Political Attitudes and the Ideology of Equality: Differentiating Support for Liberal and Conservative Political Parties in New Zealand

C. Sibley, M. Wilson Political Attitudes and the Ideology of Equality: Differentiating support for liberal and conservative political parties in New Zealand

Chris G. Sibley University of Auckland

Marc S. Wilson Victoria University of Wellington

A new scale summarizing the central and core elements of a social Treaty of Waitangi, and affirmative representation of individual versus group-based entitlement to resource- action policy in the months leading allocations in New Zealand (NZ) is presented. Item content for the Equality up to the 2005 NZ general election Positioning Scale was drawn from qualitative analyses of the discourses of (Johansson, 2004; Kirkwood, Liu, & NZ’s citizens, its political elites, and the media. As hypothesized, equality Weatherall, 2005; Sibley, Robertson, positioning differentiated between Pakeha (NZ European) undergraduates & Kirkwood, 2005). Consistent with who supported liberal versus conservative political parties.People these observations, we argue that who positioned equality as group-based tended to support the Labour ideologies of equality and issues of and Green parties and those who positioned equality as meritocracy who gets what were central to the NZ tended to support the National and NZ First parties. Regression models 2005 election campaign in much the predicting political party support in the two months prior to the 2005 NZ same what that ideologies of national general election demonstrated that the effects of equality positioning security and the war on terrorism were on political party preference were unique, and were not explained by central to election campaigns in the universal (Study 1: Big-Five Personality, Social Dominance Orientation, United States (US) that occurred at Right-Wing Authoritarianism, liberalism-conservatism) or culture-specifi c; around the same time. However, as Jost Study 2: pro-Pakeha ingroup attitudes, support for the symbolic principles (2006) concluded in a recent summary, of biculturalism) indicators derived from other theoretical perspectives. although trends in the ideologies that Taken together, these fi ndings indicate that the Equality Positioning govern voting behavior and political Scale provides a valid and reliable measure that contributes to models attitudes are often commented upon of the psychological and ideological bases of voting behaviour in NZ. anecdotally, systematic quantitative Moreover, our fi ndings suggest that the positioning of equality provided research validating such observations an axis of meaning that aided in the creation and mobilization of public remains limited. This is particularly true opinion regarding resource-allocations, land claims, affi rmative action of the NZ context. programs, and a host of other material issues in the months leading up The present research addresses this to the 2005 NZ election. lacuna and explores the impact of the ideological positioning of equality on the political party preferences of the here has been considerable On the other hand, some definitions majority ethnic group in NZ (Pakeha, or political debate regarding issues emphasize that equality should consider NZ Europeans1) in the months leading Tof who gets what in contemporary group differences, whereby it may be up to the 2005 NZ general election. New Zealand (NZ) society. Such debate necessary to allocate resources on the In order to examine this issue, we is often characterized by an underlying basis of group membership in order present a new measure of individual tension between contrasting ideologies to reduce categorical disadvantages differences in value framing, which of equality. On the one hand, some experienced by some groups within we term Equality Positioning. The defi nitions prescribe that equality should society. Equality Positioning Scale is intended be based on principles of meritocracy As various commentators have to summarize the central and core that emphasize the individual’s freedom suggested, the positioning of equality elements of an ideology of equality and to pursue economic self-interest and the provided a central axis that organized entitlement and is developed for use right to have their worth determined much of the political debate regarding in the NZ socio-political environment based solely upon their individual merit. tax cuts, the role and function of the specifi cally. The items contained in the

• 72 • New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 Political Attitudes & Equality scale are adapted from a variety of NZ- in a close second with 39.1%. The next and a signifi cant number of mainstream specifi c sources, including both our own two highest ranking parties were the NZ (primarily Pakeha) New Zealanders, the earlier qualitative work (e.g., Sibley & First party (another conservative party) policy was dropped, and reference to Liu, 2004; Sibley, Liu, & Kirkwood, with 5.7%, and the Green Party (a liberal 'Closing the Gaps' removed from policy 2006), and the insightful and content party that focuses on environmental initiatives. rich qualitative work on race talk of issues) with 5.3% (Henry, 2005). With A common argument mobilised Nairn and McCreanor (1990, 1991), the support of a number of smaller parties by members of the opposition when Wetherell and Potter (1992) and others, (primarily the Greens), Labour formed arguing against ‘Closing the Gaps’ was as well as political speeches made in the their third consecutive government that government resources should be years preceding the 2005 NZ general – an unprecedented achievement for a allocated on the basis of need rather election (e.g., Brash, 2004). Labour party. than ethnic group membership, and We present two independent Research indicates that support for that the policy implemented by the studies that explore the reliability and the National versus the Labour party Labour government was effectively predictive validity of our measure by differs amongst middle income voters advantaging Maori over other New assessing the degree to which equality (the majority of the NZ population) Zealanders. Dr. Don Brash, the leader of positioning differentiated between because of perceived ideological the National Party at the time of the 2005 support for liberal (Labour and the differences. Support for smaller and election, mobilized similar discourses Greens) versus conservative (National more extreme parties, in contrast, tends to framing equality as meritocracy in his and NZ First) political parties in the be governed more directly by economic Nationhood speech delivered to the two months leading up to the 2005 self-interest. Analyses of a random Orewa rotary club in early 2004. Brash NZ general election. Moreover, we sample of voters conducted in 1997 (2004) argued, for example, that “We examine the degree to which equality indicated, for example, that the belief are one country with many peoples, not positioning provided unique information that people (both oneself and others) have simply a society of Pakeha and Maori predicting participant’s voting the ability to determine their economic where the minority has a birthright to the preferences that could not be explained situation (and the related implication that upper hand, as the Labour Government by (a) universal predictors of political equality is most appropriately defi ned seems to believe.” Here we see the orientation (Big-Five personality, as meritocracy) predicted increased emphasis placed on treating all people Social Dominance Orientation, Right- support for National versus Labour equally as individuals, and the related Wing Authoritarianism, self-labelled (Allen & Ng, 2000). Furthermore, just as implication that not to do so would be liberalism-conservatism), and (b) other Wilson (2004) has shown that National unjust and unfair to other individuals culture-specific constructs (pro-NZ party parliamentarians ascribe less (presumably because they do not European/Pakeha ingroup attitudes, importance to equality than their Labour have ‘a birthright to the upper hand’). support for the symbolic principles counterparts, New Zealand voters tend It is somewhat ironic however, that of biculturalism). Taken together, to show the same pattern of preferences, constructing opposition to policy by these studies provide a snapshot of with Labour party voters endorsing the arguing that it is antiegalitarian (that all the psychological and ideological general concept of equality signifi cantly New Zealanders should be treated the motives associated with political party more than National party voters (Wilson, same) is inconsistent with survey results preferences in NZ during the 2005 2005). suggesting that belief in the importance election campaign. The Labour agenda over the of equality as a general principle is actually positively correlated with Political Ideology in the New last few years has been marked by support for policies based on distributive Zealand Context an egalitarian disposition toward government spending and legislation. justice rules, such as 'Closing the Gaps' NZ, like the United States (US), holds For example, one of Labour’s high (Wilson, 2005). liberal democratic values anchored profi le policies during their term in in ideals of freedom and equality as The Ideology of Equality government in 2000 was the ‘Closing central to nationhood (Liu, 2005). NZ The above analysis of political ideology the Gaps’ policy, which focused on was the fi rst country in the world to and related discourse emphasizes that identifying and addressing areas in introduce universal suffrage, was one terms such as ‘Equality’ can be used which Maori (the indigenous peoples of the first welfare states, and New to refer to distributive justice rules of NZ) were underperforming relative Zealanders have a tradition of protest that emphasize individual merit (the to Pakeha. Maori are disadvantaged against anti-egalitarian regimes. There merit principle) or rules that consider relative to Pakeha on most indicators are two major political parties in NZ, target group membership. As numerous of social and economic well-being; the Labour party (traditionally the major researchers have noted, the value of Maori form 16% of the total population liberal party), and the National party equality has the potential to cut both and 50% of the prison population; (traditionally the major conservative ways depending upon how notions of they earn 16% less income, and their party). In the 2005 NZ general election, fairness are positioned to legitimize life expectancy is 8 years lower (The these two parties achieved a remarkably or oppose the allocation of resources, Social Report, 2005). However, similar endorsement from the nation, outcomes, or other treatments that following concerted expressions of with Labour receiving 41.1% of the consider or are seen to be otherwise opposition from other political parties, nationwide vote, and National coming contingent upon group membership

New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 • 73 • C. Sibley, M. Wilson

(Dovidio & Gaertner, 1996; Kinder & value descriptions can be positioned in the societally elaborated standard Sears, 1981; McConahay & Hough, order to manage ideological dilemmas discourses of equality identified by 1976; McConahay, 1986). At one resulting from discrepancies between Sibley, Liu, and Kirkwood (2006) extreme, equality may be constructed the allocation of material resources and in work assessing opposition to as meritocracy, whereby outcomes existing social inequality (Billig, 1991; affi rmative action programs, we defi ne or treatments that consider group Myrdal, 1944). this measure of equality positioning as membership as a criterion are framed as In a recent study examining assessing the degree to which people biased and potentially discriminatory equality framing in NZ, Sibley, Liu, construe equality as meritocracy, toward individuals who belong to other and Kirkwood (2006), for example, that is, as being based solely on (typically majority) groups (Gamson presented undergraduate students with individual merit; versus the degree to & Modigliani, 1987; Arriola & Cole, an actual proposed change to university which people construe equality as a 1991). Instead, the notion of equality- affi rmative action policy, which was process whereby it may sometimes be as-meritocracy emphasizes individual framed in terms of either remedial necessary to allocate resources on the merit (performance and ability) as the action (which apportioned blame based basis of group membership rather than governing factor that should determine on historical grievances and inequality) individual merit per se in order to redress issues of who gets what. Conversely, in or bicultural partnership (which categorical disadvantages currently situations where there is a gap between emphasized the need for a mutual experienced by ethnic minorities. minority and majority group members partnership between Maori and Pakeha). We argue that equality positioning in terms of social and/or economic Irrespective of framing condition, in the NZ context constitutes an wellbeing, distributive justice rules that Pakeha students endorsed societally ideology, as defi ned by Rokeach (1968, consider minority group membership elaborated ‘standard’ discourses that pp. 123-124), who argued that “an may be a viable means of increasing positioned equality as being based ideology is an organization of beliefs social equality. Presumably, it was solely on individual merit (i.e., grades) and attitudes–religious, political, or observations of this latter type that led and opposed policies that also included philosophical in nature–that is more the Labour government to propose their ethnic group membership as a criteria or less institutionalized or shared ‘Closing the Gaps’ policy initiative in used to govern resource allocations with others, deriving from external the fi rst place. (i.e., targeted scholarships for ethnic authority”. Research in both NZ and Australia minorities). A central theme underlying Political debate in the months has shown that people tend to emphasize such discourses was that scholarships leading up to the 2005 NZ elections notions of equality-as-meritocracy for Maori students were unfair to other centered on a number of issues. Not when expressing opposition toward individuals who did not belong to that surprisingly, much of the debate was affi rmative action and reparation in ethnic group. However, when faced ideological in nature, at least to the natural discourse. Such discourses with the bicultural partnership frame, extent that different political parties typically argue that the consideration many students did make concessions and politicians sought to promote of group membership as a criterion for regarding symbolic aspects of the their own polices and denigrate those determining resource allocations may partnership between Maori and of their opponents through reference constitute preferential treatment or even Pakeha (they nevertheless continued to to ideological differences. To the reverse discrimination (Augoustinos, oppose resource allocations for Maori, extent that the positioning of equality Tuffi n, & Every, 2005; Augoustinos, however). When its comes to realistic as individual- versus group-based Tuffin, & Rapley, 1999; Nairn & issues (in the sense of their relevance provided an axis of meaning that aided McCreanor, 1990, 1991; Sibley & for material outcomes and resource in the creation and mobilization of Liu, 2004; Sibley, Liu, & Kirkwood, allocations), the discursive positioning public opinion in the months leading 2006). Discourses of this type are often of equality forms a culturally sanctioned up to the 2005 NZ election, individual unmarked and position affirmative repertoire for managing debate and differences in equality positioning action as a form of ‘preferential building consensus of opinion (and it should (a) differentiate between support treatment’ that subverts the principle seems, opposition) toward policy that for political parties understood to be of meritocracy and may divide people includes distributive rules contingent more liberal (Labour and the Greens) along ethnic lines (or other relevant upon group membership. and those understood to be more stratifi cation ideologies). As various conservative (National and NZ First). Overview of the Present Studies studies have shown, people also often Specifically, we hypothesize that raise concerns that affi rmative action We aimed to (a) develop a self-report people low in equality positioning will may risk an increase in social unrest measure assessing the degree to which express increased levels of support for and thus exacerbate rather than reduce individuals adopt a prescriptive norm the Labour and Green parties, whereas prejudice (Sibley, Liu, & Kirkwood, in which equality is positioned as people high in equality positioning will 2006; Wetherell & Potter, 1992; see also being based on individual- versus display the opposite trend and express Fraser & Kick, 2000; Riley, 2002, for group-merit, and (b) examine the increased support for the National and similar discursive analyses conducted predictive validity of this measure for NZ First parties. Furthermore, to the in other countries). Such observations understanding differences in political extent that equality positioning exerts emphasize the polemic way in which party preference. Consistent with a unique effect on political attitudes

• 74 • New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 Political Attitudes & Equality that is not due to stable underlying security that is perceived to be provided (items: 13, 15, 22, 23, 24, 25, 26, 28, individual differences, such as openness by policies that emphasize and maintain 32, and 34). Items were rated on a scale to experience or support for tradition, the traditional values, norms, and mores ranging from 0 (strongly disagree) to 6 we hypothesize that (b) equality of the ingroup. (strongly agree). positioning should predict unique Finally, we also controlled for Big-Five personality markers were variance in political party support participant’s self-placed level of assessed using the 50-item version of not explained by other universal and liberalism versus conservatism. Jost the International Personality Item Pool culture-speciﬁ c predictors. (2006) has reported that in the US, asking (Goldberg, 1999). Ten items assessed participants to rate themselves in terms each of the ﬁ ve personality dimensions: STUDY 1 of liberalism versus conservatism using Extraversion, Agreeableness, In Study 1, we first controlled for a simple one-item scale consistently Conscientiousness, Neuroticism, Big-Five personality measures of predicts support for the Democratic and Openness to Experience. Items Extraversion, Agreeableness, versus the Republican party at levels were administered using standard Conscientiousness, Neuroticism, and that exceed r = .90..90. Thus,Thus, itit appearsappears instructions (Goldberg, 1999), and Openness to Experience. We considered that self-placement on this scale were rated on a scale ranging from 0 it parsimonious to control for the effects provides an important and extremely (very inaccurate) through the midpoint of personality on political preference strong predictor of voting preference of 3 (neither inaccurate nor accurate) to before entering more ideologically- in some nations. We therefore deemed 6 (very accurate). oriented constructs given that (a) it important to include and control for Consistent with Jost (2006), personality (in particular Openness to this construct when developing models we included a single item assessing Experience) predicts lower levels of predicting political party support in the participants’ self-reported placement support for more conservative political NZ context. on a scale ranging from 0 (liberal), parties in the US (e.g., Jost, 2006); and through the mid-point of 3 (moderate) (b) personality is most appropriately Method to 6 (conservative). This item was modeled as a casual antecedent of Participants administered using the following ideological attitudes (Duckitt, 2001). Participants were 259, NZ born instructions: ‘Often, people use the We next entered the broad- undergraduate students who participated terms “liberal” or “conservative” to bandwidth ideological attitudes of for partial course credit and who self- describe their political beliefs. How Social Dominance Orientation (SDO; identified as NZ European/Pakeha would you rate yourself in these Pratto, Sidanius, Stallworth, & Malle, (the majority ethnic group in NZ). terms?’ 1994) and Right-Wing Authoritarianism Participants (62 males and 197 females) Equality positioning was measured (RWA; Altemeyer, 1996) as predictors ranged from 18-55 years of age (M = using the 8 items shown in Table 1. of political preference. Together, SDO 20.08, SD = 4.77). This research was These eight items were adapted from and RWA cast a wide ranging net that conducted in July 2005, approximately a variety of sources, primarily recent predicts variation in most domains of two months before the NZ general NZ political speeches (e.g., Brash, prejudice and related intergroup and election which occurred in September 2004) and qualitative responses and political attitudes (Sibley, Robertson, & 2005. All participants were NZ citizens summaries of responses described Wilson, 2006). The geneses of SDO and who were 18 years of age or older, in earlier work on ‘race talk’ in NZ RWA are quite different, however. SDO and were thus eligible to vote in the (e.g., Kirkwood et al., 2005; Nairn arises from perceptions of the social upcoming election. & McCreanor, 1990, 1991; Sibley & environment as a competitive dog-eat- Liu, 2004; Sibley, Liu, & Kirkwood, Materials and Procedure dog world, and predicts domains of 2006; Wetherell & Potter, 1992). Items prejudice motivated by the desire for Participants rated their support for were also revised to give a balanced group dominance. RWA arises from the four largest political parties in number of pro- and contrait statements, perceptions of the social environment as NZ at the time: Labour, National, as the discourses from which items dangerous and threatening, and predicts the Greens, and NZ First on a scale were adapted tended to be protrait in domains of prejudice motivated by the ranging from 0 (strongly oppose) to 6 nature. Items were rated on a scale desire for social conformity, control, (strongly support). These items were ranging from 0 (strongly disagree) and ingroup cohesion (Duckitt, 2001; administered using the following to 6 (strongly agree). Higher scores Duckitt & Sibley, 2007). SDO and instructions: ‘Please indicate how indicated increased endorsement of RWA should therefore tend to predict strongly you support/oppose each of ideologies positioning equality and political conservatism and support the following political parties in the fairness as being based on individual for hierarchically organized social upcoming NZ election.’ (rather than group) merit. As shown structures for different reasons: SDO SDO was measured using 10 in Table 1, exploratory factor analysis for economic and dominance oriented- balanced items from the SDO6 (Pratto indicated that the items assessing reasons associated with free market et al., 1994: items: 1, 2, 3, 4, 5, 10, 12, equality positioning all loaded on a capitalism because such economic 13, 14, and 16). RWA was measured single dimension (factor loadings > conditions foster hierarchical social using a shortened set of 10 balanced .74). Interpretation of the scree plot also structures; RWA for the collective items from Altemeyer’s (1996) scale supported a unidimensional solution,

New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 • 75 • C. Sibley, M. Wilson as the eigenvalues displayed a steeply equality should be determined solely correlated with any of the Big Five decreasing trend that levelled out after by individual merit and that ethnicity factor markers; r’s ranged from -.09 the ﬁ rst value (eigenvalues: 5.02, 1.02, should not be a factor in determining to .12. .43, .39, .33). This unidimensional the fairness of resource allocations Unique Effects of Equality solution accounted for 62.79% of tended to support the National and NZ Positioning on Political Party the total variance in item ratings, and First parties and oppose the Labour Preference as shown in Table 2, was normally and Green parties. Pakeha who tended distributed (kurtosis = -.17, skewness = to agree with statements such as ‘true Hierarchical multiple regression was -.19) and internally reliable (α = .91). equality can only be achieved once used to examine the shared and unique we recognize that some ethnic groups variance in support for political parties Results are currently more disadvantaged than predicted by Big Five personality others’, in contrast, were more likely to markers (Step 1), SDO and RWA (Step Associations Between Equality support the Labour and Green parties, 2), self-labeled liberalism-conservatism Positioning and Political Party and oppose the National and NZ First (Step 3), and equality positioning (Step Preference parties. 4). Descriptive statistics and correlations Equality positioning was also As shown on the top left side of between self-identified Pakeha weakly to moderately positively Table 3, the linear combination of Big respondents’ levels of equality correlated with SDO, r(257) = .20, Five personality markers entered at positioning, SDO, RWA, self-labeled p < .01, and RWA, r(257) = .27, p < Step 1 failed to signiﬁ cantly predict liberalism-conservatism, Big-Five .01. Equality positioning was also variance in support for the Labour party personality markers, and support 2 moderately positively correlated with (adjusted R = .02, F(5,253) = 2.12, p for NZ political parties are shown self-labeled levels of conservatism = .06). The entry of SDO and RWA at in Table 2. Equality positioning was (versus liberalism), r(257) = .29, p Step 2 predicted an additional 11% of negatively correlated with support for < .01. As expected, SDO, RWA, and the variance in support for the Labour the Labour and Green parties, r(257) 2 self-labeled conservatism were also party controlling for the Big Five, ∆R = -.25, p < .01; r(257) = -.30, p < .01, all moderately negatively correlated = .11, Fchange = (2,251) = 15.92, p < .01, respectively; and positively correlated 2 with support for the Labour and Green bringing the adjusted R for the model with support for the National and NZ parties, but positively correlated with up to .12. The direction of these effects First parties, r(257) = .33, p < .01; support the National and NZ First parties indicated that people lower in RWA r(257) = .19, p < .01. These results (see Table 2). Furthermore, equality tended to express increased support indicate that Pakeha who believe that positioning was not significantly for the Labour party. The entry of self-

Table 1. Item content and factor loadings for the Equality Positioning Scale

The Equality Positioning Scale Factor loadings Study 1 Study 2 1. We are all one nation and we should all be treated the same. No one should be entitled to .84 .78 anything more than the rest of us simply because they belong to one particular ethnic group. 2. It is wrong for any one minority to be provided with additional resources because of their ethnicity. Equality means treating all people equally regardless of whether they identify as .83 .83 Maori, NZ European, Asian, or any other ethnic group currently living in New Zealand. 3. We should provide additional resources and opportunities to ethnic minorities with a history of .81 .86 disadvantage in order to promote genuine equality in the future. (r) 4. True equality can only be achieved once we recognize that some ethnic groups are currently .81 .84 more disadvantaged than others and require additional assistance from the government. (r) 5. Given that the economic playing fi eld in New Zealand is not truly level, it is only fair to provide disadvantaged ethnic minorities with additional resources in the here and now so as to make .78 .81 things more equal in the long term. (r) 6. Everyone should be judged solely on their individual merits. People should not be given additional rights simply because of their ethnicity, even if they do belong to a ‘disadvantaged’ .77 .79 group. 7. We are all New Zealanders and the law should not make provision for minority groups .77 .78 because of their ethnicity. 8. The government should devote extra resources to disadvantaged ethnic groups in order to .74 .79 help them overcome the effects of past discrimination and inequality. (r) Note. Study 1: n = 259 self-identifi ed Pakeha, Study 2: n = 150 self-identifi ed Pakeha, (r) = item is reverse coded so that a low score on this item (e.g., a rating of 0) was coded as a high score (e.g., a rating of 6), and vice-versa.

• 76 • New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 Political Attitudes & Equality labeled conservatism at Step 3 failed to for the National party (adjusted R2 = .09, that the origins of support for this party predict additional variance in support F(5,253) = 6.23, p < .01), and the entry differed somewhat from those of the 2 for the Labour party, ∆R < .01, Fchange of SDO and RWA at Step 2 predicted other three political parties. Support = (1,250) = 2.28, p = .13. However, an additional 14% of the variance, for the NZ First party was motivated 2 as predicted, the entry of equality ∆R = .14, Fchange = (2,251) = 24.03, p primarily by the combination of stable positioning at Step 4 predicted an < .01, with people higher in SDO and regularities in personality and high additional 3% of the variance in support RWA expressing increased support for levels of RWA. Equality positioning for the Labour party above and beyond the National party. Together, the Big failed to predict additional variance in 2 all other predictors, ∆R = .03, Fchange = Five, SDO and RWA predicted nearly support for this party. (1,249) = 10.37, p < .01, yielding an a quarter of the variance in support for adjusted R2of .16 for the fi nal model. Put the National Party, adjusted R2 = .23. Discussion - Study 1 another way, equality positioning was Self-labeled conservatism (entered at Study 1 demonstrated that equality uniquely associated with support for the Step 3) also predicted a large portion positioning differentiated between Labour party. The direction of this effect (6%) of additional variance in support Pakeha who supported political parties 2 indicated that people who scored higher for the National party, ∆R = .06, Fchange understood to be more liberal (Labour on equality positioning (and thus tended = (1,250) = 21.94, p < .01, with people and the Greens) or more conservative to construe equality as meritocracy) who considered themselves more (National and NZ First), with those were more likely to oppose the Labour conservative expressing increased low in equality positioning tending to party in the upcoming election. support for the National party. Finally, support the Labour and Green parties As shown in the top right side equality positioning predicted an and those high in equality positioning of Table 3, equality positioning also additional 2% unique variance in support tending to support the National and NZ predicted unique variance in support for the National party in addition to that First parties. Importantly, the differential 2 for the National party, however, the already predicted by other variables, ∆R associations between equality positioning effects of equality positioning were in = .02, Fchange = (1,249) = 8.59, p < .01, and increased opposition to the Labour 2 the opposite direction to those observed yielding an adjusted R of .31 for the and Green parties, and increased support when predicting support for the Labour fi nal model. for the National party were unique. and Green parties. Instead, those who As also shown in Table 3, equality That is, equality positioning predicted scored higher on equality positioning positioning also predicted unique variance in levels of support (or the (and thus tended to construe equality variance in levels of support for the lack thereof) for the Labour, Green, and as meritocracy) were more likely to Green party. These trends were similar National parties that could not also be express support for the National party to those observed when predicting attributed to measures of personality, in the upcoming election. The linear support for Labour. Inspection of the SDO, RWA, or self-labeled liberalism- combination of Big Five personality regression model predicting support conservatism. markers (entered at Step 1) signifi cantly for the NZ First party (presented in the Equality positioning did not, predicted 9% of the variance in support lower right side of Table 3) suggested however, predict unique variance in

Table 2. Study 1: Descriptive statistics and correlations between Equality Positioning, SDO, WA, liberalism-conservatism, Big Five personality markers, and support for NZ political parties two months prior to the 2005 general election.

Scale 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

1. Equality positioning 2. Social Dominance Orientation (SDO) .20* 3. Right-Wing Authoritarianism (RWA) .27* .33* 4. Liberalism-Conservatism .29* .24* .50* 5. Extraversion .01 .04 .09 .10 6. Agreeableness -.03 -.35* -.04 -.08 -.26* 7. Conscientiousness .12 .01 .16* .19* .14* .12 8. Neuroticism -.05 -.12 .02 .08 -.21* .13* .06 9. Openness to Experience -.09 -.27* -.31* -.22* -.28* .25* .01 .02 10. Support for the Labour Party -.25* -.09 -.27* -.19* .15* .00 .04 -.12 -.11 11. Support for the Green Party -.30* -.25* -.32* -.33* -.02 .07 -.07 -.00 .17* .28* 12. Support for the National Party .33* .35* .40* .44* -.06 -.10 .20* -.02 -.17* -.38* -.31* 13. Support for the NZ First Party .19* .23* .35* .23* .02 -.02 .05 -.03 -.24* -.17* -.13* .33* M 3.53 2.63 3.06 3.47 2.44 4.39 3.39 3.22 3.93 3.31 3.02 3.13 2.00 SD 1.19 .84 .94 1.27 .98 .71 .81 .97 .72 1.44 1.63 1.62 1.49 Skewness -.19 .31 .26 -.01 .35 -.72 .05 .24 -.06 -.26 -.10 -.24 .11 Kurtosis -.17 -.16 -.09 -.22 -.15 1.17 -.23 -.25 -.14 -.20 -.59 -.73 -.54 α .91 .84 .83 - .90 .82 .80 .87 .80 - - - -

Note. * = p < .05; n = 259 self-identiﬁ ed Pakeha university students for all correlations. Scores for all variables ranged from 0 to 6.

New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 • 77 • C. Sibley, M. Wilson

Table 3. Study 1: Hierarchical regression analyses predicting support for NZ political parties two months prior to the 2005 general election.

Support for Labour Party Support for National Party β (SE) t R2 adj. β (SE) t R2 adj.

Step 1 Extraversion .09 (.09) 1.42 -.16 (.10) -2.83* Agreeableness .06 (.13) .87 -.05 (.13) -.78 Conscientiousness .11 (.11) 1.85 .14 (.11) 2.52* Neuroticism -.12 (.09) -1.94 -.06 (.09) -1.01 Openness to Experience -.22 (.13) -3.37* .02 -.05 (.13) -.81 .09*

Step 2 SDO -.01 (.11) -.08 .18 (.12) 2.92* RWA -.27 (.11) -3.83* .12* .15 (.11) 2.27* .23* Step 3 Liberalism-Conservatism -.06 (.08) -.93 .13 .26 (.08) 4.15* .29* Step 4 Equality positioning -.20 (.07) -3.22* .16* .16 (.08) 2.93* .31*

Support for Green Party Support for NZ First Party β (SE) t R2 adj. β (SE) t R2 adj. Step 1 Extraversion .03 (.11) .49 -.04 (.10) -.61 Agreeableness .00 (.15) .07 -.07 (.14) -1.02 Conscientiousness .00 (.12) .02 .00 (.11) -.04 Neuroticism -.00 (.10) -.08 -.04 (.09) -.66 Openness to Experience .06 (.15) .86 .02 -.15 (.14) -2.26* .05*

Step 2 SDO -.11 (.13) -1.70 .11 (.12) 1.60 RWA -.13 (.12) -1.86 .11* .24 (.11) 3.31* .13* Step 3 Liberalism-Conservatism -.18 (.09) -2.62* .14* .04 (.08) .55 .13 Step 4 Equality positioning -.19 (.08) -3.09* .17* .08 (.08) 1.23 .13 Note. Analyses were based on data from 259 self-identifi ed Pakeha university students; standardized regression coeffi cients (β), standard errors (SE), and t-values displayed for the Step 4 model. The adjusted R2 (R2 adj.) is displayed for each step, the signifi cance of the adjusted R2 indicates whether the linear combination of variables entered at that step predicted signifi cant additional variance in the dependent measure of interest (rather than whether the overall model was signifi cant at each step). * p < .05.

support for the NZ First party. Levels STUDY 2 party preference (namely SDO, RWA, of support and opposition for this party The results reported in Study 1 indicate Big Five personality markers, and were instead predicted solely by Big that equality positioning reflects a self-labeled liberalism-conservatism), Five personality markers (primarily low content rich and socially elaborated Study 2 controlled for a different set Openness to Experience) and high levels ideology governing how people think of predictors that have been shown to of RWA. More generally, these results about meritocracy and the entitlement of relate to attitudes toward ethnic groups point to an intriguing pattern suggesting group versus individual-based resource- and bicultural policy in NZ specifi cally that a substantial portion of the degree to allocations. This ideology is not (in addition to also controlling for SDO which people supported or opposed the reducible to more universal and broad- and RWA). Namely, we controlled for National and NZ First parties was due bandwidth measures of personality or pro-majority (i.e., pro-Pakeha) ethnic to individual differences in personality ideological attitudes, and the inclusion group attitudes and attitudes toward the and global ideological attitudes indexed of equality positioning in models of symbolic principles of biculturalism. by SDO and RWA. Variation in the political party preference in the months Pro-majority ethnic group attitudes degree to which people supported or leading up to the 2005 NZ elections have been shown to predict increased opposed the Labour and Green parties, signifi cantly increased the predictive levels of anti-minority ethnic group in contrast, was mostly unrelated to power of such models (for the Labour, attitudes in both NZ (Duckitt & personality, and was predicted primarily Green and National parties). Our second Parra, 2004) and elsewhere (Duckitt, by ideologically-oriented constructs (cf. study aimed to replicate and further Callaghan, & Wagner, 2005). Given Allen & Ng, 2000; Jost, 2006). examine the unique effects of equality that equality positioning is distinct positioning on political party preference from more generalized pro-ingroup in the months leading up to the 2005 ethnic attitudes, it should predict unique NZ general election. Whereas Study 1 variance in political party preference controlled for arguably universal (non- once such attitudes have been controlled. culture-specifi c) predictors of political Attitudes toward the symbolic principles

• 78 • New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 Political Attitudes & Equality of biculturalism, in contrast, assess the NZ citizens who were 18 years of age total variance in item ratings, and was degree to which people are supportive or older, and were thus eligible to vote normally distributed (kurtosis = -.31, of the incorporation of Maori values in the upcoming election. skewness = -.24) and internally reliable and culture into mainstream (primarily (α = .92). Materials Pakeha) NZ culture and national identity. All items were rated on a scale SDO was first measured using the Sibley and Liu (2004, in press); Sibley ranging from 0 (strongly disagree) to 6 full 16-item SDO scale (Pratto et al., et al. (2005); and Sibley, Wilson, & 6 (strongly agree). Political party support 1994). RWA was measured using a Robertson (2007), have argued that was assessed using the scales described shortened set of 16 balanced items from attitudes toward the symbolic aspects of in Study 1. biculturalism are distinct from attitudes Altemeyer’s (1996) scale (items: 8, 10, 12, 13, 15, 22, 23, 24, 25, 26, 28, 29, toward material interests relating to Results bicultural policy, at least in contexts 30, 31, 32, and 34). Positive intergroup like New Zealand where biculturalism attitudes toward Pakeha were assessed Associations Between Equality is part of the national ideology for using 8 items from Duckitt and Parra’s Positioning and Political Party governance. If attitudes toward the (2004) NZ ethnic attitude scale. Preference symbolic and resource-speciﬁ c aspects Attitudes toward the symbolic Correlations between self-identified of biculturalism are indeed distinct from principles of biculturalism were assessed Pakeha respondents’ levels of equality one another, then equality positioning using the 5-item scale developed by positioning, SDO, RWA, support for (which assesses attitudes toward the Sibley et al. (2005). This scale assessed biculturalism in principle, pro-majority allocation of material resources) should that degree to which people were ethnic group attitudes and support for predict additional variance in political supportive of the incorporation of Maori NZ political parties are shown in Table party preference once attitudes toward values and culture into mainstream 4. Consistent with Study 1, and as the symbolic principles of biculturalism (primarily Pakeha) NZ culture and shown in Table 4, equality positioning have been controlled. national identity. was negatively correlated with support Equality positioning and political for the Labour and Green parties, Method party support were measured using the r(144) = -.50, p < .01; r(144) = -.33, p < .01, respectively; and positively Participants and Procedure scales described in Study 1. As shown correlated with support for the National Participants were 146 NZ born in Table 1, exploratory factor analysis and NZ First parties, r(144) = .38, p undergraduate students who participated indicated that the items assessing < .01; r(144) = .31, p < .01. As with for partial course credit and who self- equality positioning all loaded on a Study 1, these results indicate that the identified as NZ European/Pakeha. single dimension (factor loadings > equality positioning scale differentiated Participants (53 males and 93 females) .78). Interpretation of the Scree plot between Pakeha who supported political ranged from 18-55 years of age (M = supported a unidimensional solution, parties understood to be more liberal 21.38, SD = 4.55). As with Study 1, as the eigenvalues displayed a steeply (Labour and the Greens) or more this research was conducted in July decreasing trend that levelled out after conservative (National and NZ First) 2005, approximately two months before the ﬁ rst value (eigenvalues: 5.36, .95, political parties. the general election which occurred in .41, .37, .31). This unidimensional September 2005. All participants were solution accounted for 65.74% of the

Table 4. Study 2: Descriptive statistics and correlations between Equality Positioning, SDO, RWA, pro-majority ethnic group attitudes, biculturalism in principle, and support for NZ political parties two months prior to the 2005 general election.

Scale 1. 2. 3. 4. 5. 6. 7. 8. 9. 1. Equality positioning 2. Social Dominance Orientation (SDO) .26* 3. Right-Wing Authoritarianism (RWA) .14 .44* 4. Support for biculturalism in principle -.38* -.40* -.18* 5. Pro-majority ethnic group attitudes .55* .46* .33* -.51* 6. Support for the Labour Party -.50* -.21* -.26* .31* -.34* 7. Support for the Green Party -.33* -.29* -.36* .30* -.37* .52* 8. Support for the National Party .38* .30* .25* -.29* .43* -.50* -.39* 9. Support for the NZ First Party .31* .32* .25* -.31* .43* -.19* -.19* .47*

M 3.65 1.60 1.83 4.03 3.12 3.37 3.11 2.84 2.26 SD 1.22 .86 .82 1.24 .71 1.32 1.56 1.56 1.50 Skewness -.19 .33 .28 -.54 -.04 -.49 -.42 -.21 .12 Kurtosis -.32 -.32 .41 .24 .15 -.05 -.21 -.71 -.47 α .92 .90 .89 .87 .72 - - - -

Note. * = p < .05; n = 146 self-identiﬁ ed Pakeha university students for all correlations. Scores for all variables ranged from 0 to 6.

New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 • 79 • C. Sibley, M. Wilson

Equality positioning was in principle (Step 3), and equality likely to oppose the Labour party in the moderately positively correlated with positioning (Step 4). upcoming election. SDO, r(144) = .26, p < .01, however As shown on the top left side of As shown in the top right side it was not signifi cantly correlated with Table 5, the linear combination of SDO of Table 5, equality positioning also RWA in this sample, r(144) = .14, p and RWA entered at Step 1 predicted predicted unique variance in support = .10. Equality positioning was also 7% of the variance in support for for the National party, however, the moderately negatively correlated with the Labour party (adjusted R2 = .07, effects of equality positioning were in support for the symbolic principles F(2,143) = 6.03, p < .01). The entry of the opposite direction to those observed of biculturalism, r(144) = -.38, p < pro-majority ethnic group attitudes at when predicting support for the Labour .01. Equality positioning was strongly Step 2 predicted an additional 6% of and Green parties. Those who scored positively correlated with Pakeha the variance in support for the Labour higher on equality positioning (and pro-ingroup ethnic attitudes, r(144) = 2 party, ∆R = .06, Fchangechange = (1,142) = thus tended to construe equality as .55, p < .01. As expected, SDO, RWA, 9.55, p < .01, bringing the adjusted R2 meritocracy) were more likely to and pro-majority ethnic attitudes were for the model up to .12. The entry of express support for the National party also moderately negatively correlated support for the symbolic principles of in the upcoming election. The linear with support for the Labour and Green biculturalism at Step 3 predicted an combination of SDO and RWA entered parties, and positively correlated with additional 3% of the variance, ∆R2 = at Step 1 predicted 9% of the variance in support the National and NZ First .03, Fchange = (1,141) = 4.33, p < .05, support for the National party (adjusted parties, whereas support for the symbolic bringing the adjusted R2 for the model R2 = .09, F(2,143) = 8.49, p < .01). principles of biculturalism displayed the up to .14. Finally, as predicted, the The entry of pro-majority ethnic group opposite pattern of correlations (see entry of equality positioning at Step attitudes at Step 2 predicted an additional Table 4). 4 predicted an additional 14% of the 10% of the variance in support for the 2 Unique Effects of Equality variance in support for the Labour party National party, ∆R = .10, Fchangechange = Positioning on Political Party above and beyond all other predictors, (1,142) = 17.22, p < .01, bringing the 2 2 Preference ∆R = .14, Fchange = (1,140) = 27.60, p < adjusted R for the model up to .19. .01, yielding an adjusted R2 of .28 for The entry of support for the symbolic Hierarchical multiple regression was the fi nal model. As with Study 1, the principles of biculturalism at Step 3 used to examine the shared and unique direction of this effect indicated that failed to signifi cantly predict additional variance in political party preferences people who scored higher on equality variance, ∆R2 < .01, F = (1,141) = predicted by SDO and RWA (Step 1), change positioning (and thus tended to construe .76, p = .39. Finally, as predicted, the pro-majority ethnic group attitudes equality as meritocracy) were more entry of equality positioning at Step 4 (Step 2), support for biculturalism

Table 5. Study 2: Hierarchical regression analyses predicting support for NZ political parties two months prior to the 2005 general election.

Support for Labour Party Support for National Party β (SE) t R2 adj. β (SE) t R2 adj. Step 1 SDO .03 (.13) .30 .08 (.17) .87 RWA -.19 (.13) -2.39* .07* .10 (.16) 1.22 .09*

Step 2 Pro-majority ethnic group attitudes .03 (.18) .29 .12* .22 (.22) 2.14* .19* Step 3 Biculturalism in principle .13 (.09) 1.55 .14* -.05 (.11) -.56 .19 Step 4 Equality positioning -.45 (.09) -5.25* .28* .21 (.11) 2.34* .21*

Support for Green Party Support for NZ First Party β (SE) t R2 adj. β (SE) t R2 adj. Step 1 SDO .04 (.16) -.39 .11 (.16) 1.17 RWA -.26 (.16) -3.17* .14* .08 (.16) 1.00 .11*

Step 2 Pro-majority ethnic group attitudes -.11 (.22) -1.05 .19* .26 (.22) 2.48* .19* Step 3 Biculturalism in principle -.11 (.11) 1.25 .19 -.08 (.11) -.92 .20 Step 4 Equality positioning -.18 (.11) -2.05* .21* .10 (.11) 1.16 .20

Note. Analyses were based on data from 146 self-identifi ed Pakeha university students; standardized regression coeffi cients (β), standard errors (SE), and t-values displayed for the Step 4 model. The adjusted R2 (R2 adj.) is displayed for each step, the signifi cance of the adjusted R2 indicates whether the linear combination of variables entered at that step predicted signifi cant additional variance in the dependent measure of interest (rather than whether the overall model was signifi cant at each step). * p < .05.

• 80 • New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 Political Attitudes & Equality predicted additional variance (3%) in with Verkuyten (2005), who observed (the merit principle) or rules that support for the National party above that majority group members who showed considered target group membership, and beyond all other predictors, ∆R2 strong ingroup identiﬁ cation were more or to use Rokeach’s (1973) parlance:

= .03, Fchange = (1,140) = 5.50, p = .02, supportive of assimilation-type policies Freedom versus Equality. Our measure yielding an adjusted R2of .21 for the which frame everyone as being equal tapped specific elements of a wider fi nal model. as individuals and thus de-emphasize social representation of these contrasting As shown on the lower left side recognition of minority group identities. ideological positions, both of which are of Table 5, equality positioning also Furthermore, equality positioning exerted framed as promoting ‘Equality’, but may predicted unique variance in levels unique effects on political preference lead to quite different outcomes. of support for the Green party. These that could not be explained by such pro- The Equality Positioning Scale results were similar to those predicting ingroup evaluations. (shown in Table 1) provided an internally support for Labour. Inspection of the reliable and normally distributed regression model predicting support General Discussion measure assessing the degree to which for the NZ First party (presented in the In two studies, equality positioning New Zealanders believe that equality lower right side of Table 5) suggested differentiated between people who should be determined on the basis of that the origins of support for this party supported liberal versus conservative group versus individual merit. The differed from those of the other three political parties in NZ. The differential scale was balanced, and analyses political parties. In short, Pakeha who associations between equality indicated that con- and protrait item supported NZ First were motivated positioning and political party support aggregates were strongly negatively primarily by RWA and SDO combined were also unique, and could not be correlated prior to the contrait items with high levels of pro-ingroup ethnic attributed to more universal (personality, being reverse scored (r’s of around -.70). attitudes. Equality positioning failed to Social Dominance Orientation, Right- Furthermore, all items loaded strongly predict additional variance in support for Wing Authoritarianism, self-labelled on a single underlying dimension. These NZ First beyond that already predicted values) or culture-specifi c (pro-Pakeha fi ndings suggest that the positioning of by the SDO, RWA, and pro-majority ingroup attitudes, support for the group- versus individual-based merit ethnic group attitudes. symbolic principles of biculturalism) reflect opposing aspects of a single constructs. It is important to note that underlying ideological continuum, rather Discussion - Study 2 these analyses examined the effects than distinct albeit correlated belief Pakeha who were low in equality of personality and ideology within dimensions, at least in the NZ context. positioning were more likely to support the individual. It is not t thehe c casease thatthat Thus, it is unlikely that New Zealanders the Labour and Green parties whereas some people’s voting preferences are will tend to endorse allocations on the those high in equality positioning governed solely by personality, whereas basis of both individual and group merit; tending to support the National party. other people are governed solely by rather our results indicate that these two These differential associations were ideology. Rather, our results indicate positions are mutually exclusive, and also unique. That is, equality positioning that ideology and personality effects on individuals will be more likely to adopt predicted variance in levels of support (or voting behaviour operate simultaneously one of these ideological positions at the the lack thereof) for the Labour, Green, within the individual to predict support expense of the other. and National parties that could not be for different political parties in the NZ It is interesting, however, that attributed to measures of SDO, RWA, environment. discourses positioning equality as pro-majority ethnic group attitudes, or The fi ndings provide good evidence group-based (such as those refl ected in attitudes toward the symbolic principles for the convergent and discriminant the contrait items contained in the scale) of biculturalism. As with Study 1, validity of our measure of equality seem to occur relatively infrequently in equality positioning did not, however, positioning and indicate that the scale public discourse and media (and should predict unique variance in support for provides important information that presumably, be less heavily anchored the NZ First party. Support for NZ First increases the accuracy of models and more malleable to change; Liu was instead predicted solely by SDO, predicting voting behaviour in NZ. The & Sibley, 2006). Indeed, qualitative RWA, and pro-majority ethnic group strong and unique predictive utility of research in NZ has commented upon attitudes. equality positioning in this context most the scarcity of socially elaborated As expected, Pakeha who positioned likely occurred because the ideological discourses that may be used to promote equality as meritocracy under one nation positioning of equality as individual- biculturalism and resource-allocations where all people are treated as individuals versus group-based provided an axis of on the basis of ethnic group membership also tended to demonstrate more positive meaning that aided in the creation and in everyday talk (e.g., Kirkwood et al., pro-ingroup evaluations, as indexed mobilization of public opinion in the 2005; Sibley & Liu, 2004; Tuffi n, Praat, by agreement with Likert items from months leading up to the election. It & Frewin, 2004). Nevertheless, when Duckitt and Parra’s (2004) scale, such seems that one of the main (perceived) asked to respond to items assessing as “NZ Europeans/Pakeha through hard ideological distinctions between Labour such discourses people can readily work and perseverance developed this and National in this election was the do so, and the manner in which they country and are entitled to their greater emphasis placed on distributive justice respond is strongly consistent with their material prosperity”. This is consistent rules that emphasize individual merit level of agreement with items assessing

New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 • 81 • C. Sibley, M. Wilson more readily apparent discourses of ethnic group), perceived themselves to Although correlational in nature, equality positioning to which they have be more conservative (versus liberal), these results are consistent with previous presumably had greater exposure: those and opposed the incorporation of Maori research examining the ideologies positioning equality as meritocracy. values and culture into mainstream surrounding affi rmative action in the The Equality Positioning Scale (primarily Pakeha) NZ culture and US (e.g., Hayley & Sidanius, 2006; was developed based upon qualitative national identity at the symbolic level. Federico & Sidanius, 2002). Our results analyses of NZ discourse, with the aim However, equality positioning was not indicate that both the competitive-driven of providing a measure of ideological signifi cantly associated with measures motivation for group dominance and positioning that was relevant and of personality, indicating that the degree superiority indexed by SDO and the directly applicable to the NZ context. to which people were extroverted threat-driven motivation for collective The positioning of equality appears to be versus introverted, agreeable versus security and social cohesion indexed a relatively universal ideology, however, disagreeable, emotionally labile versus by RWA may predict the endorsement and many studies have identifi ed similar emotionally stable, conscientious versus of ideologies that support the status discourses in measures of racism, sexism disorganized, or open versus closed quo and that facilitate the maintenance and political ideology in other countries to new experiences, was unrelated to of existing social inequality. Hence, (Sears & Henry, 2005; McConahay, the degree to which they adopted a we theorize that the emphasis placed 1986). For example, the symbolic and prescriptive belief positioning equality on the need for one nation under one modern racism scales developed in the as being based on individual (rather than common law where all people are US contain a blend of items assessing group) merit. treated ‘equally’ by people who are high belief in meritocracy and individualism Why are some people more likely in SDO occurs because such discourses (akin to those contained in our scale) and to position equality as meritocracy, and facilitate the maintenance of hierarchical items that assess generalized negative what implications does the adoption social structures and social inequality affect toward minority groups (namely of such prescriptive beliefs have (Sidanius & Pratto, 1999). This is not African Americans). Consider, for for intergroup relations in NZ? The to say that all people who position example, the similarity between items aforementioned pattern of associations equality as meritocracy are motivated contained in the Equality Positioning suggests that individual differences in by racism or group-dominance related Scale and the following item from the equality positioning are the product of motives (Sibley & Liu, 2004). It does, Symbolic Racism Scale developed for ideologically-based goals and motives, however, suggest that (a) majority group use in the US: “It’s really a matter of rather than being directly infl uenced by individuals who are motivated by such some people not trying hard enough; personality. Ostensibly, agreement with goals will be among those who are most if Blacks would only try harder they items from the Equality Positioning ardent in positioning equality in this could be just as well off as Whites”. Scale that started with statements such manner, and (b) that shared endorsement Such observations emphasize the cross- as: ‘We are all one nation and we should of such ideologies by a wide segment of cultural generality of ideologies of all be treated the same…’ and ‘We are all society may engender systemic levels of equality and their implications for New Zealanders…’ refl ect a discourse social inequality. intergroup and, in particular, race of equality for all under one nation and relations. They suggest that the one common law regardless of ethnicity, Conclusion positioning of equality as meritocracy class or creed. However, meta-analytic In conclusion, we presented a new in the NZ context may be similar in averages combining the effect sizes measure of equality positioning its form and function to discourses of from both studies indicated that Equality and demonstrated that our measure symbolic and modern racism identifi ed Positioning was moderately positively differentiated between support for liberal in the US. The positioning of equality correlated with both SDO (weighed r and conservative political parties in NZ, may function as a mechanism that = .25) and RWA (weighed r = .23).2 As even after a host of other universal and justifi es and maintains social inequality noted earlier, SDO and RWA provide culture-specific predictors had been between ethnic groups in NZ in much reliable indicators of two distinct dual considered. We contend that equality the same way that symbolic and modern processes that form the motivational positioning refl ects a socially elaborated racism are theorized to legitimize social basis for many different forms of discourse that is anchored within inequality between African Americans prejudice, negative intergroup attitudes, broader notions of liberal democracy and Whites in the US. and system justifying ideologies and freedom for all (Sibley, Liu, & (Duckitt, 2001). Ironically, people who Kirkwood, 2006). The culture-specifi c Causes and Consequences of tended to agree with statements from positioning of equality-as-meritocracy Equality Positioning the SDO scale refl ecting the belief that may have allowed political elites and To whom does the positioning of social inequality is not really such a bad their constituents to express opposition equality as meritocracy most appeal? state of affairs, such as: ‘Its OK if some to resource allocations favoring minority Our fi ndings indicate people high in groups have more of a chance in life than groups in the NZ context while still SDO and RWA were more likely to others’ and ‘Some groups of people are maintaining discourses of plausible position equality as meritocracy, as were simply inferior to other groups’, also deniability in much the same way as Pakeha who expressed more positive tended to be more likely to position symbolic racism is thought to operate pro-ingroup attitudes (and presumably equality as meritocracy. within the United States (Sears & Henry, identified more strongly with their

• 82 • New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 Political Attitudes & Equality

2005; see also Liu & Mills, 2006). In this Duckitt, J., Callaghan, J., & Wagner, C. Kirkwood, S., Liu, J.H., & Weatherall, A. way, the NZ-specifi c (re-)formulation of (2005). Group identifi cation and outgroup (2005). Challenging the standard story equality positioning assessed here may attitudes in four South African ethnic of indigenous rights in Aotearoa/New have provided an axis of meaning that groups: A multidimensional approach. Zealand. Journal of Community and Personality and Social Psychology Applied Social Psychology, 15, 1-13. aided in the creation and mobilization Bulletin, 31, 633-646. of public opinion regarding resource- Liu, J.H. (2005). History and identity: allocations, land claims, affirmative Duckitt, J., & Parra, C. (2004). Dimensions A system of checks and balances for of group identification and out-group Aotearoa/New Zealand. In J. H. Liu, T. action programs, and a host of other attitudes in four ethnic groups in New McCreanor, T. McIntosh, & T. Teaiwa, material issues leading up to the 2005 Zealand. Basic and Applied Social (Eds.), New Zealand identities: Departures NZ general election (Liu, 2005; Sibley et Psychology, 26, 237-247. and destinations, pp. 69-87. Wellington, al., 2005) – and, as our results indicate, Duckitt, J., & Sibley, C. G. (2007). Right-Wing NZ: Victoria University Press. the mobilization of public opinion in Authoritarianism, Social Dominance Liu, J.H., & Mills, D. (2006). Modern this way exerted unique effects on Orientation, and the dimensions of racism and market fundamentalism: political party preference that were generalized prejudice. European Journal The discourses of plausible deniability not reducible to universal and broad- of Personality, 21, 113-130. and their multiple functions. Journal bandwidth measures of personality, Duckitt, J., Wagner, C., du Plessis, I., & of Community and Applied Social ideological attitudes, or attitudes toward Birum, I. (2002). The psychological bases Psychology, 16, 83-99. biculturalism and ethnic group relations of ideology and prejudice: Testing a dual Liu, J.H., & Sibley, C.G. (2006). Differential in the NZ context. process model. Journal of Personality effects of societal anchoring and personal and Social Psychology, 83, 75-93. importance in determining support or opposition to (bi)cultural diversity References Federico, C.M., & Sidanius, J. (2002). Racism, ideology, and affi rmative action revisited: in New Zealand. Papers on Social Allen, M.W., & Ng, S.H. (2000). Self- The antecedents and consequences of Representations, 15, 1-15. interest, economic beliefs, and political “principled objections” to affirmative McConahay, J. B. (1986). Modern racism, party preference in New Zealand. action. Journal of Personality and Social ambivalence, and the Modern Racism Political Psychology, 21, 323-345. Psychology, 82, 488-502. Scale. In J. Dovidio, & S. Gaertner (Eds.), Altemeyer, B. (1996). The authoritarian Fraser, J., & Kick, E. (2000). The Prejudice, discrimination, and racism spectre. London: Harvard University interpretative repertoires of whites on (pp. 91-125). Orlando, FL: Academic Press. race-targeted policies: Claims making Press. Arriola, K.R.J., & Cole, E.R. (2001). of reverse discrimination. Sociological McConahay, J. B., & Hough, J., Jr. (1976). Framing the affi rmative-action debate: Perspectives, 43, 13-28. Symbolic racism. Journal of Social Issues, Attitudes toward out-group members and Gamson, W.A., & Modigliani, A. (1987). The 32, 23-45. White identity. Journal of Applied Social changing culture of affi rmative action. In Myrdal, G. (1944). An American dilemma: Psychology, 31, 2462-2483. R.G. Braungart & M.M. Braungart (Eds.), The Negro problem and modern Augoustinos, M., Tuffi n, K., & Every, D. Research in political sociology (Vol(Vol 3.3. ppp.p. democracy. New York: Harper & (2005). New racism, meritocracy and 137-177). Greenwich, CT: JAI. Brothers. individualism: Constraining affi rmative Goldberg, L.R. (1999). A broad-bandwidth, Nairn, R.G., & McCreanor, T.N. (1990). action in education. Discourse and public domain, personality inventory Insensitivity and hypersensitivity: An Society, 16, 315-340. measuring the lower-level facets of several imbalance in Pakeha accounts of racial Augoustinos, M., Tuffin, K., & Rapley, fi ve-factor models. In I. Mervielde, I. confl ict. Journal of Language and Social M. (1999). Genocide or a failure to Deary, F. De Fruyt, & F. Ostendorf (Eds.), Psychology, 9, 293-308. gel? Racism, history and nationalism in Personality Psychology in Europe, Vol. Nairn, R.G., & McCreanor, T.N. (1991). Australian talk. Discourse and Society, 7 (pp. 7-28). Tilburg, The Netherlands: Race talk and common sense: Patterns 10, 351-378. Tilburg University Press. in Pakeha discourse on Maori/Pakeha Billig, M. (1991). Ideology and opinion: Hayley, H., & Sidanius, J. (2006). The positive relations in New Zealand. Journal of Studies in rhetorical psychology. London: and negative framing of affirmative Language and Social Psychology, 10, Sage Publications. action: A group dominance perspective. 245-262. Brash, D. (2004, January 27th). Nationhood. Personality and Social Psychology Riley, S.C.E. (2002). Constructions of Retrieved on December 19th, 2006, Bulletin, 32, 656-668. equality and discrimination in professional from http://national.org.nz/article. Henry, D. (2005). General Election: Results men’s talk. British Journal of Social aspx?articleid=1614 of the offi cial count. New Zealand Gazette, Psychology, 41, 443-461. Dovidio, J. F., & Gaertner, S. L. (1996). October 2005, Issue 170. Rokeach, M. (1968). Beliefs, attitudes, and Affi rmative action, unintentional racial Johansson, J. (2004). Orewa and the rhetoric of values. San Francisco: Jossey-Bass. biases, and intergroup relations. Journal illusion. Political Science, 56, 101-119. Rokeach, M. (1973). The nature of human of Social Issues, 52, 51-75. Jost, J.T. (2006). The end of the end of values. New York: Free Press. Duckitt, J. (2001). A dual-process cognitive- ideology. American Psychologist, 61, Sears, D.O., & Henry, P.J. (2005). Over motivational theory of ideology and 651-670. thirty years later: A contemporary look prejudice. In M. P. Zanna, (Ed.), Advances Kinder, D.R., & Sears, D.O. (1981). Prejudice at symbolic racism. In M. P. Zanna, in experimental social psychology (Vol. and politics: symbolic racism versus (Ed.), Advances in experimental social 33, pp. 41-113). New York: Academic racial threats to the good life. Journal psychology (Vol. 37, pp. 95-150). New Press. of Personality and Social Psychology, York: Academic Press. 40, 414-431.

New Zealand Journal of Psychology Vol. 36, No. 2, July 2007 • 83 • C. Sibley, M. Wilson

Sibley, C.G., & Liu, J.H. (2004). Attitudes Wilson, M.S. (2004).Values and political towards biculturalism in New Zealand: ideology: Rokeach’s two-value model Address for correspondence: Social dominance and Pakeha attitudes in proportional representational towards the general principles and environment. New Zealand Journal of Chris Sibley, resource-specifi c aspects of bicultural Psychology, 33, 155-162. Department of Psychology, University of Auckland, policy. New Zealand Journal of Wilson, M.S. (2005). A social-value analysis Psychology, 33, 88-99. of postmaterialism. Journal of Social Private Bag 92019, Sibley, C.G., & Liu, J.H. (in press). New Psychology, 145, 209-224. Auckland, Zealand = bicultural? Implicit and New Zealand. explicit associations between ethnicity E-mail: [email protected] and nationhood in the New Zealand Notes context. European Journal of Social 1. There is currently considerable Psychology. debate in New Zealand regarding Sibley, C.G., Liu, J.H., & Kirkwood, S. the most appropriate name for New (2006). Toward a social representations Zealanders of European descent. theory of attitude change: The effect of Although New Zealand European message framing on general and specifi c is the most popular term (Liu et attitudes toward equality and entitlement. al., 1999), Pakeha is the term that New Zealand Journal of Psychology, most strongly implies a relationship 35, 3-13. with Maori and hence seems most appropriate for this paper. Sibley, C.G., Robertson, A., & Kirkwood, S. (2005). Pakeha attitudes toward the 2. It is worth noting that analyses of symbolic and resource-specifi c aspects the combined data from studies one of bicultural policy in New Zealand: and two indicated that SDO was The legitimizing role of collective guilt signifi cantly positively correlated for historical injustices. New Zealand with the protrait item aggregates Journal of Psychology, 34, 171-180. assessing equality positioning (i.e., the aggregate of those items where Sibley, C.G., Robertson, A., & Wilson, M.S. agreement indicated the positioning (2006). Social Dominance Orientation of equality as meritocracy) and and Right-Wing Authoritarianism: signifi cantly negatively correlated Additive and interactive effects. Political with contrait item aggregates (i.e., Psychology, 27, 755-768. the aggregate of those equality Sibley, C.G., Wilson, M.S., & Robertson, A. items where agreement indicated (2007). Differentiating the motivations the positioning of equality as group- and justifi cations underlying individual based), r(403) = .12, p = .02; r(403) differences in Pakeha opposition to = -.19, p < .01, respectively. Thus, bicultural policy: Replication and the association between SDO and extension of a predictive model. New equality positioning was not driven Zealand Journal of Psychology, 36, solely by the association between 25-33. SDO and the endorsement of only Sidanius, J., & Pratto, F. (1999). Social those items positioning equality-as- dominance: An intergroup theory of social meritocracy. People high in SDO hierarchy and oppression. Cambridge: were also just a likely to disagree Cambridge University Press. with items positioning equality as The Social Report (2005). Wellington,Wellington, NZ: group based. Ministry of Social Development. Keywords: politics, voting, ideology, social Tuffi n, K., Praat, A., & Frewin, K. (2004). policy, affi rmative action, discourses of Analysing a silent discourse: Sovereignty racism, biculturalism, intergroup relations. and tino rangatiratanga in Aotearoa. New Zealand Journal of Psychology, 33, 100-108. Verkuyten, M. (2005). Ethnic group identifi cation and group evaluation among minority and majority groups: Testing the multiculturalism hypothesis. Journal of Personality and Social Psychology, 88, 121-138. Wetherell, M. & Potter, J. (1992). Mapping the language of racism: Discourse and the legitimation of exploitation. Hemmel Hampstead: Harvester Wheatsheaf.

• 84 • New Zealand Journal of Psychology Vol. 36, No. 2, July 2007