I am indebted to the International Research and Exchanges Board for providing the opportunity to undertake research in Moscow and to the All-Union Bank for Sociological Data, the Institute of Sociology of the Academy of Sciences, Moscow, for access to their data archives and for assistance well beyond the call of duty. I especially want to thank Nina Rostegaeva, Slava Shipilov, and Aleksandr Zhavoronkov. I am grateful to Valerii Andreevich Mansurov, deputy director of the Institute of Sociology, for permission to use data from his project on the role of the intelligentsiia in the structure of Soviet society, and to Boris Grushin and Vladimir Iadov for access to the city of Taganrog data. I would also like to thank Donna Bahry for allowing me to use her 1992 survey and for helpful comments on the first draft of this paper. Additional data for this study were produced by the Soviet Interview Project. This project was supported by Contract No. 701 from the National Council for Soviet and East European Research to the University of Illinois at Urbana-Champaign, James R. Millar, principal investigator. The analysis and interpretations in this study are those of the author, not necessarily of the sponsors. An earlier version of this work has been published as number 237 in the University of Strathclyde's series, Studies in Public Policy. That version contains a technical appendix describing the coding of the variables used. It can be obtained from the Centre for the Study of Public Policy, University of Strathclyde, Glasgow G1 1XH, Scotland.

1. “Russian” in this article refers to people who live in Russia, even if they are not of Russian ethnic background.

2. The survey involved 2106 respondents representative of urban Russia. See Boeva, Irina and Shironin, Viacheslav, Russians between State and Market: The Generations Compared, Studies in Public Policy 205 (Glasgow: Centre for the Study of Public Policy, University of Strathclyde, 1992), 30–31 Google Scholar.

3. The poll was conducted between 8 and 18 August 1992. There were 1695 respondents. Most VTsIOM surveys are representative of the Russian urban population, though the report of this one gave little information on the sample. See “Ekspress-Opros ‘MN’,” Moskovskie novosti, no. 36 (September 1992): 2.

4. The survey had 1703 respondents, from 13 regions in Russia, and was conducted in July 1992 “Ekspress-Opros ‘MN',” Moskovskie novosti, no. 33 (August 1992): 2.

5. Ladodo, I.V., “Nekotorye kharakteristiki informatsionnoi kontrpropagandy (po dannym issledovanii za 1983 god)” (Moscow: Institut Sotsiologicheskikh Issledovanii MIMEO, n.d.)Google Scholar.

6. Raymond Duch and James Gibson do comment on an additional problem: about 19% of their sample offered an opinion on a fictitious organization, even when explicitly offered the option of not having an opinion. This appears to be a larger percentage than is usual in western surveys (see their “Emerging Democratic Values in Soviet Political Culture,” in Arthur H. Miller, William M. Reisinger and Vicki L. Hesli, eds., Public Opinion and Regime Change: The New Politics of Post-Soviet Societies [Boulder: Westview Press, 1993], 69-94); see also Bishop, George, Oldendick, R.W., Tuchfarber, A.J. and Bennett, S.E., “Pseudo-Opinions on Public Affairs,” Public Opinion Quarterly 44 (1980): 200–202 CrossRefGoogle Scholar.

7. Brym, Robert J. and Degtyarev, Andrei, “Anti-Semitism in Moscow: Results of an October 1992 Survey,” Slavic Review 52, no. 1 (Spring 1993): 1–12 CrossRefGoogle Scholar. Also, the discussion in Gibson, James L., “Understandings of Anti-Semitism in Russia: An Analysis of the Politics of Anti-Jewish Attitudes,” Slavic Review 53, no. 3 (Fall 1994): 801-2Google Scholar; Brym, Robert J., “Anti-Semitism in Moscow: A Re-examination,” Slavic Review 53, no. 3 (Fall 1994): 845-47.CrossRefGoogle Scholar

8. Grushin, Boris A., “Est' li u nas obshchestvennoe mnenie?” Novoe vremia, no. 30 (1988): 29–31 Google Scholar; Grushin, Boris A., “Obshchestvennoe mnenie v sisteme upravleniia,” Sotsiologicheskie issledovaniia, no. 3 (1988): 24–29 Google Scholar. Also in Soviet Sociology 28 (July-August 1989): 72-80.

9. Levykin, I.T., “Razvitie sotsial'noi struktury i sovershenstvovanie sotsialisticheskogo obraza zhizni,” in Ivanov, V.N., Kudriavtsev, V.N. and Smirnov, G.L., eds., Problemy issledovanii sotsial'noi struktury sovetskogo obshchestva (Moscow: Akademiia Nauk SSSR, 1984), 23 Google Scholar; Toshchenko, Zh.T. and Popov, V.D., eds., Obshchestvennoe mnenie: Metodicheskie rekomendatsii dlia provedeniia sotsiologicheskikh oprosov (Moscow: Akademiia Obshchestvennykh Nauk pri TsK KPSS, 1980), 34–35 Google Scholar; Ulybin, K.A., Sovremennoe ekonomicheskoe myshlenie (Moscow: Politizdat, 1986), 19–21 Google Scholar; Uledov, A.K., Obshchestvennoe mnenie sovetskogo obshchestva (Moscow: Izd-vo Sotsial'no-ekon. lit-ry, 1963), 89 Google Scholar.

10. Grushin, “Obshchestvennoe mnenie v sisteme upravleniia,” 74.

11. Raymond M. Duch and James L. Gibson, “The New Democratic Activists in the Former Soviet Union” (paper presented at the 1992 Annual Meeting of the American Political Science Association, 3-6 September 1992); Gibson, James L., Duch, Raymond M. and Tedin, Kent L., “Democratic Values and the Transformation of the Soviet Union,” Journal of Politics 54 (May 1992): 329-71CrossRefGoogle Scholar; Hahn, Jeffrey W., “Continuity and Change in Russian Political Culture,” British Journal of Political Science 2 (October 1991): 393–421 CrossRefGoogle Scholar; Tedin, Kent L., “Popular Support for Competitive Elections in the Soviet Union,” Comparative Political Studies 27 (July 1994): 241-71.CrossRefGoogle Scholar

The picture, of course, is mixed. In particular, Gibson and Duch find evidence of political intolerance in Russians. Miller, Reisinger and Hesli find minimal support for new political and economic institutions. Gibson, James L. and Duch, Raymond M., “Political Intolerance in the USSR: The Distribution and Etiology of Mass Opinion,” Comparative Political Studies 26 (October 1993): 286–329 CrossRefGoogle Scholar; Miller, Arthur H., Reisinger, William M., and Hesli, Vicki L., “Public Support for New Political Institutions in Russia, the Ukraine, and Lithuania,” Journal of Soviet Nationalities 1, no. 4 (Winter 1990-1991): 82–107.Google Scholar

12. Tsipko, Aleksandr, “What's Next?” The Nation (23 September 1991): 325.Google Scholar

13. Goldfarb, Jeffrey C., Beyond Glasnosl: The Post-Totalitarian Mind (Chicago: University of Chicago Press, 1989)Google Scholar; Grossman, Vasily, Forever Flowing, trans. Whitney, Thomas P. (New York: Harper & Row, 1972)Google Scholar; Havel, Vaclav, “The Power of the Powerless,” in Brinton, William M. and Rinzler, Alan, eds., Without Force or Lies: Voices from the Revolution of Central Europe in 1989-90 (San Francisco: Mercury House, 1990)Google Scholar; Simis, Konstantin M., “The USSR: The Corrupt Society,” in Hoffmann, Erik P. and Laird, Robbin F., eds., The Soviet Polity in the Modern Era (New York: Aldine, 1984), 309 Google Scholar; Turchin, Valentine, The Inertia of Fear and the Scientific Worldview, trans. Daniels, Guy (New York: Columbia University Press, 1981).Google Scholar

14. Havel, “The Power of the Powerless,” 49.

15. Shlapentokh, Vladimir E., “Two Levels of Public Opinion: The Soviet Case,” Public Opinion Quarterly 49 (Winter 1985): 443-59.CrossRefGoogle Scholar

16. Prothro, James W. and Grigg, C.W., “Fundamental Principles of Democracy: Bases of Agreement and Disagreement,” Journal of Politics 22 (May 1960): 276-94.CrossRefGoogle Scholar

17. Bahry, Donna and Silver, Brian, “Public Perceptions and the Dilemmas of Party Reform in the USSR,” Comparative Political Studies 23 (July 1990): 171–209.CrossRefGoogle Scholar

18. Migranian, Andronik, “Vzaimootnosheniia individa, obshchestva i gosudarstva v politicheskoi teorii marksizma i problemy demokratizatsii sotsialisticheskogo obshchestva,” Voprosy filosofii, no. 8 (August 1987): 75–91 Google Scholar; Migranian, Andronik, “Dolgii put’ k evropeiskomu domu,” Novyi mir, no. 7 (July 1989): 166-84.Google Scholar

19. Grushin, “Est’ li u nas … ?“

20. Neuman, W. Russell, The Paradox of Mass Politics: Knowledge and Opinion in the American Electorate (Cambridge: Harvard University Press, 1986), 3.Google Scholar

21. See also Converse, Philip, “The Nature of Belief Systems in Mass Publics,” in Apter, David, ed., Ideology and Discontent (New York: Free Press, 1964), 214-19Google Scholar; Rosenberg, Morris, “Some Determinants of Political Apathy,” Public Opinion Quarterly 18 (1954): 349.CrossRefGoogle Scholar

22. Dalton, Russell J., Citizen Politics in Western Democracies: Public Opinion and Political Parties in the United States, Britain, West Germany, and France (Chatham: Chatham House, 1988), 15–32.Google Scholar

23. Converse, Jean, “Predicting No Opinion in the Polls,” Public Opinion Quarterly 40 (1976): 515-30CrossRefGoogle Scholar; Ferber, Robert, “Item Nonresponse in a Consumer Survey,” Public Opinion Quarterly 30 (1966): 399–415 CrossRefGoogle Scholar; Francis, Joe D. and Busch, Lawrence, “What We Know about ‘I Don't Knows',” Public Opinion Quarterly 39 (1975): 207-18.CrossRefGoogle Scholar

24. Bennett, Stephen E., ‘“Know-Nothings’ Revisited: The Meaning of Political Ignorance Today,” Social Science Quarterly 69 (1988): 476-90.Google Scholar

25. Wolfinger, Raymond E. and Rosenstone, Steven J., Who Voles? (New Haven: Yale University Press, 1980).Google Scholar

26. Piven, Frances Fox and Cloward, Richard A., Why Americans Don't Vote (New York: Pantheon Books, 1988).Google Scholar

27. Dunham, Vera S., In Stalin's Time: Middleclass Values in Soviet Fiction (New York: Cambridge University Press, 1976)Google Scholar; Hauslohner, Peter, “Gorbachev's Social Contract,” Soviet Economy 3, no. 1 (1987)Google Scholar; Hauslohner, Peter, “Politics before Gorbachev: De-Stalinization and the Roots of Reform,” in Bialer, Seweryn, ed., Politics, Society, and Nationality inside Gorbachev's Russia (Boulder: Westview Press, 1989), 41–90 Google Scholar; Lapidus, Gail W., “Society under Strain,” in Hoffmann, Erik P. and Laird, Robbin F., eds., The Soviet Polity in the Modern Era (New York: Aldine, 1984), 691.Google Scholar

28. Bennett, ‘“Know-Nothings’ Revisited.“

29. Bennett, Stephen E. and Resnick, David, “Implications of Nonvoting for Democracy in the U.S.,” American Journal of Political Science 34 (1990): 771–803.CrossRefGoogle Scholar

30. Bogart, Leo, “No Opinion, Don't Know, and Maybe No Answer,” Public. Opinion Quarterly (Fall 1967): 342.Google Scholar

31. McClosky, Herbert and Zaller, John, The American Ethos: Public Altitudes toward Capitalism and Democracy (Cambridge: Harvard University Press, 1984), 79, 141.CrossRefGoogle Scholar

32. Private interview, Tallinn, Estonia, November 1990.

33. One of the growth industries of perestroika were articles which informed Russians about what life in the west was really like. See Voina, Vladimir, “Strana Naoborot,” Gorizont, no. 3 (1991): 39–51 Google Scholar; Zanussi, Kshishtof, “Zhizn’ v Amerike bezlika dlia evropeitsa,” Literaturnaia gazeta, no. 13 (April 1991): 14.Google Scholar

34. Faulkenberry, G. David and Mason, Robert, “Characteristics of Nonopinion and No Opinion Response Groups,” Public Opinion Quarterly 42 (1978): 533-43.CrossRefGoogle Scholar

35. This is not to imply that democracy is secure as long as it is what the people want. Rather my argument is that if popular values oppose a democratic system, then the stability of that system is uncertain. See Almond, Gabriel A. and Verba, Sidney, The Civic Culture: Political Attitudes and Democracy in Five Nations (Newbury Park: Sage Publications, 1989)Google Scholar; Inglehart, Ronald, “The Renaissance of Political Culture,” American Political Science Review 82 (December 1988): 1203-30.CrossRefGoogle Scholar

36. For evidence that apathy in America warps public policy, see Bennett and Resnick, “Implications of Nonvoting“; Piven and Cloward, Why Americans Don't Vote; Templeton, Fredric, “Alienation of Political Participation,” Public Opinion Quarterly 30, no. 2 (1966): 260;CrossRefGoogle Scholar Verba, Sidney, Schlozman, Kay Lehman, Brady, Henry and Nie, Norman H., “Citizen Activity: Who Participates? What Do They Say?“ American Political Science Review 87 (June 1993): 303-18.CrossRefGoogle Scholar The amount of distortion may not be large, however, because many of those whose opinions are inadequately expressed have few opinions (Wolfinger and Rosenstone, Who Votes'? 109).

37. The survey, identified as projects 39 and 40 of the Taganrog study, was based on a random sample of Taganrog's adults, chosen from voter lists. See Andreenkov, V.G. and Zhavoronkov, A.V., Katalog peremennykh bazy sotsiologicheskikh dannykh po problemam izucheniia ideologicheskogo protsessa, 4 vols. (Moscow: Institut Sotsiologicheskikh Issledovanii AN SSSR, 1988), 1: 34 Google Scholar; Grushin, Boris A. and Onikov, Leon A., eds., Massovaia informatsiia v sovetskom promyshlennom gorode: Opyt kompleksnogo sotsiologicheskogc issledovaniia (Moscow: Politizdat, 1980), 432-33.Google Scholar Respondents filled in surveys individually in their homes. In the Taganrog survey in general, particular care was taken to ensure the representativeness of the sample (see Tarshis, E.Ia. and Chernakova, N.E., “Organizatsiia raboty polevoi sotsiologicheskoi laboratorii,” Sotsiologicheskie issledovania, no. 3 [1982]: 157-60).Google Scholar

38. Interviews took place between March 1983 and January 1984; all but ten of the interviews were conducted in Russian. The sample was a stratified random sample based on a list of adult emigrants from the USSR to the United States between the dates mentioned. The response rate was 79%. For this study, I used the responses of the 2667 people who were of voting age five years before their lives changed as a result of the decision to emigrate. See Millar, James R., “History, Method, and the Problem of Bias,” in Millar, James R., ed., Politics, Work, and Daily Life in the USSR (New York: Cambridge University Press, 1987).CrossRefGoogle Scholar

39. See Kol'tsov, V.B. and Mansurov, V.A., “Politicheskie ideologii perioda perestroiki,” Sotsiologicheskie issledovaniia, no. 10 (1991): 22–35 Google Scholar; Mansurov, V.A., ed., Intelligentsiia o sotsial'no-politicheskoi situalsii v strane: Dannye empiricheskikh issledovanii 1989-1990 gg. (Moscow: Institut Sotsiologii Akademiia Nauk SSSR, 1990).Google Scholar

40. 1497 respondents were interviewed in 21 oblasts and republics, with 105 sampling points. The response rate was 79.5%. Interviews were conducted in November and December 1992.

41. Neuman, The Paradox of Mass Politics, 59.

42. For each data set, this variable is a count (0 to 3) of “don't know” answers to questions about local government. In Taganrog, the questions were: “In general, how often do you suppose that decisions taken by Taganrog city and district agencies coincide with the opinions of broad sectors of the population?“; “In your opinion, to what degree do Taganrog city and district agencies usually consider the opinions of wide segments of the population when solving problems of city life?“; “In your opinion, under what conditions are governmental decisions likely to correspond most fully to popular needs or preferences?” For the first two questions, the available answers were (worded slightly differently in the two cases): practically always, in most cases, half the time, mostly not, practically never, don't know. The third question offered the answers (paraphrased): when material possibilities permit, it depends on how well the population examines the problem, it depends on how important the satisfaction of the population is, it depends on whether leaders think that popular opinion is correct, don't know.

In the SIP survey, the questions were “How many of the people of the local party agencies in your city or town were competent?“; ”… were honest?“; “How many of the people in charge of local Soviet agencies in your city or town were competent?” All three questions provided the options: none, hardly, any, some, most, almost all.

For Leninskii raion, the questions were: “Are you satisfied with the work of the party raikom in solving social problems of Leninskii raion?“; “ … the local executive committee (raiispolkom) …“; “deputies to the local soviet… ?” All three questions provided the options: unsatisfactory, satisfactory, don't know.

43. See Smith, Tom W., “House Effects and the Reproducibility of Survey Measurements: A Comparison of the 1980 GSS and the 1980 American National Election Study,” Public Opinion Quarterly 46 (Spring 1982): 56.CrossRefGoogle Scholar

44. See Hough, Jerry and Fainsod, Merle, How the Soviet Union Is Governed (Cambridge: Harvard University Press, 1979), chap. 8.Google Scholar The questions asked respondents how much they trusted the police, the government of Russia, local police agencies, the Supreme Soviet of Russia, Boris Yeltsin, Ruslan Khasbulatov, and the “Committee for Security” (former KGB). Failures to answer were highest for the former KGB (26%) and Khasbulatov (15%). Otherwise, “don't knows” made up around 10% of the answers to each question.

45. Occupation variables were not included in the regressions because occupation was highly correlated with education in the Soviet Union (see Blagosostoianie sovetskogo naroda: Kratkii statisticheskii sbomik [Moscow: Informatsionno-Izdatel'skii Tsentr, 1990], 11-12). In most cases, occupation added very little to the regressions: it did not explain more of the variance and occupation variables were rarely significant. Income was not included since measures of income tend to be poor or nonexistent in Russian studies, partly because Russian sociologists have already concluded that income explains little, or at least explained little in the past, when income differences were so limited.

Generational categories were used instead of simple numerical ages for a number of reasons: the relationship between age and political attitudes is not necessarily linear, generations provide easier comparisons between surveys undertaken at different points in time and research has shown generational differences to affect political behavior. See Bahry, Donna, “Politics, Generations, and Change in the USSR,” in Millar, James R., ed., Politics, Work, and Daily Life in the USSR: A Survey of Former Soviet Citizens (New York: Cambridge University Press, 1987), 61–99 CrossRefGoogle Scholar; Hough, Jerry F., Soviet Leadership in Transition (Washington: Brookings Institution, 1980)Google Scholar; Rose, Richard and Carnaghan, Ellen, “Generational Effects on Attitudes to Communist Regimes: A Comparative Analysis,” Post-Soviet Affairs 11 (January-March 1995): 28–56.Google Scholar Due to coding differences, the exact generational breakdowns vary slightly between surveys but generally conform to the following divisions: Stalin generation—born before 1918 (1919 for Taganrog); Wartime generation—born 1919-1925 (1920-1929 for Taganrog and Leninskii raion); Postwar generation—born 1926-1940 (1930-1939 for Taganrog and Leninskii raion); 1980s generation—born 1961-1969; Reform generation—born after 1969. The omitted variable is the post-Stalin generation—born 1941-1960. For exact coding of these and other variables, see the appendix in Ellen Carnaghan, Alienation, Apathy or Ambivalence?: ‘Don't Knows’ and Democracy in Russia, Studies in Public Policy 237 (Glasgow: University of Strathclyde, Centre for the Study of Public Policy, 1994).

46. In the Taganrog survey, respondents were asked, “Which of the questions or subjects often addressed by our press, radio, television, lectures, etc., are most interesting to you?” and were allowed to choose multiple listed answers. The variable interest in politics is a count of the number of times they chose “historical revolutionary themes,” “military patriotic subjects,” “Marxist-Leninist propaganda,” “the activities of governmental agencies,” or “the activities of party, trade union, Komsomol and other social organizations.” In the SIP survey, respondents were asked: “During your LNP [“last normal period” before the decision to emigrate], how interested were you in politics and public affairs?” with answers ranging from “very interested” to “not at all interested.“

47. Since the questions in the SIP survey on which the measure of faith in other people was based were asked only of a portion of the overall sample, and since the measure had no significant effect, it is not included in the regressions reported. When “faith in people” is included, the overall regression is much weaker (R²=.17; coefficients for the wartime generation and for unfinished secondary education become insignificant, N=823).

In the Taganrog survey, respondents were asked: “There are different people. Some are not very active in public affairs, rarely expressing their opinions aloud. Others, by contrast, constantly speak out at meetings, write letters to newspapers and to Soviet and party agencies, meet with their deputies, etc. Who, in general, do you suppose these active people are? What qualities would you attribute to them?” Multiple answers were allowed. If respondents chose “these are people who forget about themselves and think first of all about public issues, other people, or the collective” or “these are people who really know and reflect the opinions and desires of the people around them,” they were considered trusting.

48. Of course, it could be true that the really fearful were afraid to express dissatisfaction with the government and, therefore, were not counted as dissatisfied although in fact they were. In fact, if people are so fearful as to never answer questions honestly, we can tell very little from analysis of survey responses. We can take some comfort, however, in the fact that Soviet citizens on the whole were not so fearful as to never answer questions or to only answer questions in ways which reflected well on the state. If Soviet citizens were entirely cowed by their state, the state would not have had to repress the findings of opinion surveys, including Grushin's Taganrog project. See Shlapentokh, Vladimir E., The Politics of Sociology in the Soviet Union (Boulder: Westview Press, 1987), 172.Google Scholar Further, analysis of means shows that those who do not say they are dissatisfied either resemble the satisfied more than the dissatisfied or show very little difference from either group.

For Taganrog, political dissatisfaction was expressed by saying that local decisions rarely or never coincided with popular opinions, or that local officials rarely or never considered public opinion when making their decisions. The questions are the same ones used in counting refusals to evaluate local government. In the SIP survey, respondents were asked to evaluate the competence of six sets of government officials: people in charge of local Soviet agencies, the military, the militsiia, the Politburo of the Central Committee of the Communist Party, the KGB, and local party agencies. This variable is an average of answers that “hardly any” or “none” of these people were competent. In the Moscow survey, this variable is an average of expressions of dissatisfaction to the questions used to count refusals to evaluate local government.

49. In the SIP sample, the correlation coefficient between support for civil liberties and refusals to evaluate local government was —.11, significant at the .001 level. That same variable added to the regressions reported in table 5, however, was not significantly correlated with refusals to evaluate local government. Other measures of support for democracy were not significantly correlated with refusals to answer either singly or in the multiple regression.

For Taganrog and Moscow, support for democracy was a five-point scale. In Taganrog, support for democracy is an average of the responses to the first three questions included in the measure of nonsubstantive responses on democratic values, discussed more extensively in the section below, as long as the respondent did not answer “don't know” more than twice. In the Moscow survey, support for political democracy is an average of valid responses to eight questions, calculated as long as the respondent answered at least four items. The questions (l=agree completely, 2 = agree on the whole, 4 = disagree on the whole, 5 = disagree completely) were: “A few volitional and devoted actions on the part of the leaders of the country can do more for society than all laws and discussions“; “Glasnost’ aggravates internationality problems“; “Martial law is the best solution to national conflicts in any Union republic“; “We should restrict the activities of informal organizations“; “People speaking against the socialist organization of society should receive punishment“; “The preservation of order in government is more important than the protection of individual freedoms for each and everyone“; “The situation in our country would improve if people held fewer demonstrations and worked more“; “Insofar as the CPSU is the ruling party, only communists may occupy higher government posts.“

For the Soviet Interview Project, the variable used is a seven-point measure of support for civil liberties: the right to strike, freedom from residence permits and protection for the rights of people accused of crimes.

50. The addition of other “don't know” measures to the regression equations attenuated the effect of education. Age remains significant only in the SIP equation; gender, only in the Taganrog equation.

51. Arthur Miller, “In Search of Regime Legitimacy,” in Miller, Reisinger, and Hesli, eds., Public Opinion and Regime Change, 104.

52. Activity measures cannot be constructed for the 1989 sample.

53. For the Russia 1992 sample, political dissatisfaction is calculated from the trust “little” or “not at all” answers to the same questions used to measure willingness to evaluate governmental institutions. Interest in politics is a measure analogous to the one in the SIP survey. Faith in people is a measure composed of people who think most people are trustworthy and/or willing to help others. The variable measuring “don't know” answers regarding earnings is also analogous to the one in the SIP survey. The people whose earnings are being evaluated are: collective farmers, workers, apparatchiki, service workers, and enterprise directors. The civil liberties and democratic values questions are discussed more fully in the section below.

54. Converse, “Predicting No Opinion.“

55. See Afanasyev, V.G., et al., Soviet Democracy in the Period of Developed Socialism (Moscow: Progress Publishers, 1979)Google Scholar; Chkhikvadze, V., The State, Democracy and Legality in the USSR: Lenin's Ideas Today (Moscow: Progress Publishers, 1972), 186-88, 202-5Google Scholar; Kerimov, D.A., Mal'tsev, G.V. and Il'inskii, LP., Demokratizatsia sovetskogo obshchestva (Moscow: Mysl', 1989), 3 Google Scholar; Krutogolov, M.A., Talks on Soviet Democracy (Moscow: Progress Publishers, 1980).Google Scholar

56. For all three data sets, this variable is a count of the number of “I find it hard to answer” responses over five questions. The questions vary among the surveys. In Taganrog, the questions were: “In your opinion, how important is it that local officials know the opinions of wide sections of the population on problems of city life?“; “In your opinion, how actively should the population express their opinions on city problems?“; “In your opinion, how necessary is it to take into consideration the opinions of wide sections of the population in solving city problems?” Respondents were also asked on what kinds of questions the population should express its opinion and under what conditions local government decisions should coincide with popular desires. Factor analysis showed that all the questions, recoded into dummy variables separating those who answered substantively from those who did not, loaded on a single factor, with Eigenvalue 2.52, explaining 50% of the variance.

In Moscow, respondents were asked to agree or disagree with the following statements: “Only the development of science and technology can ensure the solution of environmental problems, the ‘greens’ and other informal groups cannot help in this“; “Environmental problems and the protection of nature cannot be solved only with the powers of government, rather it is necessary to include wide sectors of society“; “We should restrict the activities of informal organizations“; “We should remove all restrictions on meetings and demonstrations, whatever political problems are discussed at them.” The first two statements offered three choices for answers: agree, disagree, and hard to say. The second two statements offered five choices: an agreedisagree scale, with “hard to say” in the middle. Respondents were also asked: “If it was proposed to build an object of ail-Union or citywide significance on the territory of a microraion, in what way should the opinions of residents be taken into consideration?” The offered responses were: “through a binding microraion referendum,” “through a decision by the raion soviet,” “through a national or city-level leadership decision,” “other,” “hard to say.” When the variables were coded as dummies (“answer” vs. “don't know“), factor analysis showed that all five loaded on a single factor, with Eigenvalue 1.51, explaining 30% of the variance.

In the Russia 1992 survey, respondents were offered a series of alternatives to choose between: whether people should be able to make their own decisions or whether they need constant guidance, whether political parties are necessary for democracy or not, whether elections are a waste of money or the best way to select officials, whether majorities can understand important political issues or are incapable of figuring out what is good for them. They were also asked to agree or disagree with the question, “A few strong-willed leaders would be more useful to the country, even if they sometimes broke laws, than incessantly arguing democrats.” Most questions al lowed two alternative answers, except for the first, which added the third category “it depends” and the question on purposeful leaders, which provided a four-point scale of agreement or disagreement. When the variables were coded as dummies, factor analysis showed that they loaded on a single factor, with Eigenvalue 1.84, explaining 37% of the variance.

57. In the SIP data, “don't knows” regarding civil liberties is a count of “don't know,” “refused,” or “never thought about” answers to three questions on whether people should be required to have residence permits to live in large cities, whether workers should have the right to strike, and whether the rights of individuals accused of crimes must be protected even if a guilty person sometimes goes free. For each question, a seven-point range of answers was offered, with high support for individual rights on one hand and for state prerogatives on the other. The three variables, recoded to dummy (“know”/“don't know”) form, all loaded on a single factor with Eigenvalue 1.70, explaining 56.5% of the variance.

In the Moscow data, this variable is a count of “hard to say” answers to seven statements with which respondents were asked to agree or disagree (wholly or partly). The statements were: “Almost all newspapers in our country should be absolutely independent of party agencies“; “Party committees do not have the right to concern themselves with the personal and family life of communists“; “A multiparty system is better than a one-party system“; “The system of passports and registration infringes personal freedom and should be liquidated;” “If a person considers that a decision of the Soviet of Ministers infringes upon his legal interests, he should have the right to bring an action against the Soviet of Ministers“; “We should remove all restrictions on meetings and demonstrations, whatever political problems are discussed at them“; “It is desirable that the minister of defense be a professional politician, not a professional army man.” For table 12, three questions are used to more closely parallel the SIP measure.

In the Russia 1992 data, the questions asked respondents to choose which of two alternatives was closer to their own opinions: whether the means of mass communication should show all points of view or only those useful to society, whether free speech should be protected even at the risk of exposing the population to dangerous ideas, whether individuals should be restricted from joining extremist organizations. They were also asked whether workers should have the right to strike and whether the militsiia should have the right to search homes without a warrant. For part of the sample, these last two questions offered three possible answers (right, no right, depends); for the other part of the sample, a four-point agree-disagree scale was offered. The last two questions were omitted from the variable in table 12. Recoded to dummy variables, these five variables loaded on the same factor, with Eigenvalue 2.28, explaining 38% of the variance.

58. Similar effects exist for nonsubstantive answers on civil liberties. “Don't know” answers to other questions are strong predictors of “don't know” answers to civil liberties questions. The regressions are quite robust.

59. For the Moscow 1989 survey, support for socialism is an average of valid answers (omitting “hard to say“) over six questions, calculated as long as the respondent answered at least three of the six. The questions, reworded where necessary to allow consistent coding (1=disagree completely, 2 = disagree on the whole, 4 = agree on the whole, 5 = agree completely), were: “It is necessary to strengthen ideopolitical upbringing with the goal of creating a communist type of personality“; “We should not break up collective farms and give land to peasant families for eternal use“; “People do not have the right not to work“; “We are in debt to the state for its concern for the welfare of the nation“; “There's something wrong with the fact that we now have more millionaire-cooperators“; “Mass unemployment cannot be tolerated, even if it is necessary to pull our economy out of stagnation.“

60. For the three earlier surveys, the measures of support for democracy are discussed above. In the Russia 1992 survey, support for democracy is an average of four of the questions used in the “don't know” measure for democracy, excluding the question on whether majorities can understand their own interests. Factor analysis showed that these four questions loaded on a single factor, with Eigenvalue 1.26, explaining 32% of the variance. For the Russia 1992 survey, support for socialism was based on four questions: whether social justice requires price limits (five possible answers, from “should not be” to “indispensable“); whether a market economy oppresses the poor or provides equal opportunity (two choices); whether everyone has a right not to work (four-point agree-disagree scale); and whether the profits of entrepreneurs should be limited by law (two choices). In the Soviet Interview Project data, support for socialism is an average of valid responses (“hard to say” excluded) to three questions about whether medical care, heavy industry, or agriculture should be private or controlled by the state. Using a measure more analogous to the one in the SIP survey produced virtually no changes in the Russia 1992 regressions.

61. Table 15. In the SIP survey, respondents were asked how satisfied they were with the availability of consumer goods, medical care, and their family's standard of living. In Moscow 1989, respondents were asked to state how public transportation and the availability of goods in Leninskii raion had changed during the preceding three or four years, whether worsened, improved, or stayed the same. In Russia 1992, respondents were asked how satisfied they were with their material position, their income, and their standard of living, as well as whether and how their standard of living had changed over the past five years.