¹ Lepsius, M. Raine, ‘From Fragmented Party Democracy to Government by Emergency Decree and National Socialist Takeover: Germany’, in Linz, Juan J. and Stepan, Alfred, eds, The Breakdown of Democratic Regimes: Europe (Baltimore: Johns Hopkins University Press, 1978), pp. 34–79.Google Scholar

² Sartori, Giovanni, Parties and Party Systems: A Framework for Analysis (New York: Cambridge University Press, 1976).Google Scholar

³ Lipset, Seymour Martin, ‘Some Social Requisites of Democracy: Economic Development and Political Legitimacy’, American Political Science Review, 53 (1959), 69–105.CrossRefGoogle Scholar

⁴ Easton, David, A Systems Analysis of Political Life (New York: Wiley, 1965), pp. 273–4.Google Scholar

⁵ Almond, Gabriel A. and Verba, Sidney, The Civic Culture (Princeton, NJ: Princeton University Press, 1963).CrossRefGoogle Scholar

⁶ On the measurement of Trust in Government, see Miller, Arthur H., ‘Political Issues and Trust in Government’, American Political Science Review, 68 (1974), 951–72.CrossRefGoogle Scholar On the controversy surrounding the validity and reliability of the measure, see Citrin, Jack, ‘Comment: The Political Relevance of Trust in Government’, American Political Science Review, 68 (1974), 973–88CrossRefGoogle Scholar; Miller, Arthur H., ‘Rejoinder to “Comment” by Jack Citrin: Political Discontent or Ritualism?’, American Political Science Review, 68 (1974), 989–1001CrossRefGoogle Scholar; Barnes, Samuel H. et al. , Political Action: Mass Participation in Five Western Democracies (Beverly Hills, Calif.: Sage, 1979)Google Scholar; Abramson, Paul and Finifter, Ada, ‘On the Meaning of Political Trust: New Evidence from Items Introduced in 1978’, American Journal of Political Science, 25 (1981), 297–307CrossRefGoogle Scholar; Muller, Edward N., Jukam, Thomas O. and Seligson, Mitchell A., ‘Diffuse Political Support and Antisystem Political Behavior: A Comparative Analysis’, American Journal of Political Science, 26 (1982), 240–64CrossRefGoogle Scholar; and Seligson, Mitchell A., ‘On the Measurement of Diffuse Support: Some Evidence from Mexico’, Social Indicators Research, 12 (1983), 1–24.CrossRefGoogle Scholar

⁷ See, for example, Muller, Edward N. and Williams, Carol, ‘Dynamics of Political Support-Alienation’, Comparative Political Studies, 13 (1980), 33–59.CrossRefGoogle Scholar

⁸ See Muller, Edward N., Aggressive Political Participation (Princeton: Princeton University Press, 1979)Google Scholar; Muller, Edward N. and Jukam, Thomas O., ‘On the Meaning of Political Support’, American Political Science Review, 71 (1977), 1561–95CrossRefGoogle Scholar; Muller, , Jukam, and Seligson, , ‘Diffuse Political Support and Antisystem Political Behavior’Google Scholar; and Seligson, , ‘On the Measurement of Diffuse Support: Some Evidence from Mexico’.Google Scholar

⁹ See Seligson, Mitchell A., ‘Costa Rica’, in Hopkins, Jack W., ed., Latin American and Caribbean Contemporary Record, Volume I: 1981–82 (New York: Holmes and Meier, 1983)Google Scholar; and Seligson, Mitchell A., ‘Costa Rica’, in Hopkins, Jack W., ed., Latin American and Caribbean Contemporary Record, Volume II: 1982–83 (New York: Holmes and Meier, 1984).Google Scholar

¹⁰ For details, see Seligson, Mitchell A. and Muller, Edward N., ‘Democratic Stability and Economic Crisis: Costa Rica, 1978–1983’, International Studies Quarterly, 31 (1987), 301–26CrossRefGoogle Scholar; and Seligson, Mitchell A. and Gómez B., Miguel, ‘Eleciones ordinarias en tiempos extraordinarios: La economía politica del voto en Costa Rica’, Anuario de estudios centroamericanos, 13 (1987), 5–24.Google Scholar

¹¹ Céspedes, V. H., Vega, C. G., Jiménez, R. and Lizano, E., Costa Rica: Estabilida sin Crecimiento (San José, Costa Rica: Academia de Centroamerica, 1984).Google Scholar See also Gudmundson, , ‘Costa Rica’, in Hopkins, Jack W., ed., Latin American and Caribbean Contemporary Record, Volume III: 1983–84 (New York: Homes and Meier, 1985).Google Scholar

¹² The response rates for the panels were affected not only by limited funds but also by the difficulties in locating respondents from the 1978 and 1983 surveys, neither of which were conducted with a panel design in mind. In regard to basic SES and attitudinal characteristics, however, the panel respondents differ little from the cross-section respondents, as the figures in the Appendix Table make clear. The only significant differences are in age, where the 1978–85 panel contains older respondents on average (since younger individuals could not, obviously, have been included in the original interview), and a greater proportion of primary education respondents in the cross-section than in either panel survey. But the equal proportion of university respondents is reassuring, because the generally lower degree of regime and performance support found in the university setting could have distorted the generalizability of the findings reported here. On all other variables, the distribution in the three samples are very similar.

¹³ Heidenheimer, Arnold, Heclo, Hugh and Adams, Carolyn, Comparative Public Policy (New York: St Martin's Press, 1983), pp. 126–33.Google Scholar

¹⁴ For full details on the West German sample, see Muller, , Aggressive Political Participation, pp. 8–9.Google Scholar

¹⁵ Anti-system behaviour in the rural and urban sites entailed voting for political parties of the extreme left or right; in the university sites it entailed civil disobedience and political violence. See Muller, , Aggressive Political Participation, p. 9.Google Scholar

¹⁶ See Seligson, and Muller, , ‘Democratic Stability and Economic Crisis: Costa Rica, 1978–1983’, pp. 311–13.Google Scholar

¹⁷ Seligson, and Gomez, , ‘Eleciones ordinarias en tiempos extraordinarios: la economia politica del voto en Costa Rica’.Google Scholar

¹⁸ It can be shown, with some very simple algebraic manipulations, that the following model reduces to the static score model where Regime Support in 1985 is a function of Regime Support in the previous survey, Performance Satisfaction in 1985 and Performance Satisfaction in the previous survey, and the control variables. See Kessler, Ronald and Greenberg, David, Linear Panel Analysis (New York: Academic Press, 1981), pp. 7–12.CrossRefGoogle Scholar Kessler and Greenberg note (p. 11) that ‘linear difference equations (those involving change in X as dependent variable) and equations involving only static scores (X₁ and X₂) are mathematically equivalent and can readily be transformed into one another’. Analysis of static scores is quite common with panel data (cf., e.g., Markus, Gregory, Analyzing Panel Data (Beverly Hills, Calif.: Sage Publications, 1980)Google Scholar; and Finkel, Steven E., ‘The Effects of Participation on Political Efficacy and Political Support: Evidence From a West German Panel’, Journal of Politics, 49 (1987), 441–64)CrossRefGoogle Scholar, but in this case, we use the change score transformations because our theoretical interest is in describing change in regime support as a function of change in performance satisfaction during the economic crisis.

¹⁹ Because of the relatively small number of cases in the panel surveys, we used mean substitution for missing values, which saved seven cases for the 1978–85 analysis and five for the 1983–85 sample.

²⁰ By ‘natural’ regression effects, we mean the tendency of initially high scores on a variable to decline over time, and the tendency of initially low scores on a variable to increase. One reason this pattern occurs is that initially extreme scores are caused in part by a very large positive or negative error term. When the variable is measured at another point in time, there is a high likelihood that the error term will be smaller; hence extreme scores tend to become less extreme over time. Regression to the mean can also result from measurement error and the effects of omitted variables which are captured in the error term. See Markus, , Analyzing Panel Data, pp. 45–8.Google Scholar

²¹ If an individual increased in Performance Satisfaction, then the positive change variable equalled the amount of change; otherwise, the positive change variable was 0. If an individual decreased in Performance Satisfaction, then the negative change variable equalled the amount of change; otherwise, the negative change variable was 0. Thus, the positive signs for the coefficients for the Positive Change variables in Tables 1 and 2 indicate that as individuals become more positive in performance satisfaction, they become more favourable towards the regime. The positive signs for the coefficients for the Negative Change variables indicate that as individuals evaluate incumbents more negatively, they become less favourable towards the regime.

²² On negativity in political evaluations, see Lau, Richard, ‘Negativity Effects in Political behavior’, American Journal of Political Science, 29 (1985), 119–38CrossRefGoogle Scholar; and Kernell, Samuel, ‘Presidential Popularity and Negative Voting’, American Political Science Review, 71 (1977), 44–66.CrossRefGoogle Scholar

²³ See Lewis-Beck, Michael. ‘Comparative Economic Voting: Britain, France, Germany, Italy’, American Journal of Political Science, 30 (1986), 315–46CrossRefGoogle Scholar; and Kuechler, Manfred, ‘A Trend Analysis of System Support in West Germany’, paper presented at the Annual Meetings of the Southern Political Science Association, Atlanta, Georgia, 11 1986.Google Scholar

²⁴ See Lewis-Beck, , ‘Comparative Economic Voting: Britain, France, Germany, Italy’Google Scholar; Monroe, Kristen, ed., The Political Process and Economic Change (New York: Agathon Press, 1983)Google Scholar; and the many references cited in those works.

²⁵ The slight positive effect of change in performance evaluation on change in regime support in both West Germany and Costa Rica supports theoretical speculation about the role of effectiveness as a source of legitimacy sentiment. See Conradt, David, ‘Changing German Political Culture’, in Almond, Gabriel and Verba, Sidney, eds, The Civic Culture Revisited (Boston, Mass.: Little, Brown, 1980)Google Scholar; and Malloy, James M. and Seligson, Mitchell A., Authoritarians and Democrats: Regime Transition in Latin America (Pittsburgh, Penn.: University of Pittsburgh Press, 1987).Google Scholar