Political Conservatism, Rigidity, and Dogmatism in American Foreign Policy Officials: The 1966 Mennis Data

The Journal of Psychology, 2007, I4I(\), 77-90 Copyright 2007 Heldref Publications Political Conservatism, Rigidity, and Dogmatism in American Foreign Policy Officials: The 1966 Mennis Data MARKUS KEMMELMEIER University of Nevada ABSTRACT. Researchers have established a relationship between political orientation and cognitive styles (A. Chirumbolo, 2002; J. T. Jost, J. Glaser, A. W. Kruglanski, & F Sulloway, 2003a, 2003b; M. Kemmelmeier, 1997). In this article, the author examined whether this finding is true in the political elite, whether the relationship is linear or curvilinear, and whether interest in politics moderates the relationship between political orientation and cognitive styles. He used a 1966 sample of American foreign policy officials {N = 95) to examine the relationship between self-described conservatism and party identification and individual differences in rigidity and dogmatism. Rigidity was related to selfdescribed conservatism, but this relationship was only significant among participants high in political interest, whereas dogmatism was unconditionally related to party identification. All relationships were linear and did not contain a curvilinear component. Key words: cognitive styles, conservatism, dogmatism, political involvement, rigidity WHETHER POLITICAL ORIENTATION IS ASSOCIATED with individual differences in cognitive dispositions has been an ongoing debate since the publication of The Authoritarian Personality (Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950). Most recently, Jost, Glaser, Kruglanski, and Sulloway (2003a) offered a comprehensive meta-analysis in which they demonstrated a reliable and linear relationship between intolerance of ambiguity, dogmatism, openness to experience, need for structure, and cognitive complexity, on the one hand, and political conservatism on the other hand. Despite the strong findings by Jost et al., in my study I sought to address perennial problems that have plagued research in this area. First, the majority of researchers have focused on The author thanks hem Uz and Heather Gillespie for their comments on an earlier draft of this article. Address correspondence to Markus Kemmelmeier, Interdisciplinary PhD Program in Social Psychology, Department of Sociology/300, University of Nevada, Reno, NV 89557; markusk@unr.edu (e-mail). 11

78 The Joumai of Psychology Students and, to a lesser extent, on adult members of tbe general public; hence, it is not clear to what extent the fmdings can be applied to memebers of the political elite. For people in politics and govemment, cognitive styles have more palpable consequences because the way they think about political matters is reflected in the way they decide and implement policy. Second, critics have often claimed that the oft-obtained linear relationship between conservatism and cognitive styles reflects the anticonservative biases of left-leaning academics or is simply tbe result of sampling biases. Rather, it is often proposed tbat right-wing and left-wing extremists are similar and, therefore, can be expected to have similar cognitive styles. Third, previous researchers have treated the relationship between political conservatism and cognitive orientations as overly static, neglecting to examine factors that enhance or reduce the strength of tbis link. In my study, I proposed tbat individual differences in political interest moderate tbe correlation between political conservatism and cognitive orientations. Conservatism, Cognitive Styles, and Sampling It is problematic that evidence linking individual differences in cognitive styles and conservatism is often based on student samples, witb only a minority of data coming from adult samples (for exceptions, see Kossowska & Van Hiel, 2003; Van Hiel, Pandelaere, & Duriez, 2004). Furthermore, little is known about cognitive orientation and conservatism among the political elite. One could argue that the political mindset of policymakers and government officials is much less shaped by cognitive styles tban it is for the general public. Political attitudes and decisions are presumably driven by specific policy goals or by the facts rather than by the idiosyncrasies of individual thought patterns. However, I argue that cognitive styles have a stronger effect on the political orientation of this group because cognitive styles are constantly exhibited in the ways in which they organize and think about their political knowledge. Thus, I predict that, among the political elite, the correlation between conservatism and cognitive styles should be stronger than it is in the general public. Jost et al.'s (2003a) meta-analysis did not include a sample of policymakers or government officials providing self-report data on their cognitive styles and levels of conservatism, presumably because sampling politicians and government officials often poses insurmountable obstacles (but see Altemeyer, 1988, for a study including Canadian regional politicians). The only systematic study of political orientation and cognitive style in the political elites occurred in the context of research on integrative complexity (Jost et al., 2003a). The authors read publicly available speeches of politicians or court opinions and coded them for complexity and relation to known political views of the speaker or author (Tetlock, 1983, 1984; Tetlock, Bernzweig, & Gallant, 1985; Tetlock, Hannum, & Micheletti, 1984). However, Gruenfeld (1995) found that integrative complexity varied dramatically with the social context: In her study of U.S. Supreme Court

Kemmelmeier 79 opinions, she found that majority opinions were always more complexly integrative than were minority opinions, regardless of whether there was a conservative or a liheral majority. Thus, it is unclear to what extent integrative complexity reflects a latent personality disposition. In sum, researchers have not established whether previous findings about the relationship between conservatism and cognitive styles apply equally to policymakers and government officials or whether the relationship is stronger in those populations than it is in the general public and student samples. In my study, I focused on a unique sample of American foreign policy officials, a select group of govemment officials. Linearity vs. Curvilinearity The limited empirical basis of Jost et al.'s (2003a) conclusions is particularly problematic in evaluation of the existence of a curvilinear, not merely a linear, relationship between conservatism and cognitive styles. Various theorists have challenged the validity of the typically reported Unear relationship, suggesting that it reflects political preference or even ideological biases on the part of the researchers (Durrheim, 1997; Martin, 2001; Suedfeld, 2002). Greenberg and Jonas (2003) said that left-wing and right-wing extremists are very alike in that they are more cognitively rigid and less tolerant of ambiguity than are centrists (Shils, 1954; Taylor, 1960). This hypothesis predicts a U-shaped function relating cognitive styles to cognitive orientation. On the basis of this idea, McCloskey and Chong (1985) argued that heavy reliance on student samples introduces systematic bias. Colleges and universities tend to lean somewhat left-wing in their social environments (Kemmelmeier, Danielson, & Basten, 2005; Klein & Stem, 2005; Klein & Westem, 2005; Zipp & Fenwick, 2006). Thus, researchers reporting a linear relationship between conservatism and cognitive styles may have obtained this result because extreme left-wingers were overrepresented and extreme right-wingers underrepresented in their samples. In other words, these studies may find only one side of the U-shaped relationship, thus producing the illusion of linearity. In response to this criticism, Jost et al. (2003b) demonstrated that, although such a curvilinear trend is found in some studies, it always coexists with a linear trend. Extreme left-wingers may be less tolerant of ambiguity and less cognitively complex than are centrists, but they are still more tolerant of ambiguity and more complex than are extreme right-wingers. Because it is unclear whether this pattern occurs among nonstudent samples, I reexamined the issue in a sample of foreign policy officials. Political Interest as a Moderator ofthe Conservatism-Cognitive Style Relationship Unlike researchers in other domains, researchers in cognitive dispositions and political views have treated the relationship as static, which may have led to recent

80 The Journal of Psychology efforts to demonstrate a genetic basis of conservatism and personality variables (Bouchard et al., 2003; Tesser, 1993). As a result, researchers have not explored potential contextual moderators of the strength and nature of this relationship. It is not clear under what circumstances a correlation between cognitive styles and political conservatism is stronger or weaker. In a previous study, I (Kemmelmeier, 2006) proposed political interest as a plausible moderator of the strength of this relationship. Public opinion researchers have established that political involvement and interest moderates the relationship between values and issue attitudes (Judd & Krosnick, 1989; Kemmelmeier, Bumstein, & Peng, 1999; Krosnick, 1988, 1990). People with more interest in politics tend to think more often about politics or elaborate on political issues in discussions. This cognitive elaboration sensitizes individuals to the implications of their values for issue attitudes and promotes greater consistency between values and beliefs. In comparison with individuals with low political interest, this leads to a higher correlation between attitudes and values. In the present study, I proposed a similar dynamic for the relationship between cognitive styles and conservatism (cf. Kemmelmeier, 2006). To the extent that people think and talk about politics often, it is likely that their political views reflect their cognitive style. In contrast, for individuals who are less engaged in politics, such a link is less likely to exist because their political views are not based on extensive cognitive elaboration. To test this prediction, I assessed interest in politics by examining how often individuals discuss political issues with others (see Straits, 1991). For individuals who regularly engage in political discussion, I predicted that rigidity and dogmatism would be strongly linked to political conservatism, but I did not expect to find this relationship in individuals who do not discuss politics often. Because in this research I focused on government officials, it might seem peculiar to examine interest in politics because the jobs of the participants inevitably place them within the larger domain of politics. In other words, interest in politics should be consistently high. However, individuals may differ in the extent to which they personally care about the political issues in which their work involves them. Furthermore, different government officials may feel more or less personally invested in their jobs and thus may vary in the extent to which they discuss political matters outside of the office. Therefore, it is reasonable to expect meaningful variation in political interest even among this group. In the present study, I examined the relationship between political conservatism, dogmatism, and rigidity in a unique sample of 95 American foreign policy officials, a select group of government officials who develop and implement foreign policy in the United States and abroad. The data were originally gathered in 1966 by Bernard Mennis as part of his doctoral dissertation, which formed the basis of his book about the American foreign service (Mennis, 1971). I used the data that were made publicly available by the Interuniversity Consortium for Political and Social Research (Mennis, 1966). I did not duplicate any of the analyses reported by Mennis (1971).

Kemmelmeier 81 Method Sample Respondents included 24 civilian and 34 military employees of the Department of Defense and 41 foreign service officers of the Department of State (A^ = 95). All were male with an average age of 42.60 years {SD = 5.45 years; range 27-53 years), and all respondents were White with the exception of 2 African American men. Forty-six percent had been in the service for 2 years or less, with 27% having served for 5 years or more. Fifty-five individuals held a master's degree, law degree, or PhD, and the proportion of advanced degrees did not vary between subgroups (for further details, see Mennis, 1971). Measures Respondents completed an extensive interview about their views on foreign policy and their personal biographies. Political orientation. Respondents indicated on a scale (from 1-7) whether they considered themselves strongly liberal (15.8%), not very strongly liberal (32.6%), independent but closer to liberals (13.7%), independent (8.4%), independent but closer to conservatives (12.6%), not very strongly conservative (13.7%), or very strongly conservative (2.1%). On average, respondents described themselves as somewhat liberal (M =3.16, SD = 1.77). Respondents also provided their party identification by checking on a 1-7 scale whether they considered themselves a very strong Democrat (17.9%), a not very strong Democrat (15.8%), an independent but closer to Democrats (24.2%), an independent (13.7%), an independent but closer to Republicans (11.6%), a not very strong Republican (13.7%), or a very strong Republican (3.2%). This variable served as a secondary measure of conservatism because, in the American political landscape, the Republican Party tends to adopt more conservative positions than does the Democratic Party. The average on this variable showed that overall the sample was slightly more inclined toward the Democratic Party (M=3.39, 5D= 1.76). Dogmatism. Respondents completed 20 dogmatism items from Troldahl and Powell (1965). Sample items included "Most people just don't know what's good for them" and "There are two kinds of people in this world: Those who are for the tmth and those who are against the tmth." Participants marked their responses on a 7-point scale from 1 {disagree very much) to 7 {agree very much). Because the intemal consistency of these items was low (a =.56), I performed factor analysis, and only 11 items that had their highest loading on the dominant factor were retained (a -.60). Because Cronbach's alpha in part reflects scale length (see

82 The Journal of Psychology Spearman-Brown formula, McDonald, 1999), this procedure led to only a small increase in internal consistency. However, trimming unsuitable items effectively improved scale reliability: The average interitem correlation more than doubled, increasing from.06 to.13. Rigidity. Respondents completed 22 items from the Gough-Sanford Rigidity Scale (Rokeach, 1960). Sample items included "I find it easy to stick to a certain schedule once I have started" and "I dislike to change my plans in the midst of an undertaking." Participants used the same 7-point scale for their responses as described under dogmatism. The internal consistency of this scale was satisfactory (a =.76). Political interest. Respondents were asked if they discussed politics with friends and relatives often (74.7%), sometimes (8.4%), or not at all (16.8%). Preliminary Analyses Results Table 1 shows the correlations between the two variables measuring conservatism as well as the two measures of cognitive styles. As I expected, selfdescribed conservatism and identification with the Republican Party correlated significantly, as did the two cognitive style variables, dogmatism and rigidity. Self-described conservatism also correlated with botb rigidity and dogmatism. Party identification was also related significantly to dogmatism, although the correlation with rigidity only approached statistical significance. These findings provide initial support for my hypothesis that a relationship exists between conservatism and cognitive style. Furthermore, the results demonstrate that existing TABLE 1. Zero-Order Correlations Between Measures of Conservatism and Cognitive Style Variable Conservatism 1 2 Cognitive style 3 4 1. Self-described ideology 2. Identification with Republican Party 3. Dogmatism 4. Rigidity.49*.27*.27*.35".17+.47** Note. Because of missing data, the n for individual coefficients varies )etween 1 92 and 95. *p<.\0. ><.01.*'p<.001.

Kemmelmeier 83 findings about conservatism and cognitive style, which were obtained mostly from student samples, do generalize to government officials. This finding was also reflected in the size of the coefficients, which with the exception of the correlation between rigidity and party identification were not much different from the effect size of r =.34 reported by Jost et al. (2003a; see their Table 2). However, this finding does not confirm my hypothesis that in the political elite, the relationship between political conservatism and cognitive styles would be stronger than it was in previous samples. Regressing Conservatism on Cognitive Styles I used a regression to gauge the unique effects of rigidity and dogmatism on measures of political conservatism. When I regressed ideological selfdescriptions on both cognitive styles (R-^ =.15, /? =.001), only rigidity emerged as a reliable predictor, (3 =.19, p <.03, but the effect of dogmatism was reduced to a statistical trend, p =.19, p <.10. In contrast, when I regressed party identification onto cognitive styles (R^ -.10, p <.02), the effect for rigidity was not significant, P =.04, p =.75, whereas dogmatism was now the only reliable predictor, p -.29, p <.02. Tbese fmdings suggest that tbe two measures of conservatism, although highly correlated, were each linked to a unique cognitive style. At the same time, as compared with rigidity, dogmatism appeared to be tbe sligbtly stronger cognitive predictor, because it not only was reliably related to party identification, but also approacbed statistical significance in the prediction of self-described conservatism. Linearity vs. Curvilinearity To clarify whether a linear or a curvilinear relationship exists between conservatism and cognitive styles, I computed quadratic terms for botb rigidity and dogmatism and added tbem to the regression model. Addition of tbese terms did not improve model fit for self-described conservatism, A/?^ =.03, p >.26, nor for party identification, AR^ -.Ol,p>.69. As a result, I concluded that, in this study, there was no curvilinear relationship between cognitive orientation and political conservatism. Interest in Politics as a Moderator I used a related regression model to examine whether interest in politics moderated the strength of the linear relationship between conservatism and cognitive styles. Because the majority of respondents reported discussing politics often, I collapsed the two remaining response categories, creating a variable that distinguished between high-involvement (n = 71) and low-involvement (n - 24) individuals. I entered this dichotomous variable in the first step of a hierarchical

84 The Journal of Psychology regression analysis in which I regressed self-described conservatism or party identification onto dogmatism, rigidity, and political involvement (see Table 2). For the second step of this equation, I calculated the multiplicative interaction terms between dogmatism and rigidity, on the one hand, and interest in politics (coded as 0 = low, 1 - high), on the other hand. The left column of Table 2 presents results for self-described conservatism. Because I included interaction terms, I reported only nonstandardized regression coefficients (Aiken & West, 1991). The addition of interaction terms significantly improved the model fit, AF{2, 86) = 3.54, p <.04, with the interaction between rigidity and interest in politics reaching statistical significance, B - 1.70, p <.02. Subsequent simple effects analyses demonstrated that, when interest in politics was low, there was a nonsignificant negative relationship between rigidity and selfdescribed conservatism, B = -.76, SE =.63, p <.24 (partial r = -.13). However, I found a pronounced positive relationship when interest in politics was high, B =.95, SE =.32, p <.005 (partial r =.30). In other words, the positive association between rigidity and conservatism reported in the literature (e.g., Jost et al., 2003a) emerged for respondents with high levels of involvement in the domain of politics. The corresponding hierarchical regression model for party identification, as shown in the right column of Table 2, did not show any improvement through the addition of interaction terms, AF{2, 87) = 0.75, p <.48. Therefore, I concluded that interest in politics does not moderate the previously established relationship between dogmatism and identification with the Republican Party. TABLE 2. Hierarchical Regression Models Predicting Conservatism on the Basis of Cognitive Style and Interest in Politics Predictor variable Self-described ideology" B SE AR^ Identification with Republican Party*" B SE AR^ Step 1 Dogmatism Rigidity Interest in politics -.45 0.28.66 0.29* -.35 0.40.16**.72 0.28.09 0.30 -.15 0.40.10* Step 2 Dogmatism x Interest in politics Rigidity x Interest in politics.32 0.62 1.70 0.71**.06*.80 0.65 -.19 0.75.02 Note. Entries reflect nonstandardized regression coefficients. All continuous predictor variables were centered. Interest in politics was coded as 0 = low, 1 = high. " R^ for final model =.22. ^R'^ for final model =.11.

Kemmelmeier 85 Discussion In this study, I extended the existing literature on political leanings and cognitive styles by demonstrating a linear relationship in a sample of American foreign policy officials. I concluded that fmdings previously obtained in student or nonelite samples do generalize to members of the political elite. This conclusion is corroborated by my observation that the correlations for rigidity and dogmatism found in this data were not much different from the average correlations reported in the meta-analysis by Jost et al. (2003a) for student and adult samples. At the same time, this pattern does not support my hypothesis that the association between political conservatism and cognitive styles would be greater in this sample of government officials than it is in the general population. I did not find support for the existence of a curvilinear relationship between conservatism and cognitive styles (see also Chirumbolo, 2002; Kemmelmeier, 1997). Despite having been repeatedly proposed over a period of more than 50 years (Greenberg & Jonas, 2003; McCloskey & Chong, 1985; Shils, 1954) and despite the repeated suspicion that a linear relationship reflects ideological bias (Durrheim, 1997; Martin, 2001), there is no compeuing support for the idea that extremists at both ends of the political spectrum are equally more dogmatic or more rigid than are centrists or independents. Critics sometimes argue that the U-shaped relationship between conservative and cognitive styles is not found because of sampling bias (McCloskey & Chong, 1985). As I discussed earlier, even when a curvilinear relationship exists in the population, it will not be apparent if the sample primarily includes individuals on the left and in the political center, as is allegedly the case in most college student samples. This is a potential concern because the sample J used was obtained during the administration of the Democratic President L. B. Johnson and included more self-described liberals than conservatives. At the same time, the sample included a substantial proportion of individuals with conservative leanings, and, therefore, a curvilinear pattern should have emerged, if only through visual inspection of the scatter plot. However, this relationship was absent. Thus, the results of this study corroborate the conclusion of a linear relationship between cognitive orientation and conservatism. My findings also challenge existing research in some areas. First, the finding that the relationship between rigidity and self-described conservatism was moderated by interest in politics suggests that the link between political views and cognitive styles is not as stafic as implied by previous researchers (Adorno et al., 1950; Chirumbolo, 2002; Kemmelmeier, 1997). Instead, individuals have to be personally invested enough in the domain of polifics for their cognitive styles to shape their political views. I found a substantive positive relationship between self-described conservatism and rigidity only in individuals who elaborated on political views through discussion with others. First, this finding not only illustrates the contextual conditions under which the linear relationship

86 The Journal of Psychology between political views and cognitive style exists, but it also suggests a reexamination of previous null findings (Pettigrew, 1958). To the extent that research participants did not care enough about politics, the nonexistence of a positive linear relationship is not a surprising fmding. For the same reason, researchers on conservatism and cognitive styles should include measures of interest in politics in all future endeavors. Second, because a contextual variable can be shown to moderate the relationship between conservatism and cognitive styles, previous challenges to the validity of the linear relationship between these two variables should be reexamined (Durrheim, 1997; Martin, 2001; Suedfeld, 2002). Whether it concems the selection of items or samples, it is less plausible to argue that ideological biases influence findings if it can be shown that a logically unrelated individual-difference characteristic in an otherwise quite homogeneous sample eliminates (and even slightly reverses) the relationship between the same items. Therefore, the moderator effect demonstrated here may in fact help defend against the critique of ideological bias. Note that I established the moderating role of interest in politics for the relationship between rigidity and self-described conservatism, but not for the relationship between dogmatism and party identification. Although the reason for this is not clear, it is important to consider that party identification and selfdescribed political ideology differ. Party identification and ideology are substantially correlated, but I propose that party identificafion is more variable and noncommittal than applying to the self the more broad-ranged ideological labels conservative or liberal. Both conservatives and liberals may prefer the Republican Party (or Democratic Party at times, and vice versa, especially because both parties tend to include a broad ideological spectrum. For example. President Nixon, a Republican, is considered and identified himself to be liberal (Waddan, 1998). Furthermore, party preferences may reflect momentary liking of a party's candidates. From this perspective, the relationship between dogmatism and party preference was not moderated by interest in politics, presumably because the greater cognitive elaborations associated with it are less critical in shaping party preference. Last, I found that self-described conservatism and party identification were linked to different cognitive styles. This finding was unexpected because there is often an assumption that various measures tap the same underlying dimension of conservatism. For example, Jost et al. (2003a) included samples that measured social dominance orientation or right-wing authoritarianism two constructs that, according to other researchers (Altemeyer, 1998; Heaven & Bucci, 2001; Pratto, Sidanius, Stallworth, & Malle, 1994; Whitley, 1999), are distinct. Furthermore, my results did not replicate other results from researchers that assessed both party support and self-described conservatism and found both to be related to the same dimension of cognitive style (Kemmelmeier, 1997; Kossowska & Van Hiel, 2003). I cannot offer a cogent explanation for the differential association

Kemmelmeier 87 between cognitive styles and measures of conservatism found in these studies, but both myself and Kossowska and Van Hiel relied on samples from various European societies, and not from the United States, as I did in the present study. Thus, it is possible that the discrepancy between past and present findings can be explained with regard to the often marked differences in political cultures. In any case, researchers should be aware that different measures of conservatism might show variation in the way conservatism relates to different cognitive styles. To date, only one study has taken the challenge of a more nuanced investigation into the association between these variables. In the context of their cross-cultural study, Kossowska and Van Hiel found that the relationship between political views and cognitive styles was, at least in part, contingent on the historical and political experiences of the people within each culture as well as on the specific measure of conservatism. Whereas the authors found that cultural conservatism and right-wing authoritarianism were positively related to the need for closure (NFC) in both Poland and Belgium, their measure of economic conservatism was positively related to NFC in Belgium, but negatively related in Poland. Although generalizations offered by Jost et al.'s (2003a) meta-analysis are important for a cumulative science (Hunter & Schmidt, 1996; Rosenthal, 1983). Kossowska and Van Hiel's findings illustrate that differnt samples and methodologies may lead researchers to different results regarding the relationships between political conservatism and cognitive styles. This research has limitations. Although they were government officials, the American foreign policy officers involved in this study were neither elected nor involved in defining the political goals of any administration. Thus, it is unclear to what extent my (and all previous) findings can be applied to the most powerful American leaders. Another limitation is that many participants in the sample entered the foreign service during the administrations of two Democratic presidents, John F. Kennedy and Lyndon B. Johnson. It is unclear to what extent the findings apply to individuals who joined the foreign service during Republican administrations. Furthermore, the reliance on Mennis's (1966) data necessarily imposed limitations on the choice of measures and the specific composition of the sample (e.g., the exclusion of women). Still, the sample is unique in the literature on conservatism and cognitive styles. Because I replicated and extended previous findings, the contributions made by this study outweigh its limitations, and the finidings offer important insights into the link between political views and individual cognitive styles. REFERENCES Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. New York; Harper. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA; Sage. Altemeyer, B. (1998). The other "authoritarian personality." In M. Zanna (Ed.), Advances

88 The Journal of Psychology in experimental social psychology (Vol. 30, pp. 47-92). San Diego, CA: Academic. Altemeyer, R. A. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco: Jossey-Bass. Bouchard, T. J., Segal, N. L., Tellegen, A., McGue, M., Keyes, M., & Krueger, R. (2003). Evidence of the construct validity and heritability of the Wilson-Patterson conservatism scale: A reared-apart twins study of social attitudes. Personality and Individual Differences, 34, 959-969. Chirumbolo, A. (2002). The relationship between need for cognitive closure and political orientation: The mediating role of authoritarianism. Personality and Individual Differences, 32, 603-610. Durrheim, K. (1997). Theoretical conundrum: The politics and science of theorizing authoritarian cognition. Political Psychology, 18, 625-656, Greenberg, J., & Jonas, E. (2003). Psychological and political orientation The left, the right, and the rigid: Comment on Jost et al. (2003). Psychological Bulletin, 129, 376-382. Gruenfeld, D. H. (1995). Status, ideology, and integrative complexity on the U.S. Supreme Court: Rethinking the politics of political decision making. Journal of Personality and Social Psychology, 68, 5-20. Heaven, P. C. L., & Bucci, S. (2001). Right-wing authoritarianism, social dominance orientation and personality: An analysis using the IPIP measure. European Journal of Personality, 15, 49-56. Hunter, J. E., & Schmidt, R L. (1996). Cumulative research knowledge and social policy formulation: The critical role of meta-analysis. Psychology, Public Policy, and Law, 2, 324-347. Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. (2003a). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339-375. Jost, 1. T., Glaser, J., Kruglanski, A. W., & Sulloway, F (2003b). Exceptions that prove the rule Using a theory of motivated social cognition to account for ideological incongruities and political anomalies: Reply to Greenberg and Jonas (2003). Psychological Bulletin, 129, 383-393. Judd, C. M., & Krosnick, J. A. (1989). The structural basis of consistency among political attitudes: The effects of political expertise and attitude importance. In A. R. Pratkanis, S. J. Breckler, & A. G. Greenwald (Eds.), Attitude structure and function (pp. 99-128). Hillsdale, NJ: Erlbaum. Kemmelmeier, M. (1997). Political orientation and need for cognitive closure in German university students. The Journal of Social Psychology, 137, 787-789. Kemmelmeier, M. (2006). Gender differences in conservatism: The role of cognitive orientations. Unpublished manuscript. University of Nevada, Reno. Kemmelmeier, M., Bumstein, E., & Peng, K. (1999). Individualism and authoritarianism shape attitudes toward physician-assisted suicide. Journal of Applied Social Psychology, 29, 2613-2631. Kemmelmeier, M., Danielson, C, & Basten, J. (2005). What's in a grade? Academic success and political orientation. Personality and Social Psychology Bulletin, 31, 1386-1399. Klein, D. B., & Stern, C. (2005). Political diversity in six disciplines. Academic Questions, 18, 40-52. Klein, D. B., & Westem, A. (2005). Voter registration of Berkeley and Stanford faculty. Academic Questions, 18, 53-65. Kossowska, M., & Van Hiel, A. (2003). The relationship between need for closure and conservative beliefs in westem and eastern Europe. Political Psychology, 24, 501-518. Krosnick, J. A. (1988). The role of attitude importance in social evaluation: A study of

Kemmelmeier 89 policy preferences, presidential candidate evaluations, and voting behavior. Journal of Personality and Social Psychology, 55, 196-210. Krosnick, J. A. (1990). Government policy and citizen passion: A study of issue publics in contemporary America. Political Behavior, 12, 59-92. Martin, J. L. (2001). The authoritarian personality, 50 years later: What lessons are there for political psychology? Political Psychology, 22, 1-26. McCloskey, H., & Chong, D. (1985). Similarities and differences between left-wing and right-wing radicals. British Journal of Political Science, 15, 329-363. McDonald, R. R (1999). Test theory: A unified treatment. Mahwah, NJ: Erlbaum. Mennis, B. (1966). American foreign policy officials, 1966 (Study No. 5809) [Computer file]. Ann Arbor, MI: Interuniversity Consortium for Political and Social Research [distributor]. Mennis, B. (1971). American foreign policy officials: Who they are and what they believe regarding international politics. Columbus: Ohio State University Press. Pettigrew, T. F. (1958). The measurement and correlates of category width as a cognitive variable. Journal of Personality, 26, 532-544. Pratto, R, Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741-763. Rokeach, M. (1960). The open and the closed mind. New York: Basic Books. Rosenthal, R. (1983). Meta-analysis: Toward a more cumulative social science. Applied Social Psychology Annual, 4, 65-93. Shils, E. A. (1954). Authoritarianism: "Right" and "left." In R. Christie & M. Jahoda (Eds.), Studies in the scope and method o/the authoritarian personality (pp. 24-49). Glencoe, IL: Free Press. Straits, B. C. (1991). Bringing strong ties back in: Interpersonal gateways to political information and influence. Public Opinion Quarterly, 55, 432^48. Suedfeld, P. (2002). Postmodernism, identity politics, and other political influences in political psychology. In K. R. Monroe (Ed.), Political psychology (pp. 315-327). Mahwah, NJ: Erlbaum. Taylor, I. A. (1960). Similarities in the structure of extreme social attitudes. Psychological Monographs, 74, 1-36. Tesser, A. (1993). The importance of heritability in psychological research: The case of attitudes. Psychological Review, 100, 129-142. Tetlock, P. E. (1983). Cognitive style and political ideology. Journal of Personality and Social Psychology, 45, 118-126. Tetlock, P. E. (1984). Cognitive style and political belief systems in the British House of Commons. Journal of Personality and Social Psychology, 46, 365-375. Tetlock, P. E., Bemzweig, J. & Gallant, J. L. (1985). Supreme Court decision making: Cognitive style as a predictor of ideological consistency of voting. Journal of Personality and Social Psychology, 48, 1227-1239. Tetlock, P E., Hannum, K. A., & Micheletti, P M. (1984). Stability and change in the complexity of senatorial debate: Testing the cognitive versus rhetorical style hypotheses. Journal of Personality and Social Psychology, 46, 979-990. Troldahl, V. C, & Powell, F A. (1965). A short-form dogmatism scale for use in field studies. Social Forces, 44, 211-214. Van Hiel, A., Pandelaere, M., & Duriez, B. (2004). The impact of need for closure on conservative beliefs and racism: Differential mediation by authoritarian submission and authoritarian dominance. Personality and Social Psychology Bulletin, 30, 824-837. Waddan, A. (1998). A liberal in wolf's clothing: Nixon's Family Assistance Plan in the light of 1990s welfare reform. Journal of American Studies, 32, 203-218.

90 The Journal of Psychology Whitley, B. E., Jr. (1999). Right-wing authoritarianism, social dominance orientation, and prejudice. Journal of Personality and Social Psychology, 77, 126-134. Zipp, J. F, & Fenwick, R. (2006). Is the academy a liberal hegemony? The political value orientations and educational values of professors. Public Opinion Quarterly 70 304-326. Original manuscript received March 10, 2006 Final revision accepted July 20, 2006