This work was supported by a Korea Research Foundation Grant funded by the Korean government (KRF-2004-B00021). This article builds on a paper delivered at an East Asia Institute conference in Seoul, Korea, August 19–20, 2005. For comments and suggestions we are grateful to Byung-kook Kim, Nae Young Lee, Okyeon Yi, Taehyun Kim, Victor Shih, Meredith Jung-en Woo, and Robert Y. Shapiro.Google Scholar

1. Benjamin, I. Page with Bouton, Marshall M., The Foreign Policy Disconnect: What Americans Want from Our Leaders but Don't Get (Chicago: University of Chicago Press, 2006), ch. 4–6, gives many examples of the effects of individuals' feelings toward particular countries on their policy preferences related to those countries.Google Scholar

2. For example, Popkin, Samuel L., The Reasoning Voter: Communication and Persuasion in Presidential Campaigns (Chicago: University of Chicago Press, 1994); Lupia, Arthur, “Shortcuts Versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections,” American Political Science Review 88, no. 1 (1994): 63–76; Lau, Richard and Redlawsk, David P., “Advantages and Disadvantages of Cognitive Heuristics in Political Decision Making,” American Journal of Political Science 45, no. 4 (2001): 951–971.Google Scholar

3. See Sears, David O., Huddy, Leonie, and Jervis, Robert, eds., The Oxford Handbook of Political Psychology (New York: Oxford University Press, 2003), especially ch. 6 (“The Psychology of Emotion and Politics” by George E. Marcus), ch. 12 (“Political Impressions: Formation and Management” by Kathleen M. McGraw), and ch. 13 (“Information Processing and Public Opinion” by Charles S. Taber).Google Scholar

4. Nelson, Thomas E. and Kinder, Donald R., “Issue Frames and Group-Centrism in American Public Opinion,” Journal of Politics 58, no. 4 (1996): 1055–1078; Popkin, , The Reasoning Voter; Jackman, Mary R. and Crane, Marie, “‘Some of My Best Friends Are Black …’: Interracial Friendship and Whites' Racial Attitudes,” Public Opinion Quarterly 50, no. 4 (1986): 459–486; Brown, Rupert, Eller, Anja, Leeds, Sarah, and Stace, Kim, “Intergroup Contact and Intergroup Attitudes: A Longitudinal Study,” European Journal of Social Psychology 37, no. 4 (2006): 692–703, DOI: 10.1002/ejsp.384.Google Scholar

5. Sniderman, Paul M., Brody, Richard A., and Tetlock, Philip E., Reasoning and Choice: Explorations in Political Psychology (New York: Cambridge University Press, 1991).Google Scholar

6. Brewer, Paul R., “National Interest Frames and Public Opinion About World Affairs,” Harvard International Journal of Press/Politics 11, no. 4 (2006).CrossRefGoogle Scholar

7. Maggiotto, Michael A. and Wittkopf, Eugene R., “American Public Attitudes Toward Foreign Policy,” International Studies Quarterly 25, no. 4 (1981): 601–631; Wittkopf, Eugene R., “On the Foreign Policy Beliefs of the American People: A Critique and Some Evidence,” International Studies Quarterly 30, no. 4 (1986): 425–445.Google Scholar

8. Chittick, William O., Billingsley, Keith R., and Travis, Rick, “A Three-Dimensional Model of American Foreign Policy Beliefs,” International Studies Quarterly 39, no. 3 (1995): 313–331.Google Scholar

9. Richman, Alvin, Malone, Eloise, and Nolle, David, “Testing Foreign Policy Belief Structures of the American Public in the Post-Cold War Period: Gross Validations from Two National Surveys,” Political Research Quarterly 50, no. 4 (1997): 939–955.CrossRefGoogle Scholar

10. Page, with Bouton, , The Foreign Policy Disconnect , ch. 2.Google Scholar

11. Herrmann, Richard K., Tetlock, Paul E., and Visser, Penny S., “Mass Public Decisions to Go to War: A Cognitive-Interactionist Framework,” American Political Science Review 93, no. 3 (1999): 553–573.CrossRefGoogle Scholar

12. Hurwitz, Jon and Peffley, Mark, “How Are Foreign Policy Attitudes Structured?” American Political Science Review 81 (1987): 1099–1120.Google Scholar

13. Peffley, Mark and Hurwitz, Jon, “International Events and Foreign Policy Beliefs: Public Response to Changing Soviet-US Relations,” American Journal of Political Science 36, no. 2 (1992): 431.Google Scholar

14. Page, with Bouton, , The Foreign Policy Disconnect , especially p. 15.Google Scholar

15. Ibid., pp. 125–128; 134–135; 145–146; 151–152; 188–189; 195–197.Google Scholar

16. Feldman, Stanley, “Structure and Consistency in Public Opinion: The Role of Core Beliefs and Values,” American Journal of Political Science 32, no. 2 (1988): 416–440; Wilcox, Clyde, Sigelman, Lee, and Cook, Elizabeth, “Some Like It Hot: Individual Differences in Responses to Group Feeling Thermometers,” Public Opinion Quarterly 53, no. 2 (1989): 246–257; Citrin, Jack, Green, Donald P., Muste, Christopher, and Wong, Cara, “Public Opinion Toward Immigration Reform: The Role of Economic Motivations,” Journal of Politics 59, no. 3 (1997): 858–882.CrossRefGoogle Scholar

17. When feeling thermometer scores are used as dependent variables in regression analyses, the proportion of variance accounted for (as indicated by adjusted R-squared values) is seldom very large. This suggests that the scores include substantial noise or measurement error—that different individuals may use the 100-point scale in different ways. Still, the differentiation in mean scores signals that respondents collectively make significant distinctions between countries, and the significant regression coefficients indicate that there are meaningful patterns in individuals' responses.Google Scholar

18. The standard deviations of thermometer ratings in 2002 were North Korea, 23.4; Japan 21.7; India 21.6; China 22.9, Taiwan 21.8; South Korea 22.2; Pakistan 23.8; Afghanistan 25.5.Google Scholar

19. In 2002, the standard errors of all means were less than 1 degree, ranging from .66 (Japan) to .96 (Afghanistan.) Thus, differences in mean ratings of 2 or 3 degrees are generally statistically significant at a high level of confidence.Google Scholar

20. All the CCGA surveys involve highly representative samples of the US adult population, but the 1978–1998 surveys were conducted (by Gallup) in person; the 2002 survey (by Harris Interactive) was largely by telephone; and the 2004 and 2006 surveys (by Knowledge Networks) were Web-based. Comparison of responses to questions asked at the same time using different survey modes indicates that the CCGA's mean Web-based thermometer ratings were roughly 6 or 7 percentage points lower than the in-person or telephone ratings. For a detailed discussion, see Page, with Bouton, , The Foreign Policy Disconnect , ch. 3.Google Scholar

21. Page, Benjamin I. and Shapiro, Robert Y., The Rational Public: Fifty Years of Trends in Americans' Policy Preferences (Chicago: University of Chicago Press, 1992); Holsti, Ole, Public Opinion and American Foreign Policy, rev. ed. (Ann Arbor: University of Michigan Press, 2004).CrossRefGoogle Scholar

22. In 2002, the proportion of respondents rating Asian countries at the exactly neutral, 50-degree point ranged from a low of 16 percent for Pakistan and Afghanistan to a high of 32 percent for India and South Korea.Google Scholar

23. Average ratings of India between 1978 and 2002 were 49, 48, 48, 48, 46, and 46 degrees. The similar 46-degree rating in 2006, presumably depressed by the Web-based mode of that survey, probably represents a significant warming of feelings from the telephone-based 2002 survey.Google Scholar

24. Average ratings of South Korea were 49 degrees in Web-based 2004 and 44 degrees in Web-based 2006.Google Scholar

25. Average thermometer ratings of Taiwan between 1978 and 2002 were 51, 49, 52, 48, 48, 51, and 50 degrees. They were not elicited in 2004 or 2006.Google Scholar

26. Indonesia's average thermometer rating in the Web-based 2006 survey was 41 degrees. If the mode effect depresses ratings by some 6 or 7 degrees, as it appears to do, feelings toward Indonesia in 2006 were quite similar to those toward India in the 1978–2002 period, though significantly cooler than feelings toward India in 2006.Google Scholar

27. The average rating of Japan in the 2006 Web-based survey was 58 degrees, suggesting that there may have been a small rise since the 60-degree rating in the telephone-based survey of 2002.Google Scholar

28. In the Web-based 2006 survey, Australia's average thermometer rating of 69 degrees was very nearly the same as Great Britain's 71 degrees and well above Japan's 58 degrees.Google Scholar

29. In the Web-based 2004 survey, North Korea's average thermometer rating was 28 degrees; in Web-based 2006 it was just 23 degrees.Google Scholar

30. The 2004 CCGA data are too sparse to be useful here. The 2006 data include more thermometer scores and more relevant independent variables than in 2004, but they omit several crucial independent variables—including all four goals used in the capitalism and markets index that, as we will see, is an important determinant of feelings toward Asian countries. Hence our primary reliance on 2002 data. We also did extensive analyses for 2006, however, with generally similar results; some are noted below.Google Scholar

31. After rescaling, for example, the six-category education variable takes on values of 0, .20, .40, .60, .80, and 1.00.Google Scholar

32. Multicollinearity may depress significance levels for unemployment and other SES variables, so that the slightly negative estimated effects of unemployment on feelings may be real. Multicollinearity is not a problem elsewhere, and in any case it reduces the precision of estimates but does not bias them up or down.Google Scholar

33. See Page, Benjamin I. and Rabinovich, Julia, “Twisting the Lion's Tail: Americans' Ethnic and Religious Ties to Foreign Countries,” unpubl. paper, Northwestern University, June 2005; Page, with Bouton, , The Foreign Policy Disconnect, ch. 3.Google Scholar

34. When the information index was added to demographic regressions, estimated coefficients for education dropped as follows: for Afghanistan, from 19.21^∗∗ to 11.10^∗ ; Taiwan, , 13.49^∗∗ to 7.59+; Korea, South, 14.24^∗∗ to 9.36^∗; Pakistan, 22.26^∗∗ to 18.43^∗∗; India, 17.39^∗∗ to 14.53^∗∗; and Japan, 14.37^∗∗ to 11.30^∗∗ .Google Scholar

35. See Page, with Bouton, , The Foreign Policy Disconnect , ch. 2, and appendix.Google Scholar

36. The four goals included in our capitalism and world markets index may seem heterogeneous, but in fact they cluster together rather closely in Americans' minds. Correlations between all six pairs of these goals fall in the r = .30 to .41 range. (The maximum and minimum correlations are r = .41 for financial stability with international law and r = .30 for financial stability with promotion of markets abroad. See Page, with Bouton, , The Foreign Policy Disconnect , p. 248.)Google Scholar

37. In 2006, being a strong Democrat (compared to a strong Republican) led to more negative feelings toward South Korea, North Korea, Japan, and Indonesia, by about 5, 6, 6, and 7 degrees, respectively. We also found a small negative impact of liberalism on feelings toward Indonesia and a small positive effect for North Korea in 2006.Google Scholar

38. In 2006, “active part” internationalism had similarly positive estimated effects, in the 5- to 10-degree range, on feelings toward China, South Korea, Indonesia, India, and (especially) Japan.Google Scholar

39. For 2006, our information index could include answers to an Asia-specific question, “Do you think the United States loans more money to China or that China loans more money to the United States?” as well as the earlier euro and UN secretary-general items. Estimated effects of having the highest rather than lowest levels of information, in regressions roughly analogous to those of Table 3, were feelings approximately 21 degrees warmer toward South Korea, 19 warmer toward Japan, 12 warmer toward India, 8 warmer toward Indonesia, and 13 colder toward North Korea, but no significant effect for China.Google Scholar

40. Our previous finding (reported at the Seoul conference) that support for using US troops abroad has affected feelings toward South Korea was probably mistaken. It was based on the relationship between thermometer scores and various policy preference questions concerning troop use, which are more subject to reciprocal causation (influence by, rather than influence on, feelings) than the basic goals and attitudes used here.Google Scholar

41. See Page, with Bouton, , The Foreign Policy Disconnect , p. 189.Google Scholar

42. Using the 2006 data, regression analyses analogous to those for Table 3 (but necessarily excluding the capitalism, world markets, and the defending others indexes) reveal effects similar to those in 2002. Domestic concerns led to significant and substantively cooler feelings toward all countries but Japan, particularly India, where going from one end of the scale to the other produced movement of a full 25 thermometer degrees. Antipoverty goals had positive and significant effects on feelings toward countries with a large number of citizens living in poverty (North Korea, China, India), but no significant effects for more affluent Japan or South Korea. Concern over defending the United States led to more positive assessments of South Korea and more negative assessments of North Korea, but no other significant effects. The main 2006 tables and the syntax used to produce them are available from one of the authors (d-tully@northwestern.edu) on request. Note that omission of the important capitalism and world markets variable may lead to upward biases in other coefficients.Google Scholar

43. The coefficient for the effect of antipoverty goals on feelings toward Mexico is only significant at p < 10.Google Scholar