Political Psychology, Vol 20, No 1, 1999 Social Dominance and Social Identity in the United States and Israel: Ingroup Favoritism or Outgroup Derogation? Shana Levin Department of Psychology, Claremont McKenna College Jim Sidanius Department of Psychology, University of California, Los Angeles American and Israeli university students completed questionnaires in their native languages assessing ingroup identification, social dominance orientation (SDO), and ingroup and outgroup affect The interrelationships among the variables were examined for high- and low-status groups in three intergroup contexts: whites and Latinos in the United States, Ashkenazim and Mizrachim in Israel, and Jews and Arabs in Israel Theoretical predictions of social identity theory and social dominance theory were tested Results indicated that for all high- and low-status groups, stronger ingroup identification was associated with more positive ingroup affect, and for nearly all groups, higher SDO was associated with more negative affect toward the low-status group In addition, SDO was positively associated with ingroup identification for all high-status groups, and negatively associated with ingroup identification for almost all low-status groups Explanations for cross-cultural differences in the factors driving group affect are suggested, and theoretical refinements are proposed that accommodate them KEY WORDS: social dominance; social identity; ethnicity; group affect; ingroup favoritism; cross-cultural psychology One of the most reliable and widely observed phenomena in research on intergroup behavior is the occurrence of intergroup bias, or preferential evaluation of the ingroup relative to an outgroup (for reviews, see Brewer, 1979; Hinkle & Schopler, 1979, 1986; Messick & Mackie, 1989; Mullen, Brown, & Smith, 1992; Tajfel, 1982) Intergroup bias can occur as a result of elevation of the ingroup, derogation of the outgroup, or both These two processes—ingroup favoritism and 99 0162-895X © 1999 International Society of Political Psychology Published by Blackwell Publishers, 350 Main Street, Malden, MA 02148, USA, and 108 Cowley Road, Oxford, OX4 1JF, UK 100 Levin and Sidanius outgroup hostility—have been found to be conceptually and empirically distinct (Fischer, 1994; Struch & Schwartz, 1989; Tajfel & Turner, 1986) Several studies have found mixed results with regard to the location of bias in enhanced ingroup ratings or decreased outgroup ratings In most of the studies reviewed by Brewer (1979), intergroup bias was attributed to enhanced ingroup evaluation (with outgroup ratings remaining relatively constant) (e.g., Dion, 1973; Rabbie & Horwitz, 1969; Rabbie & Wilkens, 1971; Ryen & Kahn, 1975; Wilson & Miller, 1961) However, other studies indicated that variations in bias were attributable either to a decrease in outgroup ratings alone (e.g., Rabbie, Benoist, Oosterbaan, & Visser, 1974; Worchel, Andreoli, & Folger, 1977) or to both an elevation of the ingroup and a devaluation of the outgroup (e.g., Hensley & Duval, 1976; Kahn & Ryen, 1972) These inconsistent findings raise the interesting possibility that ingroup and outgroup ratings may be differentially affected by separate, though related, processes What is clearly needed is a model that explains the factors driving mild forms of ingroup favoritism, like preferring friends of the same ethnicity, as well as the factors driving more extreme forms of outgroup aggression, like the Holocaust in Central Europe and the “ethnic cleansing” in Bosnia Previous research has identified several factors that tend to influence ingroup and outgroup affect This study examines three of these factors: ingroup status, ingroup identification, and SDO In a meta-analysis, Mullen et al (1992) found that intergroup bias tended to increase as a function of ingroup status among artificial laboratory groups and decrease as a function of ingroup status among real-world groups However, reviews of the literature have not specified whether status affects evaluation of the ingroup, evaluation of the outgroup, or both (Brewer & Campbell, 1976; Brown, 1978; Mullen et al., 1992; Sachdev & Bourhis, 1987, 1991; Turner & Brown, 1978) With regard to the effects of ingroup identification, previous research in the tradition of social identity theory has found that ingroup identification is more strongly associated with ingroup favoritism than with intergroup differentiation (Hinkle, Taylor, & Fox-Cardamone, 1989) Social dominance theory, on the other hand, emphasizes the role of SDO—a generalized desire for group-based dominance—in driving both ingroup favoritism and outgroup derogation However, social dominance researchers have not previously isolated the effects of SDO on ingroup and outgroup affect Furthermore, there have been no attempts to separate the differential effects of SDO and ingroup identification on ingroup and outgroup affect, to understand the ways in which ingroup status moderates the effects of these individual-level factors on ingroup and outgroup affect, or to test the generalizability of the effects of SDO, ingroup identification, and ingroup status across different intergroup contexts Therefore, the theoretical purpose of this study is twofold: to examine the effects of both SDO and ingroup identification on the group affect of high- and low-status group members, and to compare the moderating effects of ingroup status on the factors driving ingroup and outgroup affect across three different intergroup contexts Social Dominance and Social Identity 101 Theoretical Explanations for Hypotheses The models of group affect discussed here are analyzed within the theoretical frameworks of social identity theory and social dominance theory Social identity theory According to social identity theory, ingroup favoritism results from attempts to achieve positive group distinctiveness, or favorable evaluation of one’s own group relative to other groups The process is driven by an individual’s social identity needs, or desires to enhance those aspects of one’s self-image that relate to group membership These needs for positive social identity are assumed to drive attempts to achieve positive group distinctiveness across all ingroup status conditions and intergroup contexts Mere categorization of individuals into social groups engages social identity needs The stronger one’s identification with the ingroup, the more meaningful group membership should be to one’s self-image, and the stronger these social identity needs should be Members of all groups should, in turn, attempt to meet these enhanced social identity needs by exhibiting more positive ingroup evaluation Social identity theory therefore expects stronger ingroup identification to be associated with more positive ingroup affect across all ingroup status conditions and intergroup contexts Contrary to these predictions, however, previous research has not consistently demonstrated strong positive relationships between ingroup identification and intergroup bias (Hinkle & Brown, 1990) Hinkle and Brown suggested that two important criteria must be met in order for the identity-maintaining processes described by social identity theory to come into play To make social identity–based intergroup comparisons, members of groups must (a) feel connected with their ingroup and (b) feel concerned about how their group is faring relative to other groups In fact, the two studies reviewed by Hinkle and Brown that did find consistent and robust positive correlations between identification and bias involved rival schools and political parties that fulfilled these two criteria (Abrams, 1984; Kelly, 1988; respectively) Because the intergroup contexts examined in the present study involve real-world groups in conflict, where connections with the ingroup are strong and ingroup-outgroup distinctions are salient and comparative in nature, the predicted positive relationship between ingroup identification and ingroup affect is expected to obtain here as well Social dominance theory Because social identity theory assumes that intergroup biases are primarily expressed through ingroup preference rather than through outgroup derogation (Brewer, 1979), the model implies that ingroup identification should have little or no effect on outgroup affect What, then, are the forces driving outgroup affect? Social dominance theory proposes the existence of a fundamental human desire to establish and maintain systems of group-based social hierarchy This desire, assumed to be a normally distributed individualdifference variable, is referred to as SDO A more recent approach to intergroup relations, social dominance theory is an attempt to integrate individual-difference level social psychology with sociological 102 Levin and Sidanius perspectives concerning the interface between social ideology and social structure (see Gramsci, 1971; Michels, 1962/1991; Mosca, 1896/1939; Pareto, 1901/1979) Social dominance theory begins with the observation that human societies tend to be structured as group-based social hierarchies, with a small number of groups at the top and at least one group at the bottom In addition, these group-based hierarchies tend to be highly stable (e.g., the ethnic hierarchy in the United States; see Smith, 1991), and there tends to be a very high degree of consensuality as to which groups are “dominant” (e.g., European Americans in the United States) and which groups are “subordinate” (e.g., African Americans and Latinos; see Sidanius & Pratto, 1993) On the basis of these observations of the ubiquity, stability, and consensuality of group-based social hierarchy, social dominance theory attempts to identify the precise mechanisms by which these group-based social hierarchies are established and maintained One hierarchy-enhancing mechanism that has been investigated by social dominance researchers is a process known as ideological asymmetry (see Sidanius, 1993; Sidanius, Levin, & Pratto, 1996) Ideological asymmetry refers to systematic differences between members of high-status and low-status groups in the nature of the relationships between group members’ basic values and their attitudes toward different groups and social policies According to the ideological asymmetry hypothesis (Sidanius, 1993), anti-egalitarian values should relate more strongly to attitudes, beliefs, and ideologies that justify intergroup status differences (e.g., racism and sexism) among members of high-status groups than among members of low-status groups Members of high-status groups who value group inequality are expected to exhibit greater support for attitudes that reinforce social hierarchy because these attitudes justify the dominant ingroup’s social position, thereby directly fulfilling desires for social dominance However, because their subordinate status undermines the degree to which they are able to influence the social hierarchy, low-status group members cannot realize anti-egalitarian desires to the same extent that high-status group members can through endorsement of these hierarchy-enhancing attitudes Members of low-status groups are therefore expected to exhibit weaker correlations between their anti-egalitarian values and support for hierarchy-enhancing attitudes and beliefs In line with this ideological asymmetry hypothesis, Sidanius and his colleagues found that SDO—a general measure of anti-egalitarian values—was more strongly and positively related to support for policies that promote social inequality among high-status group members than among low-status group members (Levin, Sidanius, & Pratto, 1998; Mitchell & Sidanius, 1993; Sidanius, 1993; Sidanius et al., 1996; Sidanius, Pratto, & Rabinowitz, 1994; see also Fang, Sidanius, & Pratto, 1998) Another example of this type of ideological asymmetry appears in the work of Mercer and Cairns (1981), who found that political conservatism was significantly associated with anti-Catholic affect among Protestants in Northern Ireland (i.e., the high-status group) but was not significantly associated with anti-Protestant affect among Northern Irish Catholics (i.e., the low-status group) Similarly, Bahr Social Dominance and Social Identity 103 and Chadwick (1974) found that general political conservatism was significantly correlated with anti–Native American affect among American whites (i.e., the high-status group) but was not correlated with anti-white affect among Native Americans (i.e., the low-status group) Social dominance theory assumes that social hierarchy is enhanced to the degree that personality orientations like SDO and political ideologies like conservatism facilitate inegalitarian social attitudes and policies among high-status group members, because they are in a more powerful position than their low-status counterparts to influence intergroup hierarchy Given the logic of social dominance theory, the relationships of SDO with ingroup identification and with ingroup/outgroup affect should also reflect similar asymmetrical patterns Specifically, high SDO should be associated with increased ingroup identification among high-status group members and decreased ingroup identification among low-status group members Moreover, high-SDO group members should exhibit more positive affect toward high-status groups and more negative affect toward low-status groups, regardless of their group membership In terms of the relationships between SDO and ingroup and outgroup affect, then, high-SDO members of high-status groups should exhibit more positive ingroup affect and more negative outgroup affect, whereas their low-status counterparts should exhibit more negative ingroup affect and more positive outgroup affect SDO is expected to differentially influence the ingroup identification, ingroup affect, and outgroup affect of high- and low-status group members because this type of ideological asymmetry will help to maintain existing levels of social hierarchy For example, high-status group members who feel positively about their ingroup and their identity as a group member have been found to show stronger support for social attitudes and policies that reinforce the ingroup’s dominant social position (Levin, Sidanius, Rabinowitz, & Federico, 1998) Similarly, low-status group members who feel negatively about their ingroup and their identity as a group member have also been found to show stronger support for hierarchy-enhancing attitudes and policies, perhaps because they feel that the subordinate status of the ingroup is justified (see Levin, Sidanius, Rabinowitz, & Federico, 1998) Among high-SDO individuals, then, the increased ingroup identification and more positive ingroup affect among high-status group members may operate in concert with the decreased ingroup identification and more negative ingroup affect among lowstatus group members to mutually reinforce the social hierarchy Consistent with these ideological asymmetry predictions, previous research has shown that the relationships of SDO with ingroup identification and intergroup bias (defined as the difference between ingroup and outgroup affect) vary as a function of ingroup status: SDO was found to be positively associated with ingroup identification and intergroup bias among high-status group members, and negatively associated with ingroup identification and intergroup bias among low-status group members (Levin, Sidanius, & Pratto, 1998; Sidanius, Pratto, & Rabinowitz, 1994; see also Sidanius, Feshbach, Levin, & Pratto, 1997) 104 Levin and Sidanius To conclude, social dominance theory supports the social identity theory prediction that categorization into social groups that provide a sense of ingroup identification will trigger positive ingroup affect Moreover, social dominance theory emphasizes the role of SDO in driving ingroup identification and both ingroup and outgroup affect However, although desires for positive social identity are expected to be manifested in more positive ingroup affect, regardless of one’s group membership, desires for social dominance are expected to be manifested in more positive affect toward the high-status group and more negative affect toward the low-status group, regardless of one’s group membership Hypotheses This study considers the relationships among SDO, ingroup identification, and ingroup and outgroup affect among high- and low-status ethnic groups in three different intergroup contexts: whites and Latinos in the United States, Ashkenazim and Mizrachim in Israel,1 and Jews as a superordinate group and Arab citizens of Israel (Ashkenazim and Mizrachim drawn from two different samples are used in the analyses of Ashkenazi-Mizrachi relations and Jewish-Arab relations) Relations between these particular ethnic groups were chosen because of the clear status differences between the groups within each society Relative to their respective counterparts (whites, Ashkenazim, and Jewish Israelis), Latinos in the United States, Mizrachim in Israel, and Arab Israelis have been found to rate lower in social status along such dimensions as education, occupational prestige, and income (Harding, Proshansky, Kutner, & Chein, 1969; Kraus, 1982; Smith, 1991; Smooha, 1978, 1992) These particular intergroup contexts were chosen for study because they vary markedly along a number of important dimensions In Israel, Jews and Arabs have a history of warfare Arabs living inside the borders of Israel established by the outcome of the 1967 war were offered Israeli citizenship, while Arabs living outside the “Green Line” (e.g., in the West Bank and Gaza Strip) were not Jews, on the other hand, are automatically granted Israeli citizenship according to the Israeli “Law of Return,” regardless of where they were born (see Kretzmer, 1990) Although Israeli Arabs (about 18% of the Israeli population) are granted equal rights under Israeli law, there is a wide net of both de jure and de facto restrictions placed on their legal and social rights (see Kretzmer, 1990) The ethnic conflict between Ashkenazim and Mizrachim in Israel is not as salient a social cleavage as the conflict between Jews and Arabs According to Ben-Rafael and Sharot (1991), “most Israeli Jews will rank their Jewish and Israeli identifications before any subethnic identification based on country or area of The diverse Jewish ethnic groups in Israel are classified into two broad categories based on their region of origin: The Ashkenazim are Jews from Europe and North America, and the Mizrachim are Jews from North Africa and the Middle East Social Dominance and Social Identity 105 origin—and these ethnonational identities have no doubt been reinforced by the protracted conflict of Israeli Jews with Arab countries and with the Arab population in the West Bank and Gaza” (p 9) Therefore, while ethnic inequality still exists, Ashkenazim and Mizrachim share powerful superordinate identities as Jews and Israelis, and there is a great deal of intermarriage between the two Jewish ethnic groups (Ben-Rafael & Sharot, 1991) The conflict between American whites and Latinos falls in between these two levels of intensity Although there is some evidence that Latinos are no less patriotic than whites, and that a positive relationship exists between attachment to one’s Latino heritage and attachment to the nation as a whole (see de la Garza, Falcon, & Garcia, 1996), there is intense anti-immigrant feeling in some parts of the United States (as evidenced by, for example, California’s Proposition 187), largely focused on “Latinos” in general and Mexicans in particular Despite the marked structural differences among the three intergroup contexts, we expected to find that the same forces drive group affect within each context Hence, we examined theoretical expectations of cultural invariance with respect to relationships among SDO, ingroup identification, and ingroup and outgroup affect We expected these relationships to be the same for all low-status groups and, separately, for all high-status groups, across the different intergroup contexts Two hypotheses were tested in the cross-context comparisons: For members of high-status groups, SDO should be positively associated with ingroup identification and ingroup affect, and negatively associated with outgroup affect Ingroup identification should also be positively associated with ingroup affect For members of low-status groups, SDO should be negatively associated with ingroup identification and ingroup affect, and positively associated with outgroup affect Ingroup identification should also be positively associated with ingroup affect The first hypothesis describes the predicted model of group affect for highstatus ethnic groups within each intergroup context (i.e., whites, Ashkenazim, and Jews); the second hypothesis states the directions of the predicted effects for low-status ethnic groups within each intergroup context (i.e., Latinos, Mizrachim, and Arabs) Method Participants Data were collected from four different samples of university students: an American sample, two Jewish samples, and an Arab sample American sample: White-Latino relations Data in the American sample were collected from 294 undergraduate students at UCLA surveyed in the fall of 106 Levin and Sidanius 1993 The sample consisted of 161 whites and 133 Latinos There were 129 males and 164 females; one student did not indicate gender Jewish sample 1: Ashkenazi-Mizrachi relations Data in the Jewish sample were collected from 315 undergraduate students surveyed in 1994 at Hebrew University, Bar-Ilan University, and the Technion The Jewish sample consisted of 220 Ashkenazim and 95 Mizrachim; there were 127 males, 181 females, and seven students who did not indicate gender Jewish sample and Israeli Arab sample: Jewish-Arab relations Data in the Jewish sample and the Israeli Arab sample were also collected in 1994 from undergraduate students at Hebrew University and Haifa University The Jewish sample (N = 354) consisted of 205 Ashkenazim, 92 Mizrachim, 53 students of mixed ethnicity, students of other ethnicity, and students who did not indicate ethnicity; there were 130 males, 223 females, and one student who did not indicate gender The Israeli Arab sample (N = 181) consisted of 104 Moslems, 60 Christians, 14 Druze, and three students who did not indicate religion; there were 64 males, 116 females, and one student who did not indicate gender Data for both Jewish samples and for part of the Arab sample were collected in university classrooms where the instructors’ permission was given, and a portion of the Arab sample was collected from students in dormitories at Hebrew University and on buses traveling to Haifa University Measures The survey instruments consisted of a series of questions about SDO, ingroup identification, group affect, perceived group status, and general social and political attitudes The instruments were translated from English into Hebrew and Arabic, then back-translated and compared with the original versions to correct any errors in translation Social dominance orientation The SDO scale contains a series of 16 statements toward which participants’ reactions are measured on a 7-point scale (1, not agree at all; 7, strongly agree) (see Sidanius et al., 1996) The SDO scale is a generalized measure of one’s desires for group-based social inequality and social dominance The scale has exhibited high construct validity across a wide variety of situations and cultures, including the United States, the People’s Republic of China, New Zealand, Canada, and Taiwan (see Pratto et al., 1998; Pratto, Sidanius, Stallworth, & Malle, 1994) The scale has also been found to be distinct from the related constructs of political conservatism, racism, interpersonal dominance, and authoritarianism (see Altemeyer, 1996, 1998; McFarland & Adelson, 1996; Pratto et al., 1994) In the American sample, the 16-item SDO scale was used in its entirety; it was placed in the survey instrument after a series of questions about ethnic relations in the United States In the two Jewish samples and the Israeli Arab sample, the 16-item SDO scale was split into two balanced 8-item scales, which were placed Social Dominance and Social Identity 107 in different parts of the survey instrument All Jewish and Arab respondents completed both 8-item scales, and the order in which the two scales appeared in the survey was counterbalanced To ensure the comparability of these two split halves of the SDO scale, we conducted a reliability analysis using two independent samples The two 8-item scales had a split-half reliability of 96 in both an American sample of whites and Latinos (N = 282) and an Israeli sample of Ashkenazim and Mizrachim (N = 113) (Levin, 1996) For the purposes of the current study, one 8-item scale was used to measure SDO in the two Jewish samples and the Israeli Arab sample The 8-item scale used to measure the effects of SDO among Ashkenazim and Mizrachim (in the Jewish sample 1) appeared in the survey instrument after a series of questions about Ashkenazi-Mizrachi relations The 8-item scale used to measure the effects of SDO among Jews and Arabs (in the Jewish sample and the Israeli Arab sample) appeared in the survey instrument after a series of questions about Jewish-Arab relations Across the American, Jewish, and Arab samples, then, SDO was always measured in the intergroup context of particular relevance to the present study (i.e., American ethnic relations, Jewish ethnic relations, or Jewish-Arab relations), and the SDO scales used across samples have been shown to be essentially equivalent forms Reliability analyses were conducted for each of the three versions of the scale in English, Hebrew, and Arabic In the American sample, the 16-item SDO scale exhibited high reliability (α = 90) The 8-item SDO scales used in the Jewish and Arab samples also had adequate reliability (Jewish sample 1, α = 72; Jewish sample 2, α = 84; Israeli Arab sample, α = 78) Because the SDO scale exhibited decent reliability across the English, Hebrew, and Arabic versions, item scores were averaged within each sample to compute the SDO scale scores Ingroup identification Respondents were also asked questions about their ingroup identification For whites and Latinos in the American sample, and Ashkenazim and Mizrachim in the Jewish sample 1, ingroup identification was operationalized as ethnic group identification In the Jewish sample and Israeli Arab sample, ingroup identification was operationalized as Jewish identification or Arab identification, respectively Ingroup identification was measured by two to four items across the different samples However, for comparison purposes, only those two items that were asked in all the samples were used to measure ingroup identification These items were “To what extent you identify with other members of your group?” and “To what extent you feel close to other members of your group?” Responses were given on a 7-point scale (1, a very small extent; 7, a very large extent) The ingroup identification items were combined to form reliable scales across all samples (American sample, α = 87; Jewish sample 1, α = 86; Jewish sample 2, α = 92; Arab sample, α = 80) Group affect For the group affect measures, respondents were asked to indicate how positively or negatively they felt toward different groups on a 7-point 108 Levin and Sidanius scale (1, very negatively; 7, very positively) The American respondents rated whites and Latinos, Jewish respondents in sample gave affective ratings of Ashkenazim and Mizrachim, and Jewish respondents in sample and Israeli Arabs indicated their feelings toward Jewish and Arab citizens of Israel Perceptions of group status The perceived status of different groups within each society was assessed by single items using a 7-point scale (1, low status; 7, high status) The stem question read: “There are many people who believe that the different groups enjoy different amounts of social status in this society You may not believe this for yourself, but if you had to rate each of the following groups as most people see them, how would you so?” Paired-samples t tests were used to measure perceptions of intergroup status differences between whites and Latinos in the American sample, Ashkenazim and Mizrachim in the Jewish sample 1, and Jews and Arabs in the Jewish sample and the Israeli Arab sample The perceived status differences found in each intergroup context confirmed prior expectations: In the United States, whites (M = 6.50, SD = 69) were rated higher in social status than Latinos (M = 3.13, SD = 1.27) [t(291) = 38.12, p < 001]; in the Jewish ethnic context, Ashkenazim (M = 5.54, SD = 94) received higher status ratings than Mizrachim (M = 3.67, SD = 1.10) [t(304) = 21.50, p < 001]; and in the Jewish-Arab context, Jews (M = 6.43, SD = 83) were rated higher in social status than Arabs (M = 3.31, SD = 1.35) [t(526) = 44.83, p < 001] Moreover, the sizes of these intergroup status differences were found to vary across the three intergroup contexts A measure of the status gap between the highand low-status group in each intergroup context was computed by subtracting the status rating of the low-status group from the status rating of the high-status group A one-way analysis of variance indicated an overall difference in these status gaps across the three contexts [F(2, 1121) = 83.93, p < 0001] A Scheffé procedure was then used to test for significant differences in these status gaps at the α = 05 level The status gap was found to be equally large between whites and Latinos (status gap = 3.37) and between Jews and Arabs (status gap = 3.12) However, the status gap was found to be considerably smaller between Ashkenazim and Mizrachim (status gap = 1.87) relative to the gaps between the groups in the other two contexts Results Table I provides the descriptive statistics for SDO, ingroup identification, ingroup affect, and outgroup affect in the three intergroup contexts The theoretical predictions were tested through the use of single- and multiple-group structural equation analyses in EQS (Bentler, 1995; Bentler & Wu, 1995) The relationships among SDO, ingroup identification, and ingroup and outgroup affect were assessed separately for high- and low-status groups in the contexts of white-Latino relations, Ashkenazi-Mizrachi relations, and Jewish-Arab relations The covariance matrices were used as input in all of the analyses The first step in the analyses consisted of 112 Levin and Sidanius Figure Obtained model for Ashkenazim (high-status group)[χ2(2) = 3.32, p = 19, CFI = 99] Path coefficients are standardized regression coefficients (*p < 05, **p < 01, ***p < 001) The curved line represents correlated error terms constrained model [χ2diff(14) = 13.09, p > 50] As expected, the relationships among the variables were found to be the same for Ashkenazim, Mizrachim, and mixed Jews Therefore, the three Jewish groups, along with four individuals of another or unknown Jewish ethnicity, were combined to form the superordinate group of Jews for each of the model runs in the Jewish-Arab context Similar to the models obtained for whites and Ashkenazim, the obtained model for Jews also closely approximated the predicted model for high-status groups [χ2(1) = 60, p = 44, CFI = 1.00] Positive relationships were found between SDO and Jewish identification (β = 11, p = 05) and between Jewish identification and affect toward Jews (β = 52, p < 001) Also, a negative relationship was found between SDO and affect toward Israeli Arabs (β = –.45, p < 001) However, similar to the model for whites and Ashkenazim and again contrary to expectations, no significant direct relationship was found between SDO and affect toward Jews [nor was there a significant product-moment correlation between SDO and ingroup Social Dominance and Social Identity 113 affect (r = 07, p > 05)], but an indirect positive effect emerged: High-SDO Jews tended to identify more strongly with the ingroup, and this increased ingroup identification was associated with more positive ingroup affect (indirect effect: β = 06, p = 05) Also, a moderate, positive correlation was found between the error terms associated with ingroup affect and outgroup affect (r = 12, p < 05) With the direct path from SDO to ingroup affect excluded, the model provided a good fit to the empirical data [χ2(2) = 69, p = 71, CFI = 1.00; see Figure 3] Comparison among whites, Ashkenazim, and Jews Using multiple-group structural equation analyses, we then compared these models for whites, Ashkenazim, and Jews to see whether the relationships among SDO, ingroup identification, and ingroup and outgroup affect were the same for all high-status groups The fully unconstrained multiple-group model was tested first, in which all parameter estimates indicated by the predicted high-status model (except the paths from SDO to ingroup affect, which were not found to be significant) were free to vary across all groups This model was found to fit the empirical data quite well [χ2(6) = 3.68, p = 72, CFI = 1.00] All estimated path coefficients (between SDO and ingroup identification, identification and ingroup affect, and SDO and outgroup affect) were then constrained to equality across the three groups This constrained model provided an adequate fit to the data [χ2(12) = 12.28, p = 42, CFI = 1.00] The constrained model was then compared to the unconstrained model, and no deterioration of fit was detected in the constrained model [χ2diff(6) = 8.60, p > 19] As expected, the relationships among the variables were found to be the same for whites, Ashkenazim, and Jews For all high-status groups, the higher the SDO, the stronger the identification with the high-status ingroup and the more negative the affect toward the low-status outgroup Also, the higher the ingroup identification, the more positive the ingroup affect However, higher SDO was not found to be directly associated with more positive ingroup affect Rather, the effect of SDO on ingroup affect was mediated by ingroup identification: High-SDO individuals tended to identify more strongly with the high-status ingroup, and this increased ingroup identification was associated with more positive ingroup affect Models for Low-Status Groups An entirely different picture emerged in the comparisons among the low-status Latinos, Mizrachim, and Arabs Latinos The obtained model for Latinos varied significantly from the predicted model for low-status groups [χ2(1) = 19.01, p < 001, CFI = 68] It was expected that SDO would be negatively related to ethnic group identification and affect toward Latinos, and positively related to affect toward whites Ethnic group identification was also expected to relate positively to affect toward Latinos As predicted, we found that higher SDO was associated with weaker ethnic group 114 Levin and Sidanius Figure Obtained model for Jews (high-status group)[χ2(2) = 69, p = 71, CFI = 1.00] Path coefficients are standardized regression coefficients (*p < 05, **p < 01, ***p < 001) The curved line represents correlated error terms identification (β = –.19, p < 05) and more negative affect toward Latinos (β = –.18, p < 05) Also, stronger ethnic group identification was associated with more positive affect toward Latinos (β = 25, p < 01) Unexpectedly, however, there emerged a significant path indicating that stronger ethnic group identification was also associated with more negative affect toward whites (β = –.37, p < 001) In addition, although no significant direct relationship was found between SDO and affect toward whites [nor was there a significant product-moment correlation between SDO and outgroup affect (r = 01, p > 05)], an indirect positive effect emerged: High-SDO Latinos tended to identify less with the ingroup, and this decreased ingroup identification was associated with more positive outgroup affect (indirect effect: β = 07, p < 05) Also, the error terms associated with ingroup affect and outgroup affect were found to be positively correlated (r = 41, p < 001) With the direct path from SDO to outgroup affect excluded, the model fit the empirical data quite well [χ2(1) = 63, p = 43, CFI = 1.00; see Figure 4] Social Dominance and Social Identity 115 Figure Obtained model for Latinos (low-status group)[χ2(1) = 63, p = 43, CFI = 1.00] Path coefficients are standardized regression coefficients (*p < 05, **p < 01, ***p < 001) The curved line represents correlated error terms Mizrachim Although the predicted model for low-status groups was able to adequately fit the data obtained for Mizrachim [χ2(1) = 29, p = 59, CFI = 1.00], the obtained model indicated only two significant relationships Ethnic group identification was found to be positively associated with affect toward Mizrachim (β = 15, p < 01), and a strong positive correlation was found between the error terms associated with ingroup affect and outgroup affect (r = 86, p < 001) No significant relationships were found between SDO and ethnic group identification, between SDO and ingroup affect, or between SDO and outgroup affect [nor were the product-moment correlations statistically significant (all ps > 05)] With these paths excluded, the meager model provided an adequate fit to the data [χ2(4) = 81, p = 94, CFI = 1.00; see Figure 5] Arabs Consistent with nearly all expectations, the obtained model for Arabs very closely approximated the predicted model for low-status groups [χ2(1) = 1.94, p = 16, CFI = 98] Negative relationships were found between SDO and Arab 116 Levin and Sidanius Figure Obtained model for Mizrachim (low-status group)[χ2(4) = 81, p = 94, CFI = 1.00] The path coefficient is a standardized regression coefficient and the curved line represents correlated error terms (*p < 05, **p < 01, ***p < 001) identification (β = –.23, p < 01) and between SDO and affect toward Arabs (β = –.33, p < 001) Also, a positive relationship was found between ingroup identification and ingroup affect (β = 18, p < 05) Contrary to expectations, however, no significant direct relationship was found between SDO and affect toward Jews [nor was there a significant product-moment correlation between SDO and outgroup affect (r = 01, p > 05)] Also, contrary to results from previous models, the correlation between the error terms associated with ingroup affect and outgroup affect did not reach statistical significance With these paths excluded, the model fit the e m p i r i c a l d a t a quite well [χ 2(3) = 4.01, p = 26, CFI = 97; see Figure 6] We can only speculate as to the lack of a positive correlation between the error terms associated with ingroup and outgroup affect among Arabs It is quite possible that the intensity of the regional conflict between Arabs and Jews attenuated the positive association typically found between the measures This interpretation is given further support by the fact that the positivity of the relationship between the error terms associated with ingroup and outgroup affect among Jews vis-à-vis Arabs was also Social Dominance and Social Identity 117 Figure Obtained model for Arabs (low-status group)[χ (3) = 4.01, p = 26, CF I = 97] P ath coeffi cients are standardized regressi on coeffi cients (*p < 05, **p < 01, ***p < 001) Comparison among Latinos, Mizrachim, and Arabs Using multiple-group structural equation analyses, we then compared these models for Latinos, Mizrachim, and Arabs to see whether the relationships among SDO, ingroup identification, and ingroup and outgroup affect were the same for all low-status groups The fully unconstrained multiple-group model was tested first, in which all parameter estimates indicated by the predicted low-status model (except the paths from SDO to outgroup affect, which were not found to be significant) were free to vary across all groups This model was found to fit the empirical data well, despite the nonsignificant paths between SDO and ingroup identification/ingroup affect that were estimated for Mizrachim [χ2(6) = 5.54, p = 48, CFI = 1.00] All relatively modest (r = 12) These low correlations stand in marked contrast to the correlations found between the error terms associated with ingroup and outgroup affect among the American groups and among the Jewish ethnic groups (Ashkenazim and Mizrachim) vis-à-vis each other These low correlations might then well be another sign of the intensity of the intergroup conflict between Jewish and Arab citizens of Israel 118 Levin and Sidanius estimated path coefficients (between SDO and ingroup identification, identification and ingroup affect, and SDO and ingroup affect) were then constrained to equality across the three groups However, this constrained model did not provide an adequate fit to the data, because of unequal paths between SDO and ingroup identification/ingroup affect among the three groups [χ2(12) = 31.96, p < 001, CFI = 91] The constrained model was then compared to the unconstrained model, and a significant deterioration of fit was detected in the constrained model [χ2diff(6) = 26.43, p < 01] A second unconstrained multiple-group model was then tested, in which only those parameter estimates that achieved significance in the single-group models were free to vary (nonsignificant paths between SDO and ingroup identification/ingroup affect for Mizrachim and between SDO and outgroup affect for all three groups were not estimated in this unconstrained model) This model was found to fit the empirical data quite well [χ2(8) = 6.23, p = 62, CFI = 1.00] The estimated path coefficients between ingroup identification and ingroup affect were then constrained to equality across all three groups, and the estimated path coefficients between SDO and ingroup identification and between SDO and ingroup affect were constrained to equality across the Latino and Arab groups only This partially constrained model provided an adequate fit to the data [χ2(12) = 12.56, p = 40, CFI = 1.00] This constrained model was then compared to the unconstrained model, and no deterioration of fit was detected in the constrained model [χ2diff(4) = 6.33, p > 10] As expected, the magnitude and direction of the relationships between ingroup identification and ingroup affect were found to be the same for Latinos, Mizrachim, and Arabs As was the case for high-status groups, strong ingroup identification was associated with more positive ingroup affect among low-status groups as well However, similar relationships between SDO and ingroup identification and between SDO and ingroup affect were found only for Latinos and Arabs Unlike the case for Mizrachim, high-SDO Latinos and Arabs tended to identify less with their low-status ingroups and exhibit more negative ingroup affect Finally, contrary to all expectations, higher SDO was not found to be directly associated with more positive outgroup affect among Latinos, Mizrachim, or Arabs, but an indirect path emerged for Latinos High-SDO Latinos tended to identify less with their ethnic ingroup, and Latinos who identified less with their ingroup tended to exhibit more positive affect toward the white outgroup Discussion This study integrated the predictions of social identity theory and social dominance theory within high- and low-status models of the expected relationships among SDO, ingroup identification, and ingroup and outgroup affect, and tested the models across three very different intergroup contexts While previous researchers have examined the interactive effects of ingroup status and SDO on Social Dominance and Social Identity 119 ingroup identification and intergroup bias (Sidanius, Pratto & Rabinowitz, 1994; Levin, Sidanius, & Pratto, 1998; respectively), the interactive effects of ingroup status, SDO, and ingroup identification on intergroup bias (Levin, 1992; Sidanius, Pratto, & Mitchell, 1994), and the effects of ingroup identification on intergroup bias (for a review, see Hinkle & Brown, 1990), no one has yet compared the interrelationships among these variables across multiple intergroup contexts A cross-context comparison is vital in order to establish the generalizability of the effects of ingroup status, SDO, and ingroup identification on ingroup and outgroup affect Both social identity theory and social dominance theory expect that the same factors will drive ingroup and outgroup affect across intergroup contexts Specifically, social identity theory predicts that group members who identify more strongly with their ingroup will exhibit more positive ingroup affect This expectation was confirmed for both high- and low-status group members Social dominance theory makes specific predictions regarding the nature of the relationships between SDO on the one hand and ingroup identification, ingroup affect, and outgroup affect on the other hand for all high-status group members and, separately, for all low-status group members High-Status Groups Expectations were confirmed for high-status group members For whites, Ashkenazim, and Jews, higher SDO was associated with stronger ingroup identification and more negative outgroup affect High-status group members who want to maintain intergroup status differences may always attempt to so by increasing ingroup identification and exhibiting more negative affect toward the low-status outgroup, across all intergroup contexts, because these are effective strategies for reinforcing the dominant position of the ingroup in society However, contrary to expectations, SDO was not directly related to the ingroup affect of high-status group members; rather, its effect was mediated by ingroup identification Among high-status group members, high SDO leads to stronger ingroup identification, and it is this stronger identification that leads to more positive ingroup affect In a discussion of high-status groups, it is important to emphasize that SDO and ingroup identification had differential effects on ingroup and outgroup affect While ingroup identification consistently had a direct effect on ingroup affect, SDO only had a clear, direct effect on the outgroup affect of high-status group members These findings suggest that the ingroup and outgroup affect of high-status group members may be differentially affected by the separate, though related, needs for social identity and social dominance Among high-status group members, social identity needs may be met more through ingroup favoritism, and social dominance needs may be met more through outgroup derogation These results may partially explain the inconsistent findings of previous research regarding the location of intergroup bias in enhanced ingroup ratings or decreased outgroup ratings: 120 Levin and Sidanius Elevation of the ingroup may occur when social identity needs are salient, and devaluation of the outgroup may occur when social dominance needs are salient among high-status group members Low-Status Latinos and Arabs For low-status groups, no clear, consistent model of interrelationships emerged For Latinos and Arabs, higher SDO was directly associated with weaker ingroup identification and more negative ingroup affect, but not with more positive outgroup affect However, an indirect positive relationship between SDO and outgroup affect, mediated by ethnic group identification, emerged for Latinos Both high-SDO Latinos and high-SDO Arabs identified less with their low-status ingroup, and for Latinos this decreased ingroup identification was associated with more positive affect toward whites We can interpret this difference in the factors driving outgroup affect among Latinos and Arabs in terms of differences in the perceived extent of social stratification and perceived stability and legitimacy of the social hierarchy in the two intergroup situations Using the same samples of participants that were used in this study, Levin (1996) found that the ethnic-status hierarchy in the United States is perceived to be as steep as, but less stable and legitimate than, the Jewish-Arab status hierarchy in Israel.4 In a hierarchical society where status differences are considered to be less stable and legitimate, low-status Latinos may feel that an unfavorable social identity has been conferred unfairly by the intergroup status structure Those Latinos who identify more strongly with the low-status ingroup may then attempt to establish a positive social identity by exhibiting more positive affect toward the ingroup and more negative affect toward the white outgroup Therefore, contrary to the predictions of social identity theory, the outgroup affect of low-status group members may be driven by ingroup identification processes, but perhaps only when the status structure is more hierarchical and perceived to be less stable and legitimate Comparison Among High-Status Groups and Low-Status Latinos and Arabs In comparing the obtained models for high- and low-status groups, it is important to emphasize the asymmetrical effects of SDO across the social status Levin (1996) measured the perceived stability of group-status differences in the United States and Israel with one item on a 7-point scale (1, not agree at all; 7, strongly agree): “There will always be differences in status between ethnic groups/Jews and Arabs” [stability in the United States = 4.54; stability in Israel = 5.02; F(1, 816) = 13.76, p < 001] The perceived legitimacy of group-status differences in the United States and Israel was measured by two items on the same 7-point scale: “Differences in status between ethnic groups/Jews and Arabs are fair” and “Differences in status between ethnic groups/Jews and Arabs are the result of injustice” (reverse-coded) [legitimacy in the United States = 2.83; legitimacy in Israel = 3.09; F(1, 809) = 4.64, p < 05] Social Dominance and Social Identity 121 continuum Consistent with the earlier results of Sidanius, Pratto, and Rabinowitz (1994), SDO was positively related to ingroup identification for all high-status groups, and negatively related to identification for low-status Latinos and Arabs Furthermore, SDO was negatively related to the outgroup affect of all high-status groups, and negatively related to the ingroup affect of Latinos and Arabs The net result of SDO driving more negative outgroup affect among high-status group members and more negative ingroup affect among low-status group members is an increase in derogation of the low-status group among individuals at all levels of the social hierarchy The finding that SDO had no direct effect on the ingroup affect of high-status groups or on the outgroup affect of low-status groups supports a modified version of the ideological asymmetry hypothesis Contrary to the proposition that intergroup bias is more a function of ingroup preference than outgroup hostility (Brewer, 1979), when such bias stems from desires for group-based dominance, it may be expressed more through derogation of low-status groups than through favoritism of high-status groups, regardless of one’s group membership Even ideological asymmetry that is restricted to the derogation of low-status groups can potentially serve to maintain social inequality Consensual derogation of low-status groups will perpetuate intergroup status differences to the extent that negative affect toward low-status groups drives discriminatory attitudes, behaviors, and social policies that reinforce the inferior position of these groups Exhibiting more negative affect toward the low-status outgroup and increasing identification with the high-status ingroup may enable high-status group members to simultaneously meet social dominance and social identity needs However, given the threat to positive social identity posed by unfavorable intergroup comparisons, low-status group members cannot simultaneously meet social dominance and social identity needs by exhibiting more negative ingroup affect This study shows that one way in which high-SDO individuals may express more negative feelings toward the low-status ingroup while at the same time maintaining a positive identity is by identifying less with their inferior ingroup An important question raised by this finding is whether these high-SDO individuals who disidentify with their subordinate ingroup are engaging in personal identification strategies (i.e., attempting to enhance a positive identity for themselves personally) or social identification strategies (i.e., attempting to enhance their social identity as group members) We suggest that they are engaging in social identification strategies, but a further distinction must be made between two types of social identification, as outlined by van Knippenberg and Ellemers (1990, p 139) The first type of social identification involves the perception of the self as a (more or less prototypical) representative of a social group of which one is a member This is the form of self-definition we would expect to find among high-SDO individuals who increase identification with high-status ingroups The second type of social identification involves the perception of the self as a representative (or future representative) of a social group to which one aspires, or 122 Levin and Sidanius of which one hopes or expects to become a member This type of self-definition is more in line with the social identification strategies we would expect to find among high-SDO individuals who decrease identification with low-status ingroups Low-Status Mizrachim: An Exception Unlike the case for Latinos and Arabs, SDO had no effect at all on ingroup identification or group affect for Mizrachim The Mizrachim are a unique group in several respects First, they not only attribute greater importance to their Jewish identity than Ashkenazim, they also incorporate a strong religious component into their ethnic identities (Ben-Rafael & Sharot, 1991) If the Mizrachim perceive the Ashkenazi outgroup as members of a superordinate Jewish ingroup, then their desires for group dominance would not be expressed in weaker ethnic identification and greater interethnic bias in favor of the dominant Ashkenazi outgroup; rather, the effects of SDO would be manifested more strongly in the Jewish-Arab context Mizrachim are also unique in that their group status is not as low compared to Ashkenazim as the group status of Latinos and Arabs is compared to whites and Jews, respectively It may be the case that the effects of SDO are sensitive to the size of these intergroup status differences Hinkle and Brown (1990) suggested that unless connections with the ingroup are strong and ingroup-outgroup distinctions are comparative in nature, the identity-maintaining processes described by social identity theory not come into play It may also be the case that for low-status group members, sizable intergroup status differences must exist in order for the hierarchy-maintaining processes described by social dominance theory to come into play Effects of the Degree of Social Hierarchy For low-status Mizrachim, SDO measured in the more egalitarian context of Ashkenazi-Mizrachi ethnic relations was found to be unrelated to ethnic group identification and affect toward Mizrachim, but in the more hierarchical context of Jewish-Arab relations, where Mizrachim are part of the superordinate, high-status Jewish ingroup, SDO was found to be related to Jewish identification and affect toward Arabs Therefore, although desires for social dominance are not measured with respect to particular groups, but rather with respect to groups in general, the effects of SDO on ingroup identification and group affect may vary as a function of the context in which SDO is measured (see also Levin & Sidanius, 1997) If SDO is measured in a less hierarchical context where the ingroup has low status (e.g., Mizrachim in the ethnic context), then it may not relate to any group-relevant variables But if SDO is measured in a more hierarchical context where the ingroup has low status (e.g., Latinos and Arabs), then it should relate negatively to ingroup identification and ingroup affect Social Dominance and Social Identity 123 Interestingly, the effects of SDO on the ingroup identification and outgroup affect of high-status group members were not influenced by the degree of social hierarchy Our results suggest that if SDO is measured in a context where the ingroup has high status, then it should relate positively to ingroup identification and negatively to outgroup affect across all degrees of social hierarchy That is, high-status group members may use similar strategies to maintain their dominant social position, regardless of the extent to which their status is superior to that of the low-status outgroup It appears as though any degree of social-status difference in favor of the high-status ingroup is sufficient to evoke desires among high-status group members to maintain the status differential Conclusions This study compares the factors driving group affect among high- and lowstatus ethnic groups in the United States and Israel One possible limitation of the study is the reliance on college student samples: College students as a whole are not representative of the general population, and members of low-status ethnic groups (i.e., Latinos, Mizrachim, and Arabs) who attend college may be less representative of their group as a whole, compared to members of high-status groups While the study may be limited in its generalizability to other populations, within a college student population the factors driving the ingroup and outgroup affect of high-status group members were found to be remarkably similar across the white-Latino, Ashkenazi-Mizrachi, and Jewish-Arab contexts This similarity suggests that parallel social dynamics are operating in different intergroup contexts with distinctly different ethnic compositions and different degrees of ethnic stratification On the other hand, some significant differences were found in the factors driving the ingroup and outgroup affect of low-status group members Variations in findings were attributed to important differences in the status structures across the intergroup contexts In the future, different societies could be rated along the dimensions that this study suggests might influence the group affect of low-status group members, such as the extent of social stratification, and the perceived stability and legitimacy of the status hierarchy It may be the case that low-status groups in more hierarchical societies more closely approximate the model for Latinos and Arabs, while low-status groups in more egalitarian societies more closely approximate the model for Mizrachim ACKNOWLEDGMENTS This research was supported in part by a Lady Davis Graduate Fellowship and a Sigma Xi Grant-in-Aid of Research We are grateful to Sonia Roccas and Lilach Sagiv for translating the survey instruments into Hebrew, Wisam Marie for translating the survey into Arabic, and John Hetts, Joshua Rabinowitz, Stacey 124 Levin and Sidanius Sinclair, Colette van Laar, and Joseph Schwarzwald for providing helpful comments on earlier drafts of this paper Correspondence concerning this article should be addressed to Shana Levin, Department of Psychology, Claremont McKenna College, 850 Columbia Avenue, Claremont, CA 91711-6420 Electronic mail may be sent to shana_levin@mckenna.edu REFERENCES Abrams, D (1984) Social identity, self-awareness and intergroup behavior Doctoral thesis, University of Kent at Canterbury Altemeyer, B (1996) The authoritarian specter Cambridge, MA: Harvard University Press Altemeyer, B (1998) The other “authoritarian personality.” Advances in Experimental Social Psychology, 30, 47–92 Bahr, H M., & Chadwick, B A (1974) Conservatism, racial intolerance and attitudes toward racial assimilation among whites and American Indians Journal of Social Psychology, 94, 45–56 Ben-Rafael, E., & Sharot, S (1991) Ethnicity, religion, and class in Israeli society Cambridge: Cambridge University Press Bentler, P M (1995) EQS structural equations program manual Encino, CA: Multivariate Software Inc Bentler, P M., & Wu, E J C (1995) EQS for Windows user’s guide Encino, CA: Multivariate Software Inc Brewer, M B (1979) In-group bias in the minimal intergroup situation: A cognitive-motivational analysis Psychological Bulletin, 86, 307–324 Brewer, M B., & Campbell, D T (1976) Ethnocentrism and intergroup attitudes: East African evidence New York: Halstead Brown, R (1978) Divided we fall: An analysis of relations between sections of a factory work-force In H Tajfel (Ed.), Differentiation between social groups: Studies in the social psychology of intergroup relations (pp 395–429) London: Academic Press de la Garza, R O., Falcon, A., & Garcia, F C (1996) Will the real Americans please stand up: Anglo and Mexican-American support of core American political values American Journal of Political Science, 40, 335–351 Dion, K L (1973) Cohesiveness as a determinant of ingroup-outgroup bias Journal of Personality and Social Psychology, 28, 163–171 Fang, C Y., Sidanius, J., & Pratto, F (1998) Romance across the social status continuum: Interracial marriage and the ideological asymmetry effect Journal of Cross-Cultural Psychology, 29, 290–305 Fischer, S (1994) Effects of self-esteem and ethnicity on cognitive, behavioral and affective aspects of intergroup bias Unpublished master’s thesis, Bar-Ilan University, Israel Gramsci, A (1971) Selections from the prison notebooks London: Wishart Harding, J., Proshansky, H., Kutner, B., & Chein, I (1969) Prejudice and ethnic relations In G Lindzey & E Aronson (Eds.), Handbook of social psychology (2nd ed., vol 5, pp 1–76) Reading, MA: Addison-Wesley Hensley, V., & Duval, S (1976) Some perceptual determinants of perceived similarity, liking, and correctness Journal of Personality and Social Psychology, 34, 159–168 Hinkle, S., & Brown, R (1990) Intergroup comparisons and social identity: Some links and lacunae In D Abrams & M Hogg (Eds.), Advances in social identity theory (pp 48–70) New York: Harvester Wheatsheaf Social Dominance and Social Identity 125 Hinkle, S., & Schopler, J (1979) Ethnocentrism in the evaluation of group products In W G Austin & S Worchel (Eds.), The social psychology of intergroup relations Monterey, CA: Brooks/Cole Hinkle, S., & Schopler, J (1986) Bias in the evaluation of ingroup and outgroup performance In S Worchel & W G Austin (Eds.), Psychology of intergroup relations (pp 196–212) Chicago: Nelson-Hall Hinkle, S., Taylor, L A., & Fox-Cardamone, D L (1989) Intragroup identification and intergroup differentiation: A multicomponent approach British Journal of Social Psychology, 28, 305–317 Kahn, A., & Ryen, A H (1972) Factors influencing the bias towards one’s own group International Journal of Group Tensions, 2, 33–50 Kelly, C (1988) Intergroup differentiation in a political context British Journal of Social Psychology, 27, 319–332 Kraus, V (1982) Ethnic origin as a hierarchical dimension of social status and its correlates Sociology and Social Research, 66, 50–71 Kretzmer, D (1990) The legal status of the Arabs in Israel Boulder, CO: Westview Levin, S (1992) Intergroup biases as a function of social dominance orientation, ingroup status, and ingroup identification Unpublished master’s thesis, UCLA Levin, S (1996) A social psychological approach to understanding intergroup attitudes in the United States and Israel Unpublished doctoral dissertation, UCLA Levin, S., & Sidanius, J (1997, October) Refining the construct of social dominance orientation: An analysis of its context-free and context-specific components Paper presented at the meeting of the Society of Experimental Social Psychology, Toronto Levin, S., Sidanius, J., & Pratto, F (1998) Social dominance orientation, ingroup favoritism and ethnic status: Beyond the asymmetrical ingroup bias effect Manuscript in preparation Levin, S., Sidanius, J., Rabinowitz, J L., & Federico, C (1998) Ethnic identity, legitimizing ideologies, and social status: A matter of ideological asymmetry Political Psychology, 19, 373–404 McFarland, S G., & Adelson, S (1996, July) An omnibus study of personality, values, and prejudice Paper presented at the annual meeting of the International Society of Political Psychology, Vancouver Mercer, G W., & Cairns, E (1981) Conservatism and its relationship to general and specific ethnocentrism in Northern Ireland British Journal of Social Psychology, 20, 13–16 Messick, D M., & Mackie, D (1989) Intergroup relations Annual Review of Psychology, 40, 45–81 Michels, R (1991) Political parties: A sociological study of the oligarchical tendencies of modern democracy New York: Free Press (Original work published 1962) Mitchell, M., & Sidanius, J (1993) Group status and asymmetry in the relationship between ideology and death penalty support: A social dominance perspective National Journal of Sociology, 7, 67–93 Mosca, G (1939) The ruling class: Elements of political science New York: McGraw-Hill (Original work published 1896) Mullen, B., Brown, R., & Smith, C (1992) Ingroup bias as a function of salience, relevance, and status: An integration European Journal of Social Psychology, 22, 103–122 Pareto, V (1979) The rise and fall of the elites New York: Arno (Original work published 1901) Pratto, F., Liu, J H., Levin, S., Sidanius, J., Shih, M., Bachrach, H., & Hegarty, P (1998) Social dominance orientation and the legitimization of inequality across cultures Manuscript submitted for publication Pratto, F., Sidanius, J., Stallworth, L M., & Malle, B F (1994) Social dominance orientation: A personality variable predicting social and political attitudes Journal of Personality and Social Psychology, 67, 741–763 Rabbie, J M., Benoist, F., Oosterbaan, H., & Visser, L (1974) Differential power and effects of expected competitive and cooperative intergroup interaction on intragroup and outgroup attitudes Journal of Personality and Social Psychology, 30, 46–56 126 Levin and Sidanius Rabbie, J M., & Horwitz, M (1969) Arousal of ingroup-outgroup bias by a chance win or loss Journal of Personality and Social Psychology, 13, 269–277 Rabbie, J M., & Wilkens, G (1971) Intergroup competition and its effect on intragroup and intergroup relations European Journal of Social Psychology, 1, 215–234 Rokeach, M (1960) The open and closed mind: Investigations into the nature of belief systems and personality systems New York: Basic Books Ryen, A H., & Kahn, A (1975) Effects of intergroup orientation on group attitudes and proxemic behavior Journal of Personality and Social Psychology, 31, 302–310 Sachdev, I., & Bourhis, R Y (1987) Status differentials and intergroup behaviour European Journal of Social Psychology, 17, 277–293 Sachdev, I., & Bourhis, R Y (1991) Power and status differentials in minority and majority group relations European Journal of Social Psychology, 21, 1–24 Sidanius, J (1993) The psychology of group conflict and the dynamics of oppression: A social dominance perspective In S Iyengar & W McGuire (Eds.), Explorations in political psychology (pp 183–219) Durham, NC: Duke University Press Sidanius, J., Feshbach, S., Levin, S., & Pratto, F (1997) The interface between ethnic and national attachment: Ethnic pluralism or ethnic dominance? Public Opinion Quarterly, 61, 102–133 Sidanius, J., Levin, S., & Pratto, F (1996) Consensual social dominance orientation and its correlates within the hierarchical structure of American society International Journal of Intercultural Relations, 20, 385–408 Sidanius, J., & Pratto, F (1993) The inevitability of oppression and the dynamics of social dominance In P Sniderman & P Tetlock (Eds.), Prejudice, politics, and the American dilemma (pp 173–211) Stanford, CA: Stanford University Press Sidanius, J., Pratto, F., & Mitchell, M (1994) In-group identification, social dominance orientation, and differential intergroup social allocation Journal of Social Psychology, 134, 151–167 Sidanius, J., Pratto, F., & Rabinowitz, J L (1994) Gender, ethnic status, and ideological asymmetry: A social dominance interpretation Journal of Cross-Cultural Psychology, 25, 194–216 Smith, T W (1991) What Americans think about Jews? Working papers on contemporary anti-Semitism New York: American Jewish Committee, Institute of Human Relations Smooha, S (1978) Israel: Pluralism and conflict London: Routledge and Kegan Paul Smooha, S (1992) Arabs and Jews in Israel: Change and continuity in mutual intolerance Boulder, CO: Westview Struch, N., & Schwartz, S H (1989) Intergroup aggression: Its predictors and distinctiveness from ingroup bias Journal of Personality and Social Psychology, 56, 364–373 Tajfel, H (1982) Social psychology of intergroup relations Annual Review of Psychology, 33, 1–30 Tajfel, H., & Turner, J (1986) The social identity theory of intergroup behavior In S Worchel & W G Austin (Eds.), Psychology of intergroup relations (pp 7–24) Chicago: Nelson-Hall Taylor, S E., Peplau, L A., & Sears, D O (1997) Social psychology (9th ed.) Upper Saddle River, NJ: Prentice-Hall Turner, J C., & Brown, R J (1978) Social status, cognitive alternatives and intergroup relations In H Tajfel (Ed.), Differentiation between social groups: Studies in the social psychology of intergroup relations (pp 201–234) London: Academic Press van Knippenberg, A., & Ellemers, N (1990) Social identity and intergroup differentiation processes In W Stroebe & M Hewstone (Eds.), European review of social psychology (vol 1, pp 137–169) Chichester, England: Wiley Wilson, W., & Miller, N (1961) Shifts in evaluations of participants following intergroup competition Journal of Abnormal and Social Psychology, 63, 428–431 Worchel, S., Andreoli, V A., & Folger, R (1977) Intergroup competition and intergroup attraction: The effect of previous interaction and outcome of combined effort Journal of Experimental Social Psychology, 13, 131–140 ... (e.g., Latinos and Arabs), then it should relate negatively to ingroup identification and ingroup affect Social Dominance and Social Identity 123 Interestingly, the effects of SDO on the ingroup... SDO, and ingroup identification on ingroup and outgroup affect Both social identity theory and social dominance theory expect that the same factors will drive ingroup and outgroup affect across intergroup... identity: Some links and lacunae In D Abrams & M Hogg (Eds.), Advances in social identity theory (pp 48–70) New York: Harvester Wheatsheaf Social Dominance and Social Identity 125 Hinkle, S., &