Social Dominance Orientation. a Personality Variable Predicting Social & Political Attitudes

8/12/2019 Social Dominance Orientation. a Personality Variable Predicting Social & Political Attitudes

1/24

Social Dominance Orientation A Personality Variable PredictingSocial and Political Attitudes

(Article begins on next page)

The Harvard community has made this article openly available.Please sharehow this access benefits you. Your story matters.

Citation Pratto, Felicia, James Sidanius, Lisa M. Stallworth, and BertramF. Malle. 1994. Social dominance orientation: A personalityvariable predicting social and political attitudes. Journal ofPersonality and Social Psychology 67, no. 4: 741-763.

Published Version doi:10.1037/0022-3514.67.4.741 ccessed May 23, 2014 1:45:38 AM EDTCitable Link http://nrs.harvard.edu/urn-3:HUL.InstRepos:3207711Terms of Use This article was downloaded from Harvard University's DASH

repository, and is made available under the terms and conditionsapplicable to Other Posted Material, as set forth athttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA
http://osc.hul.harvard.edu/dash/open-access-feedback?handle=1/3207711&title=Social+Dominance+Orientation%3A+A+Personality+Variable+Predicting+Social+and+Political+Attitudeshttp://dx.doi.org/10.1037/0022-3514.67.4.741http://nrs.harvard.edu/urn-3:HUL.InstRepos:3207711http://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:3207711http://dx.doi.org/10.1037/0022-3514.67.4.741http://osc.hul.harvard.edu/dash/open-access-feedback?handle=1/3207711&title=Social+Dominance+Orientation%3A+A+Personality+Variable+Predicting+Social+and+Political+Attitudes


2/24

Journal of Personality an d Social Psychology1994, Vol.6 7,No.4,741-763Copyright1994 bythe American Psychological Association, Inc.0022-3514/94/S3.00

Social Dom inance Orientation: A Personality Variable Predicting Socialand Political AttitudesFelicia Pratto, Jim Sidanius, Lisa M. Stallworth, and Bertram F.Malle

Social dominance orientation (SDO), one'sdegreeof preference for inequality am ong social groups,is introduced. On the basis of social dominance theory, it is shown that (a) men are more socialdominan ce-oriented than w omen, (b) high-SDO people seek hierarchy-enhancing professional rolesand low-SDO people seek hierarchy-attenua ting roles, (c)SDO wasrelated to beliefs in a large num-ber of social and po litical ideologies that support group-based hierarchy (e.g., meritocracy and rac -ism) and to support forpoliciesthathaveimp lications for intergroup relations (e.g.,war,civil rights,and socialprograms), includingnewpolicies. SDO was distinguished from interpersonal dom inance,conservatism, and authoritarianism. SDO was negatively correlated with empathy, tolerance, com-munality, and altruism . The ramifications of SDO in social context are discussed.

Group conflict and group-based inequality are pervasive inhuma n existence. Currently, every continent is enduring someform of ethnic conflict, from the verbal debate over multicul-turalism in the United States and Canada to civil war in Liberiaand B osnia. Other conflicts between groups are ancient: the E u-ropean persecution of Jews, "Holy Wars" waged by Christiansand Muslims around the Mediterranean, imperialism in SouthAmerica, and anti-Black racism in northern Africa and else-where. Regardless of the intensity of the conflict, the partici-pantsjustify their behavior to others by appealing to historicalinjustices, previous territorial boundaries, religious prohibi-tions, genetic and cultural theories of in-group superiority, orother such ideologies.Prompted by the ubiquitous n ature of group-based prejudiceand oppression, we developed social dominance theory (seePratto, in press; Sidanius,1993;Sidanius&Pratto, 1993a). Thetheory postulates that societies minimize grou p conflict by cre-ating consensus on ideologies that promote the superiority ofone group over others (see also Sidanius, Pratto, Martin, &Stallworth, 1991). Ideologies that promote or maintain groupinequality are the tools that legitimize discrimination .Toworksmoothly, these ideologies must be widely accepted within a so-ciety, appearing as self-apparent truth s; hence we call them hi-erarchy-legitimizingmyths.' By contributing to consensual or

normalized group-based inequality, legitimizing myths help tostabilize oppression. That is, they minimize conflict amonggroups by indicating how individuals and social institutionsshould allocate things ofpositiveor negative social value, suchasjobs,gold,blankets, government appointme nts, prison terms,and disease. For example, th e ideology of anti-Black racism h asbeen instantiated in personal acts of discrimination, bu t also ininstitutional discrimination against African-Americans bybanks, public transit authorities, schools, churches, marriagelaws,and the penal system. Social Darwinism and meritocracyare examples of other ideologies that imply that some peopleare not as "good" as others and therefore should be allocatedlesspositive social value than others.Thu s far, we have given examples of legitimizing m yths th atenhance o r m aintain the degree of social inequality. Other ide-ologies may serve to attenuate the amount of inequality. Forexample, the "universal rights of man" and the view summa-rized by "all humans are God's children" are inclusive, egali-tarian ideologies that explicitly do not divide persons into cate-gories or groups. To the extent that such ideologies are widelyshared, there should be less group inequality. There are, then,two varieties of legitimizing myths: hierarchy-enhancing legiti-mizing myths, w hich prom ote greater degrees of social inequal-ity, and hierarchy-attenuating legitimizing myths, which pro-mote greater social equality.

Felicia Pratto , Lisa M. S tallworth, and Bertram F.Malle, Departme ntof Psychology, Stanford University; Jim Sidanius, De partm ent of Psy-chology, University of California atLosAngeles.Weare grateful to a numbe r ofpeoplefor their diligence and creativityin this research: Erron Al-Amin, Jill Andrassy, Sahr Conway-Lanz,Nick Clem ents, Magda Escobar, Jack Glaser,LouisIbarra, Kent Harber,John Hetts, Amy Lee, Johanna Jensen, John Moore, Jenn Pearson,Holly Schaefer, Margaret Shih, Stacey Sinclair, Gayatri Taneja, JackWang, and Wes Williams. Bob Altemeyer, Monisha Pasupathi, V ernonSchabert, Michael Mitchell, Steve Gangestad, Corinne Kosmitzki, TedGoertzel, and three anonymousreviewersprovided useful com mentsona draft ofthisarticle.Correspondence concerning this article shouldbeaddressedtoFeliciaPratto, Department of Psychology, Jordan Hall, Stanford University,Stanford, California 9430 5-2130.

S O CIA L D O M IN A N CE O RIEN TA TIO NGiven our theoretical postulate t hat acceptance of legitimiz-ing myths has significant influence on the degree of inequalityin societies, it is quite imp ortant to und erstand the factors th atlead to the acceptance o r rejection of ideologies that prom ote orattenuate inequality. Social dominance theo ry postulates that a1The term mythis meant to imply that everyone in the society per-ceives these ideologies as explanations for how the world isnot thatthey are false (or true). Social dominance theory is meant only to de-scribe thesocialand psychological processes that acton theseideologies,not to ascertain whether these ideologies are true, fair, moral, orreasonable.

7 41


3/24

74 PRATTO, SIDANIUS, STALLWORTH, AND MALLEsignificant factor is an individual-difference variable calledso-cial dominance orientation(SDO), or the extent to which onedesires that one's in-group dominate and be superior to out-groups.Weconsider SDO to be a general attitudinal orientationtoward intergroup relations, reflecting whether one generallyprefers such relations to be equal, versus hierarchical, that is,ordered along a superior-inferior dimension. The theory postu-lates that people who are more social-dominance oriented willtend to favor hierarchy-enhancing ideologies and policies,whereas those lower on SDO will tend to favor hierarchy-atten-uating ideologies and policies. SDO is thus the central individ-ual-difference variable that predicts a person's acceptance or re-jection of numerous ideologies and policies relevant to grouprelations.

Another way that individuals' levels of SDO may influencetheir contribution to social equality or inequality is in the kindsof social roles they take on, particularly, roles that either en-hance or attenuate inequality. We thus predict that those whoare higher on SDO will become members of institutions andchoose roles that maintain or increase social inequality, whereasthose who are lower on SDO will belong to institutions andchoose roles that reduce inequality.

The purpose of the present research was to demonstrate thatindividual variation in SDO exists and to show that this con-struct behaves according to the theory outlined above. Specifi-cally, our goals were (a) to develop a measure of SDO that isinternally and temporally reliable, (b) to show that SDO is re-lated to the attitudinal and social role variables specified by so-cial dominance theory (predictive validity), (c) to show that themeasure is not redundant with other attitude predictors andstandard personality variables (discriminant validity), and (d)to show that SDO serves as an orientation in shaping newattitudes.

HYPOTHESESThe first set of hypotheses we tested was derived from social

dominance theory and concerned those variables to which SDOshould strongly relate, termedpredictivevalidity.The second setof hypotheses, termed discriminant validity,states either thatSDO should be independent of other variables or that SDOshould have predictive value in addition to the effects of theseother variables. We also hypothesized that SDO should relatemoderately to certain other personality variables, from whichSDO is conceptually distinct. The third set of hypotheses wetested concerns SDO's power to predict new social attitudes.

Predictive ValidityGender

The world over, men and women hold different roles with re-gard to the maintenance of hierarchy. Ubiquitously, men serveas military leaders and hold leadership roles in religious, social,political, and cultural spheres(e.g.,Brown,1991,pp. 110, 137).Moreover, men hold more hierarchy-enhancing attitudes, suchas support for ethnic prejudice, racism, capitalism, and right-wing political parties, than do women (e.g., Avery, 1988; Eisler Loye, 1983;Ekehammar & Sidanius, 1982; Shapiro & Ma-hajan, 1986; Sidanius & Ekehammar, 1980; see review by Si-

danius, Cling, & Pratto, 1991). On the basis of these generalsocietal patterns, we have predicted and shown that, on average,men are more social dominance-oriented than women (seePratto, Sidanius,&Stallworth, 1993;Sidanius, Pratto,&Bobo,in press). We tested this hypothesis with the measure of SDOdeveloped in the present research.

Legitimizing MythsEthnic Prejudice

One of the major kinds of ideology concerning relative groupstatus is ethnic prejudice. In the United States, the most long-standing and widely disseminated version of ethnic prejudice isanti-Black racism. Therefore, we predicted that SDO would bestrongly related to anti-Black racism in the present U.S. sam-ples. In the United States, a theoretical and empirical debateabout how best to measure anti-Black racism has been con-ducted for some time (e.g., see Bobo, 1983; McConahay, 1986;Sears, 1988;Sniderman&Tetlock, 1986a, 1986b). Social domi-nance theory merely postulates that SDO should predict what-ever ideologies are potent within the culture at the time of mea-surement. From our theoretical viewpoint, it does not matterwhether the basis for racism is fairness (e.g., Kluegel & Smith,1986), genetic or biblical racial inferiority theories, symbolicracism (e.g., Sears, 1988), or family pathology (e.g., Moynihan,1965). Any potent ideology that describes groups as unequaland has policy implications is a legitimizing myth and should,therefore, correlate with SDO. During the period the presentresearch was conducted, our subjects' country was engaged in awar against Iraq, so we also measured anti-Arab racism andexpected it to correlate with SDO.Nationalism

A more general kind of in-group prejudice that can occur innation-statesisnationalism, chauvinism, or patriotism. Koster-man and Feshbach (1989) suggested that procountry feelings(patriotism) can be distinguished from comparative prejudice,thatis ,that one's countryisbetter than other countries (nation-alism), and as such should dominate other countries (chauvin-ism). Even so, all three reflect attitudinal bias in favor of thenational in-group, and thus we postulated that patriotism, na-tionalism, and chauvinism would all be significantly related toSDO.Cultural Elitism

All societies share the idea that one of the defining features ofthose who belong to their society (are part of the in-group, or areconsidered by them to be human)isthat they are cultured. Insome societies, including English and American society, anelitist ideology built on the cultured-not cultured distinctionpostulates that the elite class has culture not shared by mid-dle-and working-class people and is therefore more deservingof the finer things in life. We term this legitimizing mythcul-tural elitism,and we expected it to correlate with SDO as well.Sexism

Webelieve that antifemale sexismisa ubiquitous legitimizingmyth, although, as with ethnic prejudice, the content basis of


4/24

SOCIAL DOMINANCE ORIENTATION 74sexist ideology varies widely with religion, cu ltura l history, andtechnology. In the present U.S.samples, we used scales that as-sess sexism as the extent to which people believe men andwomen are "natura lly" different and shouldhavedifferent workroles outside and inside the home (Benson & Vincent, 1980;Rombough&Ventimiglia,1981)and the extenttowhich peoplebelieve that women rather than men can be blamed for un-wanted sexual advances such as rape and sexual harassment(Burt, 1980). We predicted that all of these would be positivelycorrelated with SDO, even controlling for subject sex.Political-Economic Conservatism

Political-economic conservatism is associated with supportfor capitalismversussocialism (e.g., Eysenck, 1971). Given thatcapitalism implies that some people and businesses shouldthrive, while those who are less "competitive" should not, weconsider political-economic conservatism to be a hierarchy-en-hancing legitimizing my th th at should positively correlate withSDO (see also Sidanius & P ratto, 1993b). Other policies sup-ported by conservatives, such as that women should stay homewith children and that the USSR must be kept in its place, di-vide people into groups "deserving " different treatm ent, so wefeel conservatism generally can be viewed as a legitimizingmyth. In fact, W ilson's extensive work on the body of attitudesthat make up conservatism shows that a preference for hierar-chical social relationshipsis oneof conservatism's many dim en-sions (Wilson,1973,p. 22).Noblesse Oblige

Ah ierarchy-attenuating ideology tha texistsin many culturesis that those with m ore resources should share them with thosewho have fewer resources (e.g., the M arxist maxim , "Fro m eachaccordingtohis[sic]ability, to each accordingtohis need," andthe potlatch custom of the Kwakiutl). The English-Americanversioniscallednoblesseoblige,whichweexpected to be nega-tively correlated with SDO.Meritocracy

Another hierarchy-enhancing ideology is that wealth andother social values are already distributed appropriately, basedon the deservingness of the recipients. The Protestant workethic and just world theory are examples of m eritocratic ideol-ogies, so we administered standard measures of belief in theProtestant work ethic and belief in a just world and predictedthat they would be positively correlated with SDO. In theUnited States, attributions for poverty due to laziness or tosome other inherent fault in the p oor are predicated on the ideathat equal opportunity is available to all (Kluegel & Smith,1986), so we wrote an equal opportunity scale and predictedthat it would correlate positively with SDO.Social Policy Attitudes

According to social dominance theory, individuals who aresocial dominance orientedwillfavor social practices that main-tain or exacerbate inequality among groups and will oppose so-cialpractices that reduce group inequality. The pa rticular socialpolicies that correlate w ith SDO may vary from society to soci-

ety, but we predicted tha t SDO would relate to supp ort for, oropposition to, the following policies inU.S.samples.Social Welfare, Civil Rights, and Environmental Policies

Weexpected SDO to correlate with opposition to social poli-cies that would reduce inequality between U.S. nationals andforeigners or immigrants, rich and middle class or poor, menand women, ethnic groups, heterosexuals and homosexuals,and humans versus other species. As such, we measured oursubjects' attitudes toward a variety of government social pro-grams, racial and sexual discrimination laws, gay and lesbianrights, domination of foreigners, and environmental policies. Inseveral samples we also assessed attitudes toward "interracialdating" and "interracial marriage," because miscegenation hasbeen central to the U.S.racial policy debate.Military Policy

Because the military is a symbol of nationalism and can beone of the chief means of domin ation of one nation over others,weexpectedSDOto correlate positively with expressed supportfor m ilitary program s and actions.Punitive Policies

Despite its stated creed to enact equality before the law, theU.S.crimina l justice system shows class and ethnic bias at alllevels from arrest to plea bargaining to sentencing (e.g., Bienen,Alan, Denno, Allison, & Mills, 1988; General AccountingOffice, 1990; Kleck, 1981; Nickerson , Mayo, & Smith, 1986;Paternoster, 1983; Radelet & Pierce, 1985; Reima n, 1990; Si-danius, 1988). As one example, in a review of 1,804 homicidecases in South C arolina, Paternoster (1983) found that in caseswhere Blacks killed Whites, rather th an other Blacks, prosecu-tors were 40 times more likely to request the death penalty. Forthis reason,weexpected support for "law and order"orpunitivepolicies, particularly the death penalty, to be positively relatedto SDO (see also M itchell, 1993;Sidanius, Liu, Pratto,&Shaw,1994).

Discriminant ValidityInterpersonal Dominance

SDO, or preference for unequal relationships among catego-riesof people,isconceptually distinguishable from the comm onpersonality conception of interpersonal dom inance, which con-cernsthe extent to w hich individualsliketobein charge and areefficacious. For example, people who score high on the Califor-nia Personality Inventory (CPI) Dominance scale are confident,assertive, dominant, and task oriented, whereas people whoscore low are unassuming and nonforceful (Gough, 1987,p.6).People who score high on the Jackson Personality ResearchForm (JPRF) Dominance scale attempt to control their envi-ronm ents and influenceordirect other people; they are forceful,decisive, authoritative, an d domineering (Jackson, 1965). Wetested this theoretical distinction between social and task or in-terpersonal dominance by using the CPI and JPRF Dominancesubscales in several samples reported here. We predicted thatSDO would no t correlate w ith these two measures.


5/24

744 PRATTO, SIDANIUS, STALLWORTH, AND MALLEAuthoritarianism

There is clearly some theoretical similarity in the effects ofsocial dominance theory's SDO construct and authoritarianpersonality theory 's authoritarian construct (see Adorno, Fren-kel-Brunswik, Levinson, & Sanford, 1950). High-SDO peopleand authoritarian personalities are theorized to be relativelyconservative, racist, ethnocentric, and prejudiced, and theyshould show little em pathy for lower status others. Our concep-tion of SDO , however, differs from classical auth orita riani sm inseveral respects. First, classical authoritarian theorists viewedauthoritarianism asan aberra nt and pathological condition andasa form of ego-defense against feelings of inadequacy and vu l-nerability (see also Frenkel-Brunswik, 1948, 1949). SDO , how-ever,isnot conceived of in clinical term s,asan aberrant person-ality type, oras aform of ego-defense. Rather, SD Oisconceivedof as a "norm al" h uma n propensity on which people vary. Sec-ond, authoritarian personality theory emphasized the sourcesof authoritarianism as springing from psychodynamic pro-cesses. Specifically, Adorno et al. (1950) postulated that strictand harsh parental styles would provoke conflicts between thechild and parents that would be "unresolved." As a way of re-solving these, the child as an adult w ould submit to authoritiesand be intolerant of those who would no t. In contrast, we theo-rize that such a personal history is unnecessary to developing arelatively high SDO tendency. Rather, both temperament andsocialization probably influence one's level of SDO. Third andmost important, whereas authoritarianism is primarily con-ceived as a desire for individual dominance resulting from ex-periences with a uthority figures, SDO is regarded as the desirethatsomecategories ofpeopledomin ate others. Because the twoconstructs are defined differently, m easure ments of each shouldnot be highly correlated.

Given that authoritarianism should predict many of the samevariables we postulate SDO should predict, it is impo rtant forus to show that SDO has explanatory value in addition to au-thoritarianism. We tested the "marginal utility" of the SDOconstruct by testing whether correlations between SDO andsupport for legitimizing m yths and policies are significant afterpartialing out authoritarianism.Conservatism

Political-economic conservatism serves as a legitimizingmyth in our theory, and thusweexpect it to correlate positivelywith SDO. Conservatism is also a well-known rob ust p redictorof social and political attitudes (e.g., Eysenck & Wilson, 1978;Wilson, 1973). To show that SDO has u tility in addition to po-litical-economic conservatism,wetested whether SDO substan-tially correlated with social attitudes after partialing outconservatism.

Standard Personality VariablesBecausewethink our concept of SDO is a yet unstudied per-sonality dimension, we expected it to be independent of otherstandard personality variables such as self-esteem and theBig-Five personality dimensions: Extraversion, Agreeableness,Openness, Neuroticism, and Conscientiousness (see Costa &MacRae, 1985;John, 1990, for reviews).

Empathy, Altruism, Communality, and TolerancePeople who are highly empathic with others would seem tobe lessprejudiced and discriminato ry against out-groups. T hus,it is reasonable to expect a general concern for other people tobe negatively correlated with SDO. Similarly, any general pro-

social orientation might mitigate prejudiced feelings and behav-iors toward out-group members, so altruism should be nega-tively correlated w ith SDO. Furthermo re, people who are quiteinclusive in their definitions of what constitutes an in-groupshould be less able to discriminate against out-groups, so weexpected communality to be negatively correlated with SDO.Andfinally,because tolerance is the antithesis of prejudice, wemight expect tha t a general m easure of tolerance would be neg-atively correlated with a general desire for in-group superiority.We used Davis' (1983) multidimensional empathy scale, Superand Nevill's (1985) altruism subscale, the Personal AttributeQuestionnaire (PAQ) Com munality scale (Spence, Helmreich,& Stapp, 1974), and the Jackson Personality Inventory JPI)Tolerance scale (Jackson,1976)to testthesehypotheses. If SDOhas merit as a new personality variable, none of these corre-lations should be very high.

P R E S E N T R E S E A R C HOverview

Weexamined data from 13sam ples to test the predictive anddiscriminant validity and reliability of our measure of SDO.Our logic in using this large num ber of samples is to exam inestatistically significant results that are reliable across samples.We organized the results by topic, but we report the results ineach samp le so that the reader can see the m agnitude of effectsin each sample and the stability of the results across samples.At the end of the R esults section, we provide a summ ary of theresults in the form of meta-analyses.

Data CollectionGenerally, subjects were college students who participated ina study called "Social Attitudes" for partial course credit. Allof their responses were anony mous and confidential, and theycompleted batteries of self-administered questionnaires. Sub-jects in Samples2,3b, 5,6 ,8, 9, and13spent about 1 hr in ourlaboratory completing the questionnaires. The experimenterdescribed the study as designed to measure students' social atti-tudes and personal preferences. Subjects in Samples 1 and 13completed the SDO scale after pa rticipating in un related exper-

iments, and subjects in the remaining samples completed theSDO scale and follow-up scales in two consecutive m ass-testingsessions normally conducted on subject pool participants. Allsubjects completed a demographic background sheet and our14-item SDO scale intermixed with related items, a National-ism scale based on Kosterman and Feshbach's (1989) measure,along with other attitude or experience m easures, each havingtheir own instructions and response scales. We also adminis-tered some standard personality or attitude scales accordingto the instructions of their authors. In several samples wealso administered ideological (legitimizing myths) or policyattitude items on a questionnaire entitled "Policy IssuesQuestionnaire."


6/24

SOCIAL DOMINANCE ORIENTATION 745

MeasuresSD O

In previous archival studies, we measured proxies for SDOusing items dealing with equality from the National ElectionStudy or the S6 Conservatism scale (see Sidanius, 1976). In de-veloping the present m easure of SD O, we tested over 70 item swhose content we felt related to SDO or to constructs one candefineasseparate but that m ightbeconsidered adjacent to SDO(e.g., nationalism and prestige-striving), following Loevinger's(1957) suggestion about scale construction . However, on the ba-sis of our desire to develop a simple, u nidimensional scale thatis balanced, we selected 14 items from this extensive question-naire as the SD O scale. The selected items concerned the beliefthat some people are inherently superior or inferior to othersand approval of unequal group relationships (see items in Ap-pendix A). The 14-item SDO scale was balanced in that halfthe items indicated approval of inequality and half indicatedapproval of equality (see items in Appendix A).Weassume thatthese items tap a latent con struct andso weare interested in therelationships between the scale mean and other measures ratherthan relationships between individual SDO items and othermeasures.

SDO is an attitudinal orientation, so instructions read,"Which of the following objects or statements do you have apositive or negative feeling tow ards? Beside each object or state-ment,place anumber from'1 'to'7 'w hich represents the degreeof your positive or negative feeling." The scale was labeledverypositive(7),positive(6),slightly positive(5),neitherpositive nornegative(4),slightly negative(3),negative(2),andvery negative(1).The order of the SDO items and the filler items differedamong Form A, com pleted by Samples 1, 2, 3, and 4; Form B,completed by Samples 5, 6, 7, 8, and 12; and Form C, com-pleted by Samples 9, 10, and 11. The format and instructionsfor the th ree forms w ere identical, and we saw no evidence tha tresults pertinent to reliability or validity issues differed acrossthe questionnaire form. Subsequent to the present research, wehave used just the 14 items on a questionnaire and found reli-ability coefficients of.90and predictive validity results similarto those reported below.Political-Economic Conservatism

Some of the standardscalesassessing political-economic con-servatism actually measure individuals' support for particularsocial policies (e.g., the C-scale, W ilson&Patterson, 1968). Be-cause we wished to measure political-economic conservatismseparately from policy attitudes, and because we wanted to usea m easure that should n ot vary with time and place, we used aself-identified liberal-conservative measure in all samples. Onthedemographic backgroundsheet,the political-economic con-servatism question read, "U se one of the following num bers toindicate your political views in the accom panying categories."Below these instructionswasa scale labeledvery liberal(1),lib-eral(2),slightly liberal(3),m iddle of theroad(4),slightlycon-servative (5),conservative (6), and veryconservative (7) and ablank next to each type of issue: "foreign policy issues," "eco-nomic issues," and "social issues." Political-economic conser-vatismwasthe m ean of self-ratings on these three item s.

AuthoritarianismAuthoritarianism research has been fraught with measure-ment difficulties. After surveying the authoritarianism mea-surement literature, we decided to administer two ratherdifferent measures of authoritarian ism, both of which are bal-anced: the Right W ing Authoritarian(RWA)scale by Altemeyer(1981) and Goertzel's (1987) bipolar personality measure.Goertzel (1987) intended his adjective checklist to m easure thepersonality rather tha n the ideological aspect of authoritarian -ism, but did show that it correlates with attitudes toward poli-cies falling along toughness and consistency dimensions. Alte-meyer's(1981) scale is theonly other internally reliable m easureof authoritarianism that is close to the original conception ofauthoritarianism, including conventionalism, authoritariansubmission, and authoritarian aggression (see Duckitt, 1989,for a review).

Original Legitimizing Myths and Policy A ttitudesThe consent form and instructions informed subjects thattheir opinions and preferences toward a variety ofideas,kindsof people,events,andsoforth wouldbem easured. On our "Pol-icyIssuesQuestionnaire"weincluded items from various legit-imizing myth or policy attitude scales. Items from each scalewere interspersed throughout the questionnaire. Next to eachitem was a 1-7 scale, and the instructions read, "W hich of thefollowing objects, events, or statements do you have a positiveor negative feeling towards? Please indicate your feelings by cir-cling the appropriate number alongside each item. Use one ofthe following responses. Remember, your first reaction is best.Workasquicklyasyou can ." The scale points were labeledverynegative(1),negative(2),slightly negative(3),uncertain orneu-tral(A), slightlypositive(5),positive(6),andverypositive(7).Items from the original legitimizing myths and policy atti-tude scales were selected for their content and for their internalreliability across samples. T hese scales are shown in AppendixB.Several personality m easures were used as well; these are de-scribed in the M ethod section.

MethodSubjects

Although our 1,952 subjects were college students, they representsome diversity in terms of sex, ethnicity, and income groups, comingfrom public and private universities in California. Dem ographic infor-mation about the samplesisshown in Table 1.Samples and Procedures

Sample 1 (spring 1990) consisted of98 University of California atBerkeley undergradu ateswhocompleted the CPI Dom inance, Flexibil-ity, and Capacity for Status subscales (Gough, 1987), the JPR F Dom i-nance subscale (Jackson, 1965), the JPI Tolerance subscale (Jackson,1976),and the Rosenberg (1965) Self-Esteem Scale (RSE).Sample2(fall and w inter 1990-19 91) consisted of 463 San Jose StateUniversity (SJSU) undergraduates who completed the CPI and JPRFDominance subscales; Mirels and Garrett's (1971) Protestant WorkEthic Scale; the Just World Scale (Rubin & Peplau, 1975); the four-factor Interpersonal Reactivity Index (IRI), which measures empathy(Davis, 1983); a number of policy attitude measures; and some demo-graphic descriptors.


7/24

746 PRATTO, SIDANIUS, STALLWORTH, AND MALLETable 1DescriptionofSamples

Measure

nAge range men women

%Euro-American%Asian-American%Hispanic%Black%Arab-American

Under 20K20-30K30-40K40-55K55-7OK70-100K100-150K15O-2OOK200K.+

1

9817-345050

482313151

2

46 315-564753

3840852

12911172014855

3a

8117-21

3b

5717-215149

58164146

4 5Sample

6Age and gender breakdown

1904753

14417-354951

4917-236931

Ethnic breakdown3840852

53241080

59241520Family income108510102115138

21161281914631

668101019191111

7

2245050

49251061

8

11517-59406 0

29511423

171313151713525

9

9717-36336 7

194517108

19151712159236

10

23 15446

6722441

11

100

12

1355941

50331041

13

461000

52331104

Note. Missing numbers indicate that information was not available. Samples 4, 7, 10-13 are probably similar in age distribution and range toSample 3. Income was self-reported annual family income in thousands of dollars.

Sample 3a (September, 1990) consisted of 81 Stanford University un-dergraduates who completed the SDO scale as part of a mass-testingsession. Sample 3b included 57 subjects from the same population whoparticipated in a study in our lab in December, 1990, during which theycompleted the SDO scale again and a number of attitude and personal-ity measures. The overlap of these two samples(N= 25 with completedata) was used to assess the cross-time reliability of SDO.

Sample 4 (January, 1991) consisted of 190 Stanford University un-dergraduates who completed the SDO scale and an attitude scale aboutthe Iraq war assessing environmental concerns in the war, anti-Arabracism, willingness to sacrifice for the war, willingness to restrict civilliberties for the war effort, and support for the use of military force bythe United States against Iraq.

Sample5(fall 1991) consisted of 144 SJSU undergraduates who com-pleted the RSE (Rosenberg, 1965), the Rombough and Ventimiglia(1981) Tri-Dimensional Sexism Scale, the Sexist Attitudes TowardWomen Scale (Benson & Vincent, 1980), the Rape Myths Scale (Burt,1980), the Altruism subscale from the Values Scale (Super & Nevill,1985), and the IRI (Davis, 1983). We also measured policy attitudestoward gay rights, women's equality policies, militarism, punitiveness,racial policies, and environmental policies. In addition, we measuredideologies such as anti-Black racism, elitism, patriotism, belief in equalopportunity, and opposition to miscegenation.

Sample 6 (September, 1991) consisted of 49 Stanford undergraduateswho completed the same measures as subjects in Sample 5.

Sample 7 (September, 1991) consisted of 224 Stanford undergradu-ates who completed a battery of personality questions, including Malleand Horowitz's (1994) bipolar descriptions of Factors I (Extraversion),II (Agreeableness), IV (Neuroticism), and V (Conscientiousness) of theBig-Five personality dimensions (see John, 1990, for a review). A fewweeks later, in the 3 days including and following the day ClarenceThomas was confirmed to the Supreme Court, those subjects who had

given their prior permission were telephoned and asked four questionsabout their opinions regarding this Supreme Court nomination. In all,149 subjects were reached by telephone, and the response rate was100%.

Sample 8 (February, 1992) consisted of 115 Stanford undergraduateswho completed the PAQ (Spence et al., 1974), CPI Dominance scale(Gough, 1987), JPRF Dominance scale (Jackson, 1965), JPI Tolerancescale (Jackson, 1976), IRI (Davis, 1983), RSE (Rosenberg, 1965), apost-Iraq war attitude survey, a general war attitude survey, and a num-ber of other policy attitude measures similar to those in Sample 5.

Sample 9 (April, 1992) consisted of 97 SJSU undergraduates. Theycompleted the CPI and JPRF Dominance subscales; the JPI Tolerancesubscale; the IRI; the Protestant Work Ethic Scale; all 19 of the author-itarian bipolar adjective choices (Goertzel, 1987); Altemeyer's (1981)30-item RWA Scale; John, Donahue, and Kentle's (1992) Big-Five Per-sonality Inventory; the PAQ; McConahay's (1986) Modern RacismScale; and Katz and Hass' (1988) Pro-Black, Anti-Black, and Humani-tarian-Egalitarian Scales. They also completed a number of policy atti-tude items similar to those for Sample 5.

Sample 10 (March, 1992) consisted of 231 Stanford undergraduateswho completed the SDO scale. Two weeks later, 176 of these subjectscompleted a comprehensive survey about their ideologies and generalattitudes about the death penalty and their attitude about the executionof Robert Alton Harris, who was executed by the state of California theday before the survey was administered.

Sample 11 (March, 1991) consisted of 100 Stanford University un-dergraduates who completed the SDO scale and a battery of other ques-tionnaires including Snyder's (1974) self-monitoring scales; Fenigstein,Scheier, and Buss' (1975) Self-Consciousness scales; and Malle and Ho-rowitz' (1994) bipolar adjective versions of Factors I and IV of the Big-Five personality dimensions.

Sample 12 (January, 1992) included 139 Stanford undergraduates


8/24

SOCIAL DOMINANCE ORIENTATION 747who completed the SDO scale in a mass-testing session. Of these, 70also completed Malle and Horowitz' (1994) measures of Factors Iand IV.Sample13included 46 undergraduate men at Stanford during 1990-1991 who were selected to be in an experiment on the basis of havingeither extremely high or low SDO scores in Sam ples 3,4 , and 11.Theyparticipated in theexperiment between6weeks and8mo nths after theirfirst testing and completed the SDO scale again.

ResultsWe first present the internal and temporal reliability of ourSDO scale. We then examine whether this measure related tothe ideological, policy attitude, and h ierarchy role variables pre-dicted by social dom inance theory. We show that SDO was ei-ther independent of other personality variables with which itmight be confused or that it predicted the attitudin al outcomesoverand above the effects of these other variab les.Wealso showthat it was not redundant with other personality measures. Fi-nally,weshow that SDO predicted new social and political atti-

tudes.Tosumm arize the results across samples, we report sim-ple averagesof the inte rnal reliab ility coefficients across samplesand averaged correlations across samples using Fisher's z-to-rtransformation.Reliability of he SDO M easure

UnidimensionalityWe conducted two kinds of analyses to confirm that the 14SDO items assessed a single construct. First, within each sam-ple, principal-components analyses of the 14 SDO itemsshowed that a single dimension captured the bulk of the vari-ance in these items. That is, there was a precipitous drop be-

tween the values of the first and second eigenvalues in everysample. Second, we subjected our largest sam ple, Sample 2 {N= 446 with complete data on all SDO items) to confirmatoryfactor analysis. Using maximum-likelihood estimation, wetested a model in which all 14 items were driven by a singlelatent construct. Each item had a statistically significant rela-tionship to the latent factor(ps


9/24

748 PRATTO, SIDANIUS, STALLWORTH, AND MALLETable 2CoefficientAlphas,Correlation With SubjectGender, an dAverageItemMeansan dVariancesbySample or14-Item Social Dominance Orientation Scale

MeasureaMVariance

1.85.29**2.440.14

2.83.27**2.740.22

3a.84.32**2.550.18

3b.85.31*2.310.17

4.84.32**2.590.21

5.81.112.970.40

Sample6

.84.36*2.500.24

7.89.28*2.590.23

8.82.27**3.020.18

9.80.033.120.36

10.83.30**3.130.66

11.81

2.910.27

12.83.26**2.600.23

Note. Positive correlations with gender indicate that men were higher than women*p


10/24

SOCIAL DOMINANCE ORIENTATION 749Table 3Coefficient Alphaso fLegitimizingMyth Scalesan dCorrelations With Social Dominance OrientationSample

No of Sample2 Sample 3b Sample5 Sample 6 Sample8 Sample9 Sample 10Measure items (n= 408) (n=57) Sample 4 (n= 144) (n= 49) (=11 5) (n= 95) (n=156)Coefficienta

Political-economic conservatism 3 .69 .83 .89*Protestant Work Ethic 19 .68 .75Just W orld 20 .55 .42Nationalism 6 .75 .88 .80Patriotism 12Cultural elitism 7Equal opportunity 6Noblesse oblige 6 .58 .80 .69Anti-Black racism 5 .68 .77

8

.68.83.67.65.68.70

.7 1

.86.89.78.76.73.74

8

78

.56.60

7273.66.80.59.49.72.77

.78

.62

.54

Political-economic conservatismProtestant Work EthicJust WorldNationalismPatriotismCultural elitismEqual opportunityNoblesse obligeAnti-Black racism

.26**- . 0 3.0 9.52**

- . 3 9 ".57**

Correlations.28*.33*.43**.41**

-.54**.42**

44 b

.53**"

_43b

.11

.43**

.43**.51**.51**- .60**.49**

.72**

.67**

.65**.2 3.51**

.61**

.24*

.51**

- .69**.65**

.17*

.0 3

.47**

.22*.44**.34**-.50**.52**

.55**

.72**

-.72**

n = 18 0 . = 90.*p


11/24

750 PRATTO, SIDANIUS, STALLWORTH, ANDMALLETable 5Coefficient Alphaso fPolicyScales,Correlations WithSocialDominanceOrientation, andPartial CorrelationsControlling or Conservatism,AcrossSamples

Policy scale

ChauvinismLaw and orderMilitary programsGay& lesbian rightsWomen's rightsSocial program sRacial policyMiscegenyEnvironmentalpolicies

ChauvinismLaw and orderMilitary programsGay& lesbian rightsWomen's rightsSocial programsRacial policyMiscegenyEnvironmentalpoliciesRepublican partypreference

ChauvinismLawand orderMilitary program sGay & lesbian rightsWomen's rightsSocial programsRacial policyMiscegenyEnvironmentalpolicies

No .ofitems

8432410725

2(=455)

.6 4.67".8 2.6 3.7 8.71.9 6

3b(n =50) 4

Coefficient a.7 3.7 1.7 5.91.7 2.77.81.97

Sample5

(n= 129)

.5 9.7 3.8 5.6 9.7 9.6 8.9 3.7 1

6(n= 37)

.7 3.IT.67.8 6.6 3.8 6.7 2.94.8 0

Correlations of social dominance orientation and policy items.0 8.33**-.32**-.42**-.50**-.42**- .31**

.15**

.23*.27*- . 5 0 * *- . 3 2 * *- . 3 1 * *- . 4 6 * *- . 1 5

.25*

.37**.30**

.33**- . 2 9 * *- . 3 9 * *- . 2 9 * *- . 2 3 * *- . 3 0 * *- . 2 7 * *

.24**

.49**.59**.70**- . 5 5 * *- . 3 4 *- . 7 0 * *- . 6 2 * *- . 3 1 *- . 4 0 * *

.45**Partial correlations removing political-economic conservatism

- .02.16***-.28***-.38**-.30***-.33***- . 2 8 * * *

.1 5

.18- . 3 2 * *- . 3 1 * *- . 2 7 *- . 3 0 * *- . 1 9

.40**.29***

.31***- . 2 9 * * *- . 3 5 * * *- . 3 0 * * *- . 2 2 * * *- . 3 1 * * *- . 2 7 * * *

.16

. 31*

.40**- . 1 4- . 2 7 *- . 4 9 * *- . 3 8 * *- . 0 8- . 3 1 *

8(n= 100)

.6 7

.80.66 C.6 0 c.91

.24*

- . 5 2 * *- . 5 5 * *- . 5 4 * *- . 2 5 *

.33**

.25***

- . 4 6 * * *- . 5 0 * * *- . 4 9 * * *- . 2 3 * *

9(n =89)

.58.77".5 9.8 3.7 4.81".7 7.87.7 6

.14.1 9.47**- . 1 7- . 4 2 * *- . 3 9 * *- . 3 4 * *- . 1 8- . 4 7 * *

.27*

.0 6.1 5

.46***- . 1 5- . 4 0 * * *- . 3 7 * * *- . 3 1 * * *- . 1 7 *- . 4 6 * * *

*T h r e ei tems. bT woi tems. cSixi tems. dSeven i tems.*p


12/24

SOCIAL DOMINANCE ORIENTATION 751utilityasa predictor of policy attitudes over and above political-economic conservatism,wecompu ted the correlations betweenSDOand the policy attitudes reported above after partialing outpolitical-economic conservatism. Of the41significant zero-or-der correlations between SD O and policy attitudes in Table 5,only 5 become nonsignificant when political-economic conser-vatism is partialed out.4A few ofthevery high zero-order cor-relations werereduced substantially, but many more partia l cor-relations were almost the same as the zero-order correlations(see Table5).Across all the sam ples, then, there was no consis-tent evidence that political-economic conservatism could re-place SDO as a predictor of the policy attitudesweassessed.In Sample 9, we assessed another rival predictor of policy at-titudes, namely authoritarianism, using two measures. Alte-meyer's 30-item RWA scale had good internal reliability {a =.78);Goertzel's measure was adequate for a bipolar scale (a =.53).Both measures of authoritarianism correlated with politi-cal-economic conservatism (r = .31 for RWA,r = .29 for theGoertzel measure, both ps < .01), confirming their validity.Neither,however,correlated strongly with SDO.RWAcorrelated.14 (ns) with SDO, and the Goertzel measure correlated .18with SDO(p


13/24

752 PRATTO, SIDANIUS, STALLWORTH, AND M A L L ETable 6Coefficient AlphasofInterpersonal DominanceandSelf-Esteema ndCorrelationsWith Social Dominance Orientation WithinSamples

Measure

CPI DominanceJPRF DominanceRosenberg Self-Esteem

CPI DominanceJPRF DominanceRosenberg Self-Esteem

N o.ofitems

351910

1(n=98)

.82

.81

.87

-.11-.04-.09

2( = 403)

.79

.81

.87

-.03.13**-.18

3a 3b(n=80) (n=57)

Coefficient a.79.74

.88 .88Correlations

-.17-.17

.09 .01

Sample4

(n=90)

.9 0

.1 6

5( = 144)

.87

- . 2 3 * *

6( = 56)

.8 5

-.01

8( = 115)

.71

.73

.84

.24**

.01-.29**

9(n=95)

.71

.69

.83

.20.04- . 1 4 *Note. CPI = California Psychological Inventory; JPRF = Jackson Personality Research Form.*p


14/24

SOCIAL DOMINANCE ORIENTATION 75Lest high levels of dominance orientation be confused withagency, it is importan t to emphasize th at there was no correla-tion between SDO and thePAQAgency scale in Samples8and9 (see Table 7 ).

ToleranceSDO was negatively related to the JPI Tolerance subscale (rs= - .27, - .36, and - .27, allps


15/24

754 PRATTO, SIDANIUS, STALLWORTH, AND MALLETable 8Coefficient AlphasandCorrelationsofSocial Dominance Orientation WithNew Attitudes

Scale or item

Favors military action by U.S.Willing to m ake sacrifices for warFavors suspending liberties for warConcerned about environment in war

Favors appointing a Black personFavors appointing a conservativeFavored Clarence T hom as' confirmationBelieved Anita Hill's testimony

Specific deterrenceFavored executing HarrisPainful executionsBelief in retrib ution

No.ofitemsSample 45444Sample 7

Sample 103285

Coefficient

.78.56.51.57

.70.96.86.74

Correlation withsocial dom inanceorientation

.48**.45**.45**.28**

- .20*.32***.22**-.26***

.35**.36**.42**.51**" p < . 0 5 . * * p < . 0 1 . /x 001

Death PenaltyElsewhere, we have argued th at SDO should also be relatedto support of legal institutions that are discriminatory or in-egalitarian in their effects (Sidanius, 1993; Sidanius, Liu,Pratto,&Shaw, 1994). Mitchell (1993) showed that SDO is re-lated to ideologies that legitimize the use of the d eath penalty,such as the belief in legal retribution or the belief that the death

penalty has a deterrent effect on crime. We assessed SDO inMarch, 1992, in Sample 10; 2 weeks later, we administered anextensive survey about death penalty ideologies and about theexecution of Robert Alton Harris, who was executed by thestate of California the day before the subjects were surveyed.Belief that executions have a specific deterrent effect, supportfor Ha rris' execution, supp ort of painful executions, and beliefin retribution were positively correlated with SDO (seeTable 8).S u m m a r y

Toprovide a sum mary of the correlations withSDO, weusedFisher'sz-to-rformula to average the correlations across sam-ples (e.g., Rosenthal, 1986, p. 27). Totest the average statisticalsignificance of the correlation coefficients, we computed stan-dard normal (Z) scores corresponding to each correlation co-efficient using the large-sample formula Z= rX (N)*.The totalof the Zscoresacross samples divided by the square ro ot of thenumber of samples can be compared with the standard normaldistribution to test the null hypothesis that the pattern of corre-lations obtained over samples occurred because of chance asso-ciations between SDO and the variable in que stion. 7

On average, subject sex correlated .26 with SD O (Z =9.92, p< .001). Thus, the datawereconsistent with our prediction thatmen will be higher on SDO tha n w omen.In terms of discriminant validity, overfivesamples, the aver-

age correlation between SDO and C PI Dom inance was .03 andthe average correlation between SDO and JPRF Dominancewas .01.Over nine samples, the average correlation betweenSDO and RSEwas -.08 . Averaged over four samples, SD O cor-related .03 with Extraversion and .10 with Neuroticism.Clearly SDO is independent of all of these constructs. As ex-pected, SDO had moderate negative correlations with concernfor others, communality, tolerance, and altruism (see averagecorrelations in Table 9). The averaged correlations were clearlydifferent from zero, but they were not high enough that theyindicate redundancy between these measures and SDO either,given tha t they ranged from .28 for altruism to .46 for con-cern for others.SDO strongly and consistently related to belief in a numberof hierarchy-legitimizing myths, most strongly to anti-Blackracism and nationalism. SDO also was strongly negatively re-lated to a hierarchy-attenuating ideology, noblesse oblige. SDOcorrelated consistently positively with beliefs in sexism, equalopportunities, patriotism, cultural elitism, conservatism, and aJust World. The Protestant Work Ethic was the only legitimiz-ing ideology thatwedid not find to relate to SDO reliably oversamples (see Table 9).Finally, SDO showed strong consistent correlations withscales assessing opposition to social programs, racial policies,and women's rights, and with support for military programs.SDO wasalsoconsistently correlated with opposition togayandlesbian rights, environmental programs, and miscegeny andwas consistently correlated with support for U.S. chauvinism,law-and-order policies, and Republican party identification (seeTable 9). SDO also predicted attitudes toward new political

7AZ statistic with smaller variance could also have been computedby using the sample variance to weight the Z from each sam ple, whichwould cause allthepvalues given below to be smaller.


16/24

SOCIAL DOMINANCE ORIENTATION 755Table 9Average Correlations and Significance Tests AcrossSamplesBetween Social Dominance Orientationand PersonalityVariables, Ideologies, and Policy Attitudes

Measure Mean rPersonality variables

Concern for othersCommunalityToleranceAltruism

Anti-Black racismNoblesse obligeNationalismSexismEqual opportunitiesPatriotismCultural elitism

- .46- .3 3- .30- .2 8Ideologies

.5 5- . 5 7.5 4.47.46.4 5.4 0Political-economic conservatism .38Just WorldProtestant Work Ethic

Social programsRacial policyWomen's rightsMilitary programsGay& lesbian rightsEnvironmental programChauvinismMiscegeny

.2 7.11Policy attitudes

- . 47- . 44- . 40.4 4- .37s - .38.34- .25Republican party preference .28Law and order .2 8

n

6232

610812333823

6665533666

Z

-8 .92-4.84-5.31-3 .98

15.0520.3015.9614.917.516.846.9410.263.581.25

-12.74-11.74-11.52-10.12-8 .79-6 .165.34-7 .367.086.38

Note. All Zs were significant atp =.0002) and Prote stant W ork Ethic {p=.10).The meanrwascomputedusing Fisher'sz;ndenotes number of samples.

events, including the 1991 war against Iraq, Clarence Thom asasa Supreme Co urt Justice, and a state execution. Those aspir-ing to hierarchy-enhancing careers had higher SDOscoresthanthose aspiring to hierarchy-attenuating careers. Thus, we haveprovided substantial evidence that SDO (a) can be measuredreliably, (b) is stable over time, (c) is higher among men thanamong women, (d) is higher among those who sup port hierar-chy-enhancing ideologiesandis lower among those whosupporthierarchy-attenuating ideologies, (e) is higher among those whosupport hierarchy-enhancing policies and lower among thosewho support hierarchy-attenuating policies, (f)ishigher amongthose who choose hierarchy-enhancing social roles and loweramong those whochoose hierarchy-attenuatingsocialroles, and(g)serves to orient new social and political attitudes.DISCUSSION

On the basis ofsocialdominance theory, we postulated thatthereisan im portan t individual difference in general preferencefor group dom ination, whichwecallsocial dominanceorienta-tion.As its definition and name implies, SDO m ay best be con-sidered a general social-attitudinal o rientation or implicit valuerelevant to intergroup relations. Some of our data inform us

that there may be a significant relationship between on e's ori-entation toward other persons in general and one's orientationtoward other groups. People who are highly emp athic (specifi-cally, concerned with others' well-being) and to a lesser extent,those whofeel interdependentorcommunalwithothers, tend toprefer egalitarian relationships among g roups. Given tha t SDOpredisposes people to believe in legitimizing myths and dis-crimina tory policies, this would seem to make them less likelyto believe in ideologies that relegate certain persons to "infe-rior" categories and to policies that disadvantage certain groupssystematically. The recent work being done on interdependence(e.g., Depret & Fiske, 1993; Markus & K itayama, 1991) mayend up being quite informative about how to mitigate in-tergroup discrimination.We might note that we came to postulate the existence ofSDO not by thinking about "p ersonality" in the traditional, in-dividualistic sense, but by think ing abou t how group-based hu-man social life is and the considerable data generated by re-searchers of political attitudes. We consider individual differ-ences on SDO to be important not for showing the uniquenessof each person, nor for enabling us to classify persons into tax-onomies. Rather, ourfindingof individual variation on SDO iscentral to our postulation of a dynamic model of human op-pression in which different kinds of people (e.g., with high orlow SDO) play different roles (e.g., enhance or attenuate in-equality) and have different effects on each other (e.g., in howmuch they discriminate in the allocation of resources).Despite significant variations in the degree of oppressionfrom one society to another, it seems to us that many societiesshare the basic social-psychological elem ents that contribute toinequality: socially shared myths that define "superior group"and "inferior gro up" and tha t attem pt to justify this distinctionand the policies that "shou ld" follow from it. As such, we pos-tulate tha t individual variation on SDO could be reliably mea-sured in m any other societies and would show the same p atternof relationships to ideologies, policy attitudes, and hierarchyroleasthose shown here. Some of ourpreviousresearch showedthat group-dom inance o rientation significantly correlated w ithsupport for military programs, capitalism (Sidanius & Pratto,1993b), racism, and sexism (Sidanius, Devereux, & Pratto,1992; Sidanius&Pratto , 1993a) in Sweden, a much more egal-itarian country than the United States, and showed the samegender difference in Australia, Sweden, Russia, and the UnitedStates (see Sidanius, Pratto, &Brief, 1993). In the future, wehope to examine whether SDO relates to rathe r different kindsof legitimizing myths that exist in very different culturalcontexts, such as anti-Semitism in Poland, xenophobia in Ja-pan, or fatalism in Ch ina.

Another kind of research endeavor that could further showthe dynam ic link between SDO and societal oppression woulduse SDO to predict attitudes toward new ideologies or policies.Weexpect that even when societies undergo substantial change,as with mass migration, technological innovation, or dramaticchanges in borders or political leadership, such changes will beaccepted only to the extent that they are satisfactorily legiti-mized and meetthepublic'slevelof desire forsocialdominance.Hence, the operation ofSDOin the invention ofnewlegitimi-zations and processes that assort persons into hierarchy rolesmay inform studies of political and social change.


17/24

756 PRATTO, SIDANIUS, STALLWORTH, AND MALLESDO and Social Class

Several political psychologists and sociologists have postu-lated or investigated the relationship between social class andconservatism, racism, or authoritarianism, with Lipset (1960)postulating that the working class is more authoritarian, andStacey and Green (1968, 1971) and m any others presenting ev-idence tothe contrary.We havemade no predictions concerningwhether SDO should be correlated with social class in eitherdirection; instead we suspect that SDO's relationship to thesevariables may vary as a function of hierarchy-group mem ber-ship,which in some societies would be designated by class andin others by racial group, caste, and so forth.Wetested for cor-relations between SDO and class in the present sam ples, and wefound no statistically significant relationships between SDOand respondents' social class or family income category. Al-though there was substantial heterogeneity in these samples onthese variables for the respondents' families of origin, the edu-cation level in these samples is clearly restricted. In contrast, ina random survey of the Los Angeles area, Sidanius et al. (inpress) found statistically significant and m onotonic decreases inscores from an abbreviated SDO scale with increasing familyincome level and increasing respond ent's education level. Withthe datanowin hand,wecannotsaywhether socialclassesdifferon SDO.

Authoritarianism ReconsideredIn the sample in which we assessed SDO and authorita rian-ism, the correlations betweentwomeasures of authoritarianismand SDO were weak. However, because SDO p redicts many ofthe social attitudes conceptually associated with a uthoritaria n-ism (e.g., ethnocentrism, punitiveness, and conservatism), a

more thorough com parison of these constructsiswarranted. Tobegin w ith, there is little theoretical consensus on the constructof authoritarianism. The original and some contemporary re-searchers have described authoritarianism as a multifacetedconstruct; for example, Heaven (1985) suggested tha t au thori-tarianismispart achievement motivation, part dominance, par tconventionalism, part militarism, part punitiveness, and partethnocentrism. Although we think that the tendency for theseconstructs to covary is extremely important, calling this com-pendium authoritarianismis conceptually unsatisfying. It isneither a clear definition of a un itary c onstruct nor a theory ofwhy the separate constructs covary. In our view, punitiveness,ethnocentrism, conservatism, and sometimes conventionalismcan function as legitimizing myths. Because legitimizing mythstend to be correlated with SDO, they are often spuriously cor-related with one another. Th us, it is entirely possible tha t SDOunderliesthesecorrelations.Byseparately defining th e presum -ably causative value orientation (SDO) and ideological stances(legitimizing myths and policy attitudes) and using distinctmeasures of the constructs, we have avoided the conceptualproblem of describing an individual tendency as a set of corre-lations (see also Duckitt, 1989).

Other definitionsofauthoritarianismhave avoided the com -pendium problem by radically restricting the meaning of theterm. Ray (1976) postulated that part of authoritarianism isleadership desire or directiveness. That SDO did not signifi-cantly correlate with two robust measures of interpersonal

dominanc e or with CPI Capacity for Status orPAQAgency sug-gests that SDO cannot be interpreted as leadership desire. Onthis dimension, SDO and authoritarianism are distinct.There is a new view of authoritaria nism that we see as com-plementary to SDO, namely, Duckitt's (1989) description ofauthoritarianism as the desire for individuals to submit to au-thority figures within the in-group. The three classical dimen-sions of authoritarianism that covary empiricallysubmittingto in-group conventions, submitting to in-group authorities,and desiring to punish m emberswho donot submit to in-groupnorm s and authorities (Altemeyer, 1981)all concern the rela-tion of the individual to the group. T hus, in Duck itt's view, au-thoritarianism is primarily an intragroup phenomenon con-cerning individuals' or groups' attitudes about what the rela-tionship between individuals and their in-groups shouldbe.Bycomparison, SDO concerns individuals' attitudes about whatkind of relationship should exist between in-groups and out-groups, whichisan intergroup phenomenon.

SDO Versus Political-Economic ConservatismThe other well-known individual predictor ofsocialand po-litical attitudes is political-economic co nservatism. The powerof political-economic conservatism to pred ict social and politi-cal attitudes and candidate preferenceisfartoorobust to refute.In fact, we postulate th at political-economic conservatism is aprototypic legitimizing myth: an ideology that separates peopleinto groups and suggests that some groups should be accordedmore positive social value (e.g., tax breaks, funds for schools,and access to health care), whereas other groups should be allo-cated m ore negative social value (e.g., prison term s, censorship,and layoffs).Our analysis of conservatism as a legitimizing my th can ex-plain why, in many previous studies, conservatism was corre-lated with racism (e.g., Dator, 1969; Levinson, 1950; Sidanius&Ekehammar, 1979; Stone &Russ, 1976).We believepolitical-economic conservatism and racism are spuriously correlatedand that both are "drive n" by SDO. In fact, recent studies haveshown that there was no significant residual correlation be-tween political-economic conservatism and racism once SDOwas controlled (see Sidanius & Pratto , 1993a, 1993b; Sidaniusetal., 1992).We believe SDO has significant power to predict policy atti-tudes over and above political-economic conservatism. In thepresent study, virtually all partial correlations between policyattitudes and SDO, controlling for political-economic conserva-tism, were reliable. Part oftheadvantage SDO may have overpolitical-economic conservatism is that SDO is an attitudinalorientation rathe r than a policy doctrine and therefore does notrequire expertiseordeliberate application. Th atis,to formulatea policy attitude consistent with one's political ideology, onemust have a thorough u nderstanding of that ideology and thin kthrough or know how it "should" apply to the acceptance ofnew policy initiatives. In contrast, one's SDO level will providea gut reaction to new policy initiatives that imply changes inintergroup relations, essentially " I like it" or "I do n't like it." Inother words, we expect SDO to be a better predictor of group-relevant social and political attitudes than political-economicconservatism among nonideologues, whenever thoughtful ideo-logical reason ingisnot engaged, and for new social attitudes.


18/24

SOCIAL DOMINANCE ORIENTATION 757The present results can be compared with Wilson's (e.g.,1973)extensive work on conservatism as an attitud inal orienta-tion, rather thanas anideology.Wilson found that conservatismwas a convenient label for describing the dimension underlyinga similar constellation of ideological beliefs and policy attitudesto those we have shown to relate to SD O: racism, nationalism,

ethnocentrism, militarism, law and order, and proestablish-ment politics. Conceptually, our definition ofSDOdiffers fromWilson's conservatism in tha twehave not included fundamen-talist religiosity, antihedonism, or strict morality as part ofSDO. Our suppositionisthatthosebeliefs are particu larly West-ern legitimizing myths that happen to be held by people whomake distinctions between superior and inferior or deservingand undeserving people in a Calvinist vein, but tha t they wouldnot necessarily be related to SD O in all cultural con texts. Like-wise, it seems that forms of ethnic prejudice other than anti-Black racism would be more powerfully related to SDO in cer-tain other cultural contexts. The comp arison between SDO andattitudinal conservatism highlights that ou r mea sure of SDO isrelatively independent of particular cultural beliefs or policies.That is, our items do not specify which groups of people arereferred to (with the exception ofnations)because we felt thata general orientation toward groups could predict attitudes to -ward specific groups o r specific group-relevant policies.We modified our SDO scale so that it related to o ur concep-tual definition more strongly by making each item refer only tothe generic conceptgroup.We compared the 14-item scale inAppendix A with this new scale in a sample of 199 Stanfordstudents.Weadministered the 14-item SD O scale(a=.88)anda brief policy attitude survey with a1(very negative)to 7 (verypositive)response format. Fourweekslater,weadministered thebalanced 16-item SDO scale shown in Appendix C ( a =.91).The two scales correlated .75 with one another (p


19/24

758 PRATTO, SIDANIUS, STALLWORTH, AND MALLEgroup identification were more positively correlated in higherstatus than in lower status groups. Using various ethnic identi-fication measures, we showed that the covariance between eth-nic identification and SD O was statistically significant and pos-itive within a high-status group (i.e.,Euro-Am ericans), whereasthis relationshipwasstatistically significant and negative withina low-status group (Hispanics and African-Americans).These experimental and correlational findings suggest thathigh-status groups will be the m ost discriminatory against out-groups because their m embers are most likely to have both highSDO levels and high levels of in-group identification. It is im-portant to remember that these two variables, in experimentalstudies, interact to ca.use extremely severe out-group discrimi-nation. In total, these results support social dominance theory'scontention that higher status groups will tend to be more in-group serving than lower status groups. That is, higher statusgroups are more discrimina tory against out-groups than lowerstatus groups, and the aggregate affect of this asymmetry is notequal groups in conflict, but the maintenance of hierarchicalgroup relations (see Sidanius&Pratto, 1993a).

SDO and Social RoleBy performing organizational roles, individuals greatly ex-pand their capacity for group discrimination because collectiveinstitutions can often allocate resources or costs on a far largerscale than individuals can. Therefore, institutional discrimina-tion is one of the major contributors to the creation and main-tenance of social inequalities and social hierarchy (e.g., Feagin& Feagin, 1978). The individual organizational m ember, how-ever, is not insignificant. An organization's members help aninstitution perform its hierarchy role by endorsing legitimizingmyths and adapting their SDOlevelsto the institution's norms.

Several processes may contribu te to a m atch between individu-als and institutions. There is mounting evidence that peopleseek roles in an institution compatible with their SDO levelsand ideological beliefs. The present research showed two repli-cations of this relationship between SDO and aspired hierarchyrole. A previous study showed the same relationship betweenaspired hierarchy role and belief in legitimizing myths (Sida-nius, Pratto, Martin,&Stallworth, 1991). In anotherstudy,po-lice recruits were found to have significantly higher S DO levelsand related attitudes than public defenders (Sidanius, Liu,Pratto,&Shaw, 1994). Self-selection into roles based on h ierar-chy-relevant ideologies may be a contributing factor to institu-tional discrimination.An institution also reinforces and contributes to the match

between individuals' attitudes and institutional hierarchy role.For example, White police academy recruits became increas-ingly m ore negative toward Blacks during their first 18 monthsas police (Teahan, 1975). Also, the initial racial attitudes of in-tended hierarchy enhancers resisted the usual liberalizing in-fluence of college the longer they stayed in college (Sidanius,Pratto, M artin, & Stallworth, 1991). Any number of socializa-tion or social influence processes may have caused such effects.Hierarchy-enhancing behaviors or attitudes may even be re-warded by hierarchy-enhancing institutions. O ne study of cam-pus police officers showed that those who were m ost successfulin their careers, as evidenced by several measures such as supe-rior's evaluations and salary increases, tended to score higher

on measures of racism and ethnocentrism (Leitner&Sedlacek,1976). More recently, in a study oftheLos A ngeles Police De-partm ent, the Christopher Comm ission (1991) found that thosepolice officers with the highest number of civilian complaintsfor brutality and excessive force received unusually positive per-formance evaluations from their supervising officers. In addi-tion, these supervisor evaluations "were uniformly optimisticabout the officers' progress and prospects on the force" (Chris-topher etal ., 1991,p. 41).Apparently, individuals and institutions reinforce each oth-er's hierarchy-enhancing tendencies, which we believe makestheir discriminatory behaviors powerful and difficult to change.More research on the detailed processes by which individualsand institutions reinforce one another's prejudices may be use-ful to those seeking to reduce institutional discrimination.To summarize, the present research indicates that SDO, thedesire for group dominan ce, is a significant predictor ofsocialand political attitudes pertaining to intergroup relations andalso of hierarchy roles. Social dominance theory suggests thatthe confluence of this individual-difference variable and a num -ber of social factors including lack of common identity, high in-group status, and social role, contributes to the oppression ofsocial groups.

ReferencesAdorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N.(1950). The authoritarian personality. New York: Norton.Altemeyer, B. (1981). Right-wing authoritarianism. Mani toba: Univer-sity of Manitoba Press.Avery, P. G. (1988). Political tolerance among adolescents. Theory andResearch in Social Education, 16, 183-201 .Benson, P. L., & Vincent, S. (1980). Development and validation of thesexist attitudes toward women scale. Psychology of Women Quarterly,5 , 2 7 6 - 2 9 1 .Bienen, L., Alan, N., Denno, D. W., Allison, P. D., & Mills, D. L.(1988). The reimposi t ion of capi tal punishment in New Jersey: Therole of prosecutorial discretion. Rutgers Law Review(Fall).Bobo, L. (1983). Whites ' opposi t ion to bu sing: Symbol ic racism or re-alistic gro up conflict? Journal of Personality and Social Psychology,

45 , 1196-1210.Brown, D. E. (1991).Human universals.N ew York: McGraw -Hil l .Bur t , M. R. (1980). Cul tural myths and suppor ts for rape. Journal ofPersonality and Social Psychology, 38, 2 1 7 - 2 3 0 .Carm ines, E. G. , & Mclver , J. D. (1981). Analyzing models wi th uno b-served variables: Analysis of covariance structures. In G. W. Bohin-stedt & E.F.Borgatta (Eds.),Social measurement: Current issues(pp .65 -11 5) . Beverly Hills, CA: Sage.Christopher, W., Arguellas, J. A., Anderson, R. A., Barnes, W. R., Es-trada, L. F., Kantor, M., Mosk, R. M., Ordin, A. S., Slaughter, J. B.,

& Tranquada, R. E. (1991).Report of the Independent Comm issionon the Los Angeles Police Department. Sui te 1910, 400 South HopeStreet, Los Angeles, CA 90 071 -289 9.Costa, P. T. , & M acRae, R. R. (1985). The NEO Personality InventoryManual. Odessa, FL: Psychological Assessment Resources.Crocker, J., & Luhtanen, R. (1990). Collective self-esteem and in-groupbias.Journal of Personality and Social Psychology, 58, 6 0 - 6 7 .Dator , J. A. (1969). Wha t 's lef t of the econo mic theory of discr imina-tion? In S. Shulm an & W. Darity, Jr. (Eds.), The question of discrimi-nation: Racial inequality in the U.S. labor market (pp . 335-374) .Middletown, CT: Wesleyan University Press.Davis, M. H. (1983). Measuring individual differences in empathy: Ev-idence for a mul t idimensional approach. Journal of Personality andSocial Psychology,44 , 113-126.


20/24

SOCIAL DOMINANCE ORIENTATION 759Depret,E. E,& Fiske, S. T. (1993). Social cognitionandpower: Somecognitive consequencesofsocialstructure asasourceofcontrolde-privation.In G.Weary,F.Gleicher, &K. L.Marsh (Eds.), Controlmotivation and social cognition(pp. 176-20 2). NewYork:SpringerVerlag.Duckitt,J.(1989). Authoritarianismandgroup identification:A new

view of an old construc t.PoliticalPsychology, 10, 63-84.Eisler,R., &Loye,D.(1983).The"failure"ofliberalism:A reassess-mentof ideology from a feminine-masculine perspective.PoliticalPsychology,4,375-391.Ekehammar,B.,&Sidanius, J . (1982). Sex differences in socio-politicalideology: A replication and extension.BritishJournalofSocialPsy-chology, 21,249-257.Eysenck, H .J.(1971). Social attitude s and social class.BritishJournalof Social andClinicalPsychology, 10,210-212.Eysenck, H.J.,&Wilson, O. D. (1978).T hepsychological basisof ide-ology.Baltimore, MD : University Park Press.Feagin,J. R.,& Feagin,C. B.(1978).Discrimination American style:Institutional racism andsexism.Malabar, FL: Krieger.Fenigstein, A., Scheier, M.F.,& Buss, A.H.(1975). Public and privateself-consciousness: Assessment and theory.Journal ofConsulting andClinicalPsychology, 43,522-527.Frenkel-Brunswik, E . (1948). A study of prejudiceinchildren.HumanRelations, 1,295-306.Frenkel-Brunswik, E. (1949 ). Intolerance of ambiguityasan emotionaland perceptual variable.JournalofPersonality, 18 , 108-143.General Accounting Office.(1990).Deathpenaltysentencing:Researchindicates patternof racialdisparities.United States G eneralAc-counting Office, Re port to Senate and House Com mittees on theJu-diciary (GAO/GG D-90-57). Washington,DC:Government PrintingOffice.Goertzel,T. G. (1987). Authoritarianism ofpersonalityandpoliticalattitudes.JournalofSocialPsychology,127,7-18.Gough,H.(1987).California PsychologicalInventory:Administrator sguide.Palo Alto, CA: Consulting Psychologists Press.Heaven, P.C. L.(1985). Constructionandvalidationofa m easureofauthoritarian personality. JournalofPersonality Assessment,49,545-551.Jackson,D. N.(1965).Personality ResearchForm.Goshen, NY: Re-search Psychologists Press.Jackson, D. N. (1976).JacksonPersonalityInventory.Goshen, NY: Re-search Psychologists Press.John, O. P. (1990). The Big Five factor taxonom y: Dim ensions of per-sonality in the na tural language andinquestionnaires. In L. A. Pervin(Ed.),Handbookof personalitypsychology:Theory and research(pp.66-100). Hillsdale, NJ: Erlbaum.John, O. P., Donahue,E.,& Kentle,R.L. (1992).Th e BigFiveInven-tory:Versions4aand 54.Technical Report, InstituteofPersonalityAssessment and Research.Katz, I.,&Hass, R. G. (1988). Racial ambivalence and Am erican valueconflict: C orrelationalandpriming studiesofdual cognitive struc-tures.JournalofPersonalityand SocialPsychology,55, 893-905.Kleck, G. (1981). Racial discriminationincriminal sentencing:Acrit-ical evaluation of the evidence with additional evidence on the deathpenalty. AmericanSociologicalReview,46, 783-805.Kluegel,J. R.,& Smith, P. M. (1986).Beliefs aboutinequality: Ameri-cans'viewsofwhatis andwhat oughtto be. Hawthorne, NY: Aldinede Gruyter.Kosterman,R.,& Feshbach, S. (1989). Towardameasureofpatrioticand nationalistic attitudes.PoliticalPsychology, 10,257-274.Leitner, D.W,& Sedlacek, W. E. (1976). Characteristicsof successfulcampus police officers. JournalofCollege StudentPersonnel, 17,304-308.Levin,S. L.,& Sidanius,J.(1993).Intergroup biasesasa functionofsocial dominance orientationandin-group status. Unpublishedmanu script, University of California atLosAngeles.

Levinson,D.J. (1950). Politico-economic ideology and group m ember-shipsinrelationtoethnocentrism.InT. Adorno, E. Frenkel-Bruns-wik,D. J.Levinson, &R. N. Sanford (Eds.),T heauthoritarianper-sonality(pp. 151 -207). NewYork:Norton.Lipset,S.M . (1960).Politicalman.London: Heinemann.Loevinger,J.(1957). Objective tests as i nstrum ent of psychological the-ory.Psycho logical Reports,M onograph Supplement9.Malle,B.E ,&Horowitz, L. M . (1994).T he puzzle ofnegative self-views:An explanation usingth eschemaconcept. Unpublished manuscript,Stanford University.Markus,H. R.,& Kitayama,S.(1991). Cultureand the self:Implica-tionsfor cognition, emotion,andmotivation.PsychologicalReview,98 ,224-253.McConahay,J.B. (1986). Modern racism, ambivalence,andth emod-ern racism scale. In J. F. Dovidio&S. L. Gaertner (Eds.),Prejudice,discrimination, an dracism(pp. 91-125).SanDiego, CA: AcademicPress.Mirels, H. L.,&Garrett, J.B.(1971). The Protestant ethicasaperson-ality variable.JournalofConsulting and ClinicalPsychology,36,40-44 .Mitchell, M. (1993).Attitudes towards the death penalty and useofexe-cutions:A socialdominanceperspective. Unpublished dissertation,Dep artme nt of Psychology, University of CaliforniaatLosAngeles.Moynihan,D.P.(1965).The Negro am ily: The case or national action.Washington, DC: OfficeofPolicy Plan ningandResearch, U.S.De-partment of Labor.Nickerson, S., Mayo,C,&Smith, A. (1986). Racisminthe courtroom.InJ.F. Dovidio & S.L.Gaertner (Eds.),Prejudice, discrimination,and racism(pp.255-27 8). San Diego, CA: Academic Press.Paternoster,R.(1983). Race of victimandlocation of crime : The deci-sion to seek the death penalty in South Ca rolina.Journalo f CriminalLaw andCriminology, 74 , 754-785.Peterson,B.E., Doty, R. M.,&Winter, D. G. (1993). A uthoritarianismand attitudes towards contem porary social issues.Personality and So-cial Psychology Bulletin, 1 9,174-184.Pratto,F.(in press). Sexual politics: The gender gap in the bedroom andthe cabinet.In D. M.Buss &N.Malamuth (Eds.),Sex,power,and

conflict: Evolutionaryand feminist perspectives. NewYork: OxfordUniversity Press.Pratto, E, Sidanius, J.,&Stallworth, L. M. (1993). Sexual selection andthe sexual and ethnicbasisof social hierarchy. In L.Ellis(Ed.),Socialstratificationan dsocioeconomicinequality:Acomparative biosocialanalysis(pp.111-137). NewYork:Praeger.Radelet, M . L., & Pierce, G.L.(1985). Raceandprosecutorial discre-tioninhomicide cases.Law and SocietyReview, 19 ,587-621.Ray,J.(1976). Do authoritarians hold authoritarian attitudes?HumanRelations,29 , 307-325.Reiman,J.(1990).Therichget richer andthe poor get prison:Ideology,class,andcriminal ustice.New York:M acmillan.Rombough,S.,& Ventimiglia,J.C. (1981). Sexism: A tri-dimensionalphenomenon.Sex Roles,7 747-755.Rosenberg, M. (19 65).Societyandthe adolescentselfmage.Princeton,NJ: Princeton University Press.Rosenthal, R. (1986).Meta-analytic procedures orsocialresearch.Bev-erly Hills, CA: Sage.Rubin, Z.,&Peplau, L. A. (1975).Who believes ina just w orld?Journalof SocialIssues, 31,265-289.Sears,D. O. (1988). Symbolic racism. In P. A.Katz& D. A.Taylor(Eds.),Eliminating racism:Profilesincontroversy(pp. 5 3-84). New\brk: Plenum.Shapiro, R. Y,&Mahajan, H. (1986). G ender differences in policypref-erences:Asummaryoftrends from the 1960sto the 1980s. PublicOpinionQuarterly, 50,42-61.Sidanius,J.(1976).Furthertests of aSwedish Scale of Conservatism.Reports from the Department ofPsychology, University ofStock-holm, Supplement 36.


21/24

760 PRATTO, SIDANIUS, STALLWORTH, AND MALLESidanius, J. (1988). Race and sentence severity: The case of Americanjustice.JournalofBlackStudies,18 ,273-281.Sidanius, J. (1993). The psychology of group conflict and the dyna micsof oppression: A social dom inance perspective. InW.McGuire & S.Iyengar (Eds.),Current approaches topoliticalpsychology(pp. 183219). Durham , NC: Duke University Press.Sidanius, J., Cling, B. J., & Pratto, F. (1991). Ranking and linking be-havior as a function of sex and gender: An exploration of alternativeexplanations.JournalofSocial Issues,47,131 -149 .Sidanius, J., Devereux, E., & Pratto, F. (1992). A comparison of sym-bolic racism theory and social dominance theoryasexplanations forracial policy attitude s.Journalof SocialPsychology, 132, 377-395.Sidanius, J., & Ekehammar,B .(1979). Political socialization: A multi-variate analysis of Swedish political attitud e and preference data .Eu-ropean Journalo f SocialPsychology,9,265-279.Sidanius, J.,&Ekehammar,B.(1980). Sex-related differences in socio-political ideology.ScandinavianJournal ofPsychology, 21,17-26.Sidanius, J., Liu, J., Pratto, F, & Shaw, J. (1994). Social dominanceorientation, hierarchy-attenuators and hierarchy-enhancers: Socialdominance theory and the criminal justicesystem.Jou rnal of AppliedSocialPsychology,24 ,338-366.Sidanius, J., & Pratto, F. (1993a). The dynamics of social dominanceand the inevitability of oppression. In P. Sniderman & P.E. Tetlock(Eds.),Prejudice,politics, andrace in America today(pp. 173-211).Stanford, CA: Stanford University Press.Sidanius, J., & Pratto, F. (1993b). Racism and support of free-marketcapitalism: A cross-cultural analysis.PoliticalPsychology, 14 , 383-403.Sidanius, J., Pratto, F, & Bobo, L. (in press). Social dominance orien-tation and the political psychology ofgender:A case of invariance?JournalofPersonalityand Social Psychology.Sidanius, J., Pratto, F, &Brief,D. (1993).Group dominanceand thepolitical psychologyo fgender:A cross-culturalcomparison. Unpub-lished m anuscript, U niversity of California atLosAngeles.Sidanius, J., P ratto, F , M artin, M., & Stallworth, L. M . (1991). Con-

sensual racism and career track: Some implications ofsocialdomi-nance theory.PoliticalPsychology, 12 , 691-720.Sidanius, J., Pratto, F , & Mitchell, M. (1994). In-group identification,social dominance o rientation, and differential intergroup social allo-cation.Journa l of SocialPsychology, 134,151-167 .Sidanius, J., Pratto, F, & Rabinowitz, J. (1994). Gender, ethnic status,and ideological asymmetry: Social dominance interpretation.Jour-nalof Cross-CulturalPsychology,25 ,194-216.Sniderman,P.M .,&Tetlock,P.E.(1986a). Symbolic racism: Problemsof motive attribution in political analysis.Journalof Social Issues,42 ,129-150.Sniderman, P. M., & Tetlock, P. E. (1986b). Reflections on Americanracism.Journalo f Social Issues,42 ,173-188.Snyder, M . (1974). The self-monitoring of expressive behavior.Journalof Personalityand SocialPsychology,30 ,526-537.Spence, J. X, Helm reich, R.,&Stapp, J. (1974). The personal attributesquestionnaire: A measure of sex role stereotypes and masculinity-femininity.JSASCatalogof SelectedDocuments inPsychology,4,1 -42.Stacey,B.G., & Green, R.T.(1968).The psychologicalbasesof politicalallegiance among white-collar males.British Journal of Social andClinicalPsychology,7 45-60.

Stacey, B. G., & Green, R. T. (1971). Working-class conservatism: Areview and an e mpirical study.British JournalofSocial and ClinicalPsychology, 10, 10-26.Stone,W.F ,&Russ, R. C. (1976). Machiavellianismastough-minded-ness.JournalofSocialPsychology, 98, 213-220.Super, D. E., & Nevill, D. D. (1985).Th evaluesscale.Palo Alto, CA:Consulting Psychologists Press.Tajfel, H., & Turner, J. C. C. (1986). The social identity theory of in-tergroup behavior. In S. Worchel &W.G. Austin (Eds.), Psychologyof intergroup relations(pp. 7-24 ). Chicago: Nelson-Hall.Teahan, J. E. (1975).Alongitudinal study of attitud e shifts am ong Blackand White police officers.JournalofSocial Issues, 31,47-56.Wilson, G.D.(Ed.). (1973).The psychology of conservatism.San Diego,CA: Academ ic Press.Wilson, G. D., & Patterson, J. R. (1968). A new measure of conserva-tism.British JournalofSocial andClinicalPsychology, 1, 264-269.

Appendix AItems on the Social Dom inance O rientation Scale

1. Some groups of people are simply not the equals of others.2. Some people are just m ore worthy than others.3. This country would be better off if we cared less about how equalall people were.4. Some people are just more deserving than others.5. Itisnot a problem ifsome people havemore ofachance in life thanothers.6. Some people are just inferior to othe rs.7. To getahead in life, itissometimes necessary to step on others.8. Increased economic equality.9. Increased social equality.

10. Equality.11. If people were treated m ore equallywewould have fewer problem sin this country.12. In an ideal world, all nations would be equal.13. We should try to trea t one anoth er as equals as much as possible.(All hum ans should be tre ated equally.)14. Itisimportant thatwetreat other countriesasequals.

All items were measured on a very negative(1) to very positive(7 )scale. Items 8-1 4 were reverse-coded. The version of Item 13in paren-theseswasused in Sam ples 5-1 2. Th e order of items differed from aboveand across samples.


22/24

SOCIAL DOMINANCE ORIENTATIONAppendix B

Items Used on Scales

761

Samples2,3b, 4, 5,6 , 8, 9: Items Comprising the OriginalLegitimizing Myths ScalesAnti-BlackRacism Scale

A Black president."Racial integration."White superiority.Blacks are inherently inferior.Civil rights activists."Anti-ArabRacism Scale

Most of the terrorist s in the world today are A rabs.Historically, Arabs have made important contributions to worldculture."Iraqis have little appreciation for dem ocratic values.People of the Muslim religion tend t o be fanatical.Muslims value peace and love."

Cultural ElitismScaleThe poor cannot appreciatefineart and m usic.No amount of education can make up for the wrong breeding.Qualifications and not personality should determine whether a can-didategetsvotes.The ideal w orldisrun by those who are m ost capable.Western civilization has brought more progress