring their traditional roles.

Studies in diverse cultural contexts have shown that hostile and benevolent attitudes toward women are positively correlated with each other (Glick et al. 2000; Glick et al. 2004) as are hostile and benevolent attitudes toward men (Glick et al. 2004). Both types of attitudes were found to have negative associations with indicators of gender equality across 16 nations (Glick et al. 2004).

Ambivalent Sexism and Religiosity

A wealth of research has been conducted on ambivalent sexism and its implications for domains as varied as preferences for romantic partners (Chen et al. 2009 (U.S.); Eastwick et al. 2006 (Globally); Lee et al. 2010 (U.S.); Travaglia et al. 2009 (New Zealand)), endorsement of beauty standards (Forbes et al. 2007 (U.S.)), attitudes toward wife beating (Glick, Sakalli-Ugurlu et al. 2002 (Turkey, Brazil)) and breastfeeding (Forbes et al. 2003 (U.S.)). However, very few studies have explored the role of religiosity in shaping ambivalent sexism, and the empirical evidence regarding the effects of religiosity on attitudes toward men is scant.

Three studies have explored the effects of Christian religiosity on ambivalent sexist attitudes. Burn and Busso (2005) examined the relationships between religiosity and attitudes toward women in a sample of Christian American students. They found that religiosity and scriptural literalism were positively related to benevolent sexism. An additional study on Christian American students found that men’s religiosity was related to the protective paternalism component of benevolent sexism (Maltby et al. 2010). No relationships were found in this study between women’s religiosity and sexism. Finally, in a study of a Catholic Spanish sample, Glick et al. (Glick 2002a) found that religiosity was positively related to benevolent attitudes toward both men and women. In all three studies, Christian religiosity was a significant predictor of benevolent sexism but was unrelated to hostile attitudes.

Somewhat different findings were obtained in a study on Muslim Turkish university students (Tasdemir and Sakalli-Ugurlu 2010). In addition to positive relationships between religiosity and benevolent sexism for both men and women, this study documented positive relationships between men’s religiosity and hostile sexism. This finding presumably reflects the cultural context in which hierarchy and dominance are highly valued (Tasdemir and Sakalli-Ugurlu 2010).

As evident from these findings and in line with scholars’ suggestions (Glick, Lameiras, and Castro 2002; Maltby et al. 2010), the effects of religiosity on ambivalent sexist attitudes may vary depending on the specific religious context and the beliefs it promotes. The goal of the present study is to examine the influence of religiosity on ambivalent sexism within the context of Judaism in Israel. It attempts to elaborate on previous findings and broaden our understanding of ambivalent sexism and its determinants.

Ambivalent Sexist Attitudes and Jewish Religiosity

As the other western world’s major religions—Christianity and Islam—Judaism fosters gender hierarchy and traditional gender ideologies. In the Orthodox daily prayer service, women thank God for “making me according to his will,” whereas the parallel blessing for men thanks God “who has not made me a woman” (Dashefsky et al. 2003). The Jewish tradition is strongly characterized by all three content domains of ambivalent sexist attitudes. Patriarchy is reflected in Jewish law according to which women’s issues are delineated in terms of their relationship to men. Their credibility as witnesses is severely limited, and they are powerless to effect changes in their own marital status (Adler 1999). Gender differentiation is similarly extensive. The Jewish tradition defines separate spheres for men and women, with men occupying the public sphere and women limited to the private sphere. Accordingly, women are exempted from many of the religious rituals that could undermine their devotion to domestic responsibilities. They are not counted as part of the quorum of 10 persons required for services, are not permitted any service leadership role in the synagogue, and were traditionally exempted from the high status activity of Jewish learning (Dashefsky et al. 2003). Finally, the notion of heterosexual intimacy is deeply rooted in Judaism, which views heterosexual marriage as natural and women as men’s essential counterparts (Sacks 1995).

Importantly, the Jewish tradition assumes a heavy tone of benevolence to all three attitudinal domains. For example, protective paternalism is reflected in the Jewish law which dictates a married man’s obligation to provide his wife with adequate food and clothing. The notion of complementary gender differentiation is expressed in Jewish rabbinical literature where women are described as having greater powers of discernment and as being especially tenderhearted (Greenberg 1998). Consistently, this literature states that a man must love his wife at least as much as himself, but honor her more than himself. Finally, the idea of heterosexual intimacy is deeply embedded within Judaism’s basic assumption that to be whole, women must be partnered with men. Judaism has no role for a woman who has not fulfilled herself as companion to man (Alpert 1992). The Jewish law also dictates a married man’s obligation to provide his wife with regular pleasurable sex, and the rabbinical literature further states that a man without a wife lives without joy, blessing, and good (Greenberg 1998).

It is important to note that the traditional gender ideology embedded in the Jewish religion is particularly strong in the Israeli context, where Orthodox Judaism prevails (Deshen et al. 1995). Reform and Conservative Judaism endorse more liberal and egalitarian gender attitudes and practices. However, these movements constitute a negligible minority in Israel (Deshen et al. 1995; Tabory 1991). Only Orthodox Judaism is officially recognized, and all marriages, divorces, or conversions performed by Conservative and Reform rabbis are legally invalid (Tabory 1991).

Accordingly, several studies on Israeli Jews have shown that their levels of religiosity correlate with traditional gender attitudes and behaviors. For example, studies found that Jewish Israelis’ traditional gender attitudes correlated positively with both their family of origin’s religiosity (Kulik 2002) and their own level of religiosity (Gaunt 2006a; Kulik 2004). Other studies similarly showed that Israeli Orthodox women tended to adhere to more traditional gender ideologies (Moore 2000, 2006), which, in turn, shaped their employment characteristics. Finally, findings revealed that the religiosity level of Jewish Israeli women correlates with their gatekeeping tendencies; that is, their attempts to maintain responsibility for family work, their tendency to associate doing family work with an affirmation of being a good mother, and their expectations for a clear division of labor and distinct spheres for men and women (Gaunt 2008).

Sexism in the Israeli Context

Ambivalent sexist attitudes have been studied in many nations around the world, and have generally been found to be correlated with national measures of gender inequality (Glick et al. 2004). Thus, both hostile and benevolent sexism scores are generally higher in developing than in developed countries (Glick and Fiske 2001). Although Israel is considered to be a developed country with relatively low levels of gender inequality (ranked 22nd on the UN gender inequality measure, United Nations Human Development Report 2011), sexist attitudes may be particularly prevalent in the Israeli context, which combines growing religious dominance, patriarchal ideologies and a strong family orientation (Gavriel-Fried and Ajzenstadt 2011; Katz and Lavee 2005; Strier and Abdeen 2009). As a modern secular state, Israel formally grants women equality and freedom with respect to legal, economic and political rights (Halperin-Kaddari 2004). However, along with the intrusion of religion in the political, legal and cultural spheres (Halperin-Kaddari 2004), the Israeli society is strongly affected by a dominant militaristic discourse, massive waves of immigration from North African and Asian countries which strengthen the patriarchal orientation (Rabin and Lahav 2001), and a familist, pro-natalist- and child-oriented culture which idealizes motherhood as the ultimate form of femininity (Katz and Lavee 2005; Strier and Abdeen 2009; Remennick 2000). The combination of these institutional and cultural circumstances is likely to yield relatively strong sexist attitudes.

Socio-Demographic Characteristics and Sexism

In addition to the role of cultural and religious context in shaping sexist attitudes, other socio-demographic variables may also play a role. Gender effects on ambivalent sexism may be particularly important and have been documented in many different national and cultural contexts (Glick et al. 2000; 2004). Like other subordinate groups, women tend to accept prejudiced ideologies that justify the status quo (Jost and Burgess 2000). However, ostensibly benevolent ideologies are more attractive to women than overtly hostile ideologies that appear to contradict individual and group interests (Glick et al. 2000). This is because benevolent sexism not only justifies hierarchic gender relations but also promises rewards from the dominant group (e.g. protection and admiration). In line with this reasoning, findings in many countries have shown that women tend to endorse benevolent sexism more than hostile sexism, and that men outscore women on hostile but not on benevolent sexism (Glick et al. 2000). Similarly, results indicate that men score lower than women on hostility towards men, and higher on benevolence towards men (Glick et al. 2004).

In addition to gender, Glick, Lameiras, and Castro (2002) highlighted the crucial role of education in decreasing sexist attitudes. Presumably, education helps to promote more liberal gender attitudes and leads highly educated individuals to endorse more egalitarian gender attitudes (Fitzpatrick Bettencourt et al. 2011). In line with this rational, Glick, Lameiras, and Castro (2002) found that more educated Spanish respondents were less likely to endorse sexist attitudes toward men and women. In fact, respondents’ level of education was the strongest and most consistent predictor of sexist beliefs.

Aside from education, age may have an effect on the endorsement of sexist attitudes as well. Several studies have indicated a general increase in egalitarian attitudes among younger generations (e.g., Fitzpatrick Bettencourt et al. 2011; Young-DeMarco and Thornton 2001). For instance, Glick, Lameiras, and Castro (2002) found that older respondents in Spain were more likely to express sexist attitudes, with the exception of hostile sexism which did not increase with age. Lemus et al.’s (2010) study on Spanish teenagers, however, found that respondents generally became less sexist as they progressed through adolescence (Lemus et al. 2010). Specifically, adolescent boys showed less endorsement of benevolent sexism with increased age, whereas adolescent girls showed less endorsement of hostile sexism (Lemus et al. 2010).

The Lemus et al.’s study (2010) further clarified that these effects of age masked a contrasting effect of experience in heterosexual relationships. In their sample, relationship experience predicted greater endorsement of benevolent sexism (but not hostile sexism) among boys, and greater endorsement of hostile sexism (but not benevolent sexism) among girls. Glick, Lameiras, and Castro (2002) similarly found that married men were less likely to endorse hostile sexism and benevolence toward men. The authors speculated that being in a relationship may affect men’s gender beliefs, because of the less sexist beliefs of their female partners (Glick, Lameiras, and Castro 2002).

Overview and Hypotheses

The current study explores the role of Jewish religiosity in ambivalent sexist attitudes while examining the effects of Israeli participants’ age, education level and marital status and controlling for their potential confounding effects with religiosity. This study extends previous work in four important ways. First, Glick, Lameiras, and Castro (2002) suggested that the association of religiosity with sexism should depend on the specific religion and the ideology it endorses. In line with this argument, examining the effect of Jewish religiosity on sexism may shed more light on the various forms this association may take. Second, whereas most of the earlier studies have centred exclusively on the role of religiosity in sexist attitudes toward women, the current study examines hostile and benevolent attitudes toward men as well. Third, most previous studies have not included sociodemographic variables that may play a role in ambivalent sexist attitudes. To fill this gap, the present study examines the role of religiosity while controlling for the effects of participants’ age, education and marital status. Finally, whereas the generalizability of most of the earlier findings was limited by the use of relatively homogeneous samples of university students, the current study follows Glick, Lameiras, and Castro (2002) in using a more diverse community sample.

Consistent with the reasoning delineated above, the following hypotheses were tested:

  1. Hypothesis 1a

    Jewish religiosity will be positively related to benevolent sexism. The effect of religiosity will remain significant even after controlling for participants’ age, education and marital status.

  2. Hypothesis 1b

    Jewish religiosity will be positively related to benevolent attitudes towards men. This effect will remain significant after controlling for participants’ age, education and marital status.

  3. Hypothesis 2

    Consistent with prior research, it is hypothesized that (2a) gender differences will interact with sexism sub-scale such that men will score higher than women on hostility toward women and benevolence toward men, but lower on hostility toward men; (2b) a higher level of education will be associated with a lower endorsement of sexist attitudes; (2c) greater age will be associated with greater endorsement of sexist attitudes; and (2d) marriage will be associated with lower endorsement of hostile sexist attitudes among men.

In addition to tests of the above hypotheses, exploratory tests were also conducted concerning the relationships between Jewish religiosity and hostile attitudes. As no specific predictions were made regarding these relationships, their examination remains exploratory in nature.

Method

Participants

Data were collected from a convenience sample of 854 participants as part of a larger research project on ambivalent sexism and gender attitudes. Participants who identified Islam (n = 16) or Christianity (n = 8) as their religion were excluded from analyses. Data from four additional respondents were discarded because important demographic information was missing. The remaining participants were 826 adults (355 men and 471 women).

Table 1 shows the demographic characteristics of the men and women in the final sample. No gender differences were found in these variables (all t’s and χ2’s were non-significant). The participants’ ages ranged from 18 to 59 years (M = 26, SD = 4.68), with 90 % of the participants between 20 and 30 years of age. Most of the participants were unmarried, and approximately half of them had a university degree. Finally, 61 % of the participants self-identified as secular, 15 % as traditional, 20 % as Orthodox religious and 4 % as Ultra-Orthodox. Although random sampling procedures were not applied, these percentages were very similar to those reported by the Israel Central Bureau of Statistics (2009), with a minor overrepresentation of secular and Orthodox individuals (3 % and 4 % more than in the general population respectively) and a minor underrepresentation of traditional and Ultra-Orthodox individuals (4 % and 2 % respectively).

Table 1 The demographic characteristics of the participants
Full size table

Measures

Ambivalent Sexism

Participants’ attitudes toward women were measured using the 22-item Ambivalent Sexism Inventory (ASI; Glick and Fiske 1996). The 22 items were translated into Hebrew and back-translated as in previous cross-cultural studies with the ASI (Glick et al. 2000; Glick et al. 2004). Following Glick et al.’s recommendation (Glick et al. 2000), non-reversed wording was used for all items in the translated version. Participants responded to the items by using a 6-point scale labeled disagree strongly (0), disagree somewhat (1), disagree slightly (2), agree slightly (3), agree somewhat (4), and agree strongly (5). The ASI consisted of two sub-scales: hostile sexism (HS) which assesses sexist antipathy toward women (e.g., “most women fail to appreciate fully all that men do for them”, “women seek to gain power by getting control over men”), and benevolent sexism (BS) which assesses subjectively positive but patronizing attitudes toward women (e.g., “many women have a quality of purity that few men possess,” “women should be cherished and protected by men”). The average score for each sub-scale was computed to obtain the respondent’s HS and BS scores. A high score reflected more hostile or benevolent attitudes. Cronbach’s alphas for these measures were .90 and .84 respectively.

Ambivalent Attitudes Toward Men

Participants’ attitudes toward men were measured using the 20-item Ambivalence toward Men Inventory (AMI; Glick and Fiske 1999). The 20 items were translated into Hebrew and back-translated as in previous cross-cultural studies with the AMI (Glick et al. 2004). Non-reversed wording was used for all items in the translated version (as recommended by Glick et al. 2000). Participants responded to the items by using a 6-point scale labeled disagree strongly (0), disagree somewhat (1), disagree slightly (2), agree slightly (3), agree somewhat (4), and agree strongly (5). The AMI consisted of two sub-scales: hostility toward men (HM) which assesses resentment toward male dominance (e.g. “men usually try to dominate conversations when talking to women”, “a man who is sexually attracted to a woman typically has no morals about doing whatever it takes to get her in bed”), and benevolence toward men (BM) which assesses appreciation toward men as providers and protectors (e.g., “men are more willing to put themselves in danger to protect others“, “every woman ought to have a man she adores”). The average score for each sub-scale was computed to obtain the respondent’s HM and BM scores. A high score reflected more hostile and benevolent attitudes toward men. Cronbach’s alphas for these measures were .85 and .87 respectively. The complete Hebrew ASI and AMI measures are available from the author.

Religiosity

Participants’ religiosity was indicated on a 4-point scale, labeled as follows: 1 (secular), 2 (traditional), 3 (orthodox), 4 (ultra-orthodox). This self-identification measure is most frequently used to assess Jewish religiosity in Israel, and reflects varying degrees of commitment to Judaism (Yuchtman-Yaar and Peres 2000). Previous studies on Jewish samples showed that this measure is highly correlated with ratings of subjective levels of religiosity (from “not at all religious” to “very religious”, Schwartz and Huismans 1995), as well as with scores on the Systems of Belief Inventory (Baider et al. 2001; Hasson-Ohayon et al. 2009) and men’s synagogue attendance (Lewin-Epstein et al. 2000).

Demographic Variables

Participants reported their gender, age, marital status and level of education. Educational span was a continuous variable, with five options ranging from elementary school to graduate degree.

Procedure

Participants were recruited in five localities in the centre, north and south of Israel. They were personally approached by male and female graduate research assistants in public areas such as cafes, work places and university campuses. The study was introduced to participants as an attitude survey. Completion of the questionnaire took approximately 15 min. Participants were not compensated and all responses were anonymous.

Results

Gender Differences in Sexist Attitudes

Gender differences in mean scores on the ASI and AMI scales were examined using a 2 (Gender) X 4 (ASI/AMI sub-scales: HS, BS, HM, BM) MANOVA (see Table 2). This analysis revealed a significant Gender X Scale interaction, F(3, 821) = 80.89, p < .001. In line with previous research, men scored higher than women on both hostile sexism, F(1, 823) = 67.30, p < .001, and on benevolence toward men, F(1, 823) = 13.51, p < .001, whereas women scored higher than men on hostility toward men, F(1, 823) = 19.09, p < .001. No gender differences were found in benevolent sexism, F(1, 823) = .014, ns. Given these gender differences in mean scores, subsequent analyses are reported separately for men and women.

Table 2 Means, standard deviations and gender differences in ambivalent gender attitudes
Full size table

Intercorrelations Among Sexism Subscales

Table 3 presents correlations among the hostile and benevolent attitudes scores computed separately for men and women. Particularly high correlations were found between benevolent attitudes toward men and women (cf. Glick and Fiske 2001; Glick, Lameiras, and Castro 2002).

Table 3 Correlations between ambivalent gender attitudes and socio-demographic variables
Full size table

Religiosity and Ambivalent Sexist Attitudes

Correlation Analyses

As an initial test of hypotheses 1a and 1b regarding the positive associations between religiosity and benevolent attitudes toward men and women, the correlations between these variables were computed (see Table 3). Following Glick and Fiske’s recommendation (Glick and Fiske 1996, 1999), partial correlations were used to test the relationships of religiosity with benevolence while controlling for hostility and vice versa. As hypothesized, religiosity was consistently positively related to benevolent sexism and benevolence toward men for both male and female participants. As Table 3 further shows, religiosity was also negatively correlated with men’s hostile sexism and hostility toward men.

Regression Analyses

The hypotheses suggested that religiosity would predict benevolent sexism (Hypothesis 1a) and benevolence toward men (Hypothesis 1b) even after controlling for participants’ age, education and marital status. In order to test these hypotheses, a series of multiple regression analyses was performed separately for men and women. To control for the positive correlations between hostile and benevolent attitudes, hostility was entered at the first step as a control variable in the analysis of benevolence and vice versa (cf. Lemus et al. 2010). Religiosity was then entered in a second step, to assess its role in sexist attitudes towards men and women. Finally, a third step assessed the effect of the other demographic variables (age, education, marital status), and whether the effect of religiosity remains significant after controlling for these variables. All variables were assessed for possible multicollinearity using tolerance and the variance inflation factor (VIF). VIF values greater than 10 and tolerance values smaller than .10 would indicate multicollinearity among variables. However, there were no signs of multicollinearity in any of the regression models (VIF values ranged from 1.00 to 1.09 and tolerance values ranged from .92 to .99).

The regression results for male participants are presented in Table 4 (upper part). As hypothesized, religiosity was a significant predictor in the regression equations of benevolent sexism (BS) and benevolent attitudes toward men (BM), β = .29 and β = .24 respectively, p’s < .001. In line with the hypotheses, as Jewish men’s religiosity increased, they were more likely to endorse benevolent attitudes toward men and women. Unanticipated results occurred with respect to hostile attitudes. As can be seen in Table 4, religiosity was a significant negative predictor in the regression equations of hostile sexism (HS) and hostility toward men (HM), β = −.18 and β = −.14 respectively, p’s < .001. Thus, as Jewish men’s religiosity increased, they were less likely to endorse hostile attitudes toward men and women. These effects of religiosity on men’s sexist attitudes remained significant once the other demographic variables were entered into the equation.

Table 4 Hierarchical regression analyses predicting ambivalent sexism and attitudes toward men from religiosity and sociodemographic variables
Full size table

The regression results for female participants were in a similar direction although with lower magnitudes (see lower part of Table 4). In line with the hypotheses, religiosity was a significant predictor in the regression equations of benevolent sexism (BS) and benevolent attitudes toward men (BM) β’s = .16, p’s < .001. That is, similarly to men, as Jewish women’s religiosity increased, they were more likely to express benevolent attitudes toward men and women. The results regarding women’s religiosity and hostile attitudes were mostly insignificant, although pointing to the same direction as those for men. Moreover, when adding the other socio-demographic variables to the regression equation of hostile sexism, women’s religiosity accounted for a unique portion of the variance in hostility. That is, as women’s religiosity increased, they were less likely to endorse hostile attitudes toward women, β = −.08, p < .05.

Socio-Demographic Variables

It was hypothesized that ambivalent sexism would increase with age (Hypothesis 2c) but decrease with level of education (Hypothesis 2b) and men’s marriage (Hypothesis 2d). The regression results for these variables appear in Table 4. The effects of participants’ age were weak and inconsistent. As can be seen in the table, men’s age was a significant negative predictor in the regression equations of both hostile sexism (HS) and benevolent attitudes toward men (BM), β’s = −.12, p’s < .05. That is, older men were less likely to express hostility toward women and benevolence toward men. Table 4 also indicates that women’s age was a negative predictor in the regression equations of hostile sexism (HS), β = −.12, p < .01, but a positive predictor in the equation of benevolent sexism (BS), β = .09, p < .05. That is, older women endorsed less hostile sexist attitudes but more benevolent sexist attitudes.

The results with respect to the effect of education were similarly weak and inconsistent. In line with Hypothesis 2b, more educated men were less likely to express benevolent sexism (BS), β = −.15, p < .001, and more educated women were less likely to express hostile sexism (HS), β = −.08, p < .05. However, education was also a positive predictor in the equation of men’s hostile sexism (HS). That is, more educated men were more likely to express hostility toward women, β = .13, p < .01.

Finally, the correlational results in Table 3 confirmed the hypothesized negative link between marriage and hostile sexism among men (r = −.14, p < .01), along with a positive link with benevolent sexism (r = .12, p < .05). However, these associations were eliminated in the regression analyses (see Table 4) when religiosity and age were taken into account (β’s < .05, ns).

Discussion

This study sought to explore the relationships between religiosity and ambivalent sexism in the context of Judaism in Israel. Due to Judaism’s emphasis on gender complementarity and heterosexual intimacy, it was hypothesized that Jewish religiosity would be positively related to benevolent sexism (Hypothesis 1a) and benevolent attitudes toward men (Hypothesis 1b). No specific predictions were made regarding the relationships between religiosity and hostile attitudes toward men and women.

The results provided strong support for the two hypotheses. As predicted, controlling for participants’ age, education and marital status, religiosity predicted participants’ benevolent attitudes toward men and women. That is, more religious Jews endorsed more benevolent attitudes toward both genders. These findings parallel findings from previous studies with Christian (e.g., Glick, Lameiras, and Castro 2002) and Muslim (Tasdemir and Sakalli-Ugurlu 2010) samples and add to our knowledge of sexism and religiosity in the three Western world’s major religions.

Intriguingly, the findings also revealed negative associations between Jewish religiosity and hostile attitudes, mainly among men. That is, more religious men were less likely to express hostile attitudes toward men and women. Apparently, these findings are unique to the Jewish context. Three previous studies with Christian samples found no relationships between religiosity and hostile sexism (Burn and Busso 2005; Glick, Lameiras, and Castro 2002; Maltby et a. 2010), and a positive association was found among Muslim men (Tasdemir and Sakalli-Ugurlu 2010).

Whereas the positive associations between Jewish religiosity and benevolent sexism presumably reflect Judaism’s views of gender relations, the negative associations between religiosity and hostile sexism may stem from a different feature of Judaism, unrelated to gender issues. According to the Jewish tradition, derogatory speech about other people is strictly forbidden (Goldstein 2006; Strassfeld 2006). This prohibition is based on the Bible and emphasized in the rabbinic literature, where slander in all its forms is subjected to the strongest moral disapproval. The extensive rabbinic discussion encompasses listening to or producing written or verbal slander, direct or insinuating, of an individual or of a whole group of people (Goldstein 2006; Strassfeld 2006). It is therefore possible that more religious respondents were particularly reluctant to express hostile attitudes that are unequivocally negative and derogative to other people.

Several interesting findings emerged with respect to the other socio-demographic variables. As previously found in cross-national studies (Glick et al. 2000; Glick et al. 2004), men scored significantly higher than women on hostile sexism and on benevolence toward men, whereas women scored significantly higher than did men on hostility toward men. As in several other nations, there was no gender difference in benevolent sexism (cf. Glick et al. 2004). These findings reinforce the claim that women tend to accept prejudiced ideologies that justify the status quo, as long as these are not overtly hostile ideologies that contradict individual and group interests (Glick et al. 2000; Jost and Burgess 2000). While justifying hierarchic gender relations, benevolent sexism also promises rewards for women (e.g. protection, admiration) and therefore may be perceived as positive and beneficial.

The findings regarding the effect of education on sexist attitudes stand in contrast to those obtained by Glick, Lameiras, and Castro (2002). Education was the strongest and most consistent negative predictor of sexist attitudes in Glick et al.’s Spanish sample (see also Fitzpatrick Bettencourt et al.’s U.S. sample, 2011), whereas weak and inconsistent associations were found in the current study. Although highly educated men expressed less benevolent sexism and highly educated women expressed less hostile sexism, it was also found that highly educated men expressed more hostile sexism. In addition, education was unrelated to participants’ attitudes toward men. The correlational nature of the data leaves room for various speculations regarding these associations. It is possible, for example, that more liberal women pursue higher education and express less hostile sexism. It is also possible that men who pursue higher education frequently encounter nontraditional women who elicit hostile sexism in them. More research is needed to better account for the complex relationships between education and sexism.

The findings regarding the effect of participants’ age were also weak and inconsistent. Contrary to evidence for an increase in egalitarian attitudes among younger generations (e.g., Fitzpatrick Bettencourt et al. 2011; Young-DeMarco, and Thornton 2001), older men and women in the current study expressed less hostile sexism. Older men also expressed less benevolence toward men, whereas older women endorsed more benevolent sexist attitudes. These findings contrast with Glick, Lameiras, and Castro (2002) findings, where age was associated with more sexist beliefs on all the scales except hostile sexism. Again, the correlational nature of these data does not allow for drawing causal conclusions. It is possible that a third variable that was not taken into consideration in the analysis mediated the relationships between age and sexism. More research is needed to explore this possibility.

The correlational design similarly limits the conclusions that can be drawn regarding the causal links between religiosity and sexist attitudes. It is possible to argue that more conservative sexist persons are attracted to more religious environments that support their views. However, this argument is implausible given that religiosity is largely inherited (Myers 1996). Specifically, studies on intergenerational transmission of religiosity have shown that the most important determinant of adults’ religiosity is the religiosity of their parents (e.g., Flor and Knapp 2001; Kapinus and Pellerin 2008). The powerful role of the family of origin is evident in Israel as well, where religiosity is one of the most prominent divisions of society (Ben-Rafael and Sharot 1991; Deshen et al. 1995). It is therefore more likely that individuals who grew up in a more religious environment adopt more benevolent sexist attitudes than vice versa.

The generalizability of this study is obviously limited by its use of a convenience sample of Jews in Israel. A random representative sample would allow for more confident inferences from the findings, especially with regard to the role of socio-demographic variables. In addition, further studies with diverse samples and a variety of religions would enhance our understanding of the relationships between religiosity and sexist attitudes. Nevertheless, the findings from this study underscore the importance of religiosity in shaping gender attitudes. In the context of Judaism in Israel, religiosity was the strongest and most consistent predictor of ambivalent sexism toward men and women, to a much greater extent than were education and age. Given the crucial role played by sexist beliefs in the maintenance of gender inequalities (e.g., Gaunt 2006b), understanding the social and structural sources of such beliefs is highly important. Whereas Jewish religiosity’s strong associations with benevolent gender attitudes parallel similar associations found for Christianity and Islam, the intriguing associations between Jewish religiosity and reduced hostile attitudes seem unique to Judaism, and their exact nature has yet to be explored.