¹ Converse, Philip E., ‘Information Flow and the Stability of Partisan Attitudes’, Public Opinion Quarterly, 26 (1962), 578–99CrossRefGoogle Scholar; Converse, Philip E., ‘The Concept of a Normal Vote’, in Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald, Elections and the Political Order (New York: Wiley, 1966), pp. 9–39Google Scholar; Falter, Jürgen W. and Rattinger, Hans, ‘Parties, Candidates and Issues in the German Federal Election of 1980: An Application of Normal Vote Analysis’, Electoral Studies, I (1982), 65–94CrossRefGoogle Scholar; Zaller, John R., The Nature and Origins of Mass Opinion (Cambridge: Cambridge University Press, 1992).CrossRefGoogle Scholar

² See, for example, Alger, Dean E., The Media and Politics (Englewood Cliffs, NJ: Prentice Hall, 1989)Google Scholar; Patterson, Thomas E., ‘The Press and Its Missed Assignment’, in Nelson, Michael, ed., The Elections of 1988 (Washington, DC: Congressional Quarterly Press, 1989), pp. 93–110.Google Scholar

³ Popkin, Samuel, The Reasoning Voter: Communication and Persuasion in Presidential Campaigns (Chicago: University of Chicago Press, 1992)Google Scholar; McCubbins, Mathew D., ‘Party Decline and Presidential Campaigns in the Television Age’, in McCubbins, Mathew D., ed., Under the Watchful Eye (Washington, DC: Congressional Quarterly Press, 1992), pp. 9–58.Google Scholar

⁴ Lazarsfeld, Paul, Berelson, Bernard and Gaudet, Hazel, The People's Choice (New York: Columbia University Press, 1944)Google Scholar; Berelson, Bernard R., Lazarsfeld, Paul F. and McPhee, William N., Voting (Chicago: University of Chicago Press, 1954)Google Scholar; Mendelsohn, Harold and O'Keefe, Garrett J., The People Choose a President (New York: Praeger, 1976)Google Scholar; Patterson, Thomas E. and McClure, Robert D., The Unseeing Eye (New York: Putnam, 1976)Google Scholar; Patterson, Thomas E., The Mass Media Election (New York: Praeger, 1980).Google Scholar

⁵ See the reviews in Graber, Doris, ‘Media and Politics’, in Crotty, William, ed., Political Science: Looking to the Future, Volume 3 (Evanston, Ill.: Northwestern University Press, 1991), pp. 91–124Google Scholar; Graber, Doris, ‘Political Communication: Scope, Progress, Promise’, in Finifter, Ada W., ed., Political Science: The State of the Discipline II (Washington, DC: American Political Science Association, 1993), pp. 305–32Google Scholar; also Bartels, Larry M., ‘Messages Received: The Political Impact of Media Exposure’, American Political Science Review, 87 (1993), 267–85CrossRefGoogle Scholar; Farah, Barbara G. and Klein, Ethel, ‘Public Opinion Trends’, in Pomper, Gerald, ed., The Election of 1988: Reports and Interpretations (Chatham, NJ: Chatham House, 1989), pp. 103–28Google Scholar; Frankovic, Kathleen, ‘Public Opinion in the 1992 Campaign’, in Pomper, Gerald, ed., The Election of 1992 (Chatham, NJ: Chatham House, 1993), pp. 110–31Google Scholar; Granberg, Donald and Holmberg, Soeren, The Political System Matters (Cambridge: Cambridge University Press, 1988)Google Scholar; Markus, Gregory B. and Converse, Philip E., ‘A Dynamic Simultaneous Equation Model of Electoral Choice’, American Political Science Review, 73 (1979), 1055–70CrossRefGoogle Scholar; Markus, Gregory B., ‘Political Attitudes During an Election Year: A Report on the 1980 NES Panel Study’, American Political Science Review, 76 (1982), 538–60CrossRefGoogle Scholar; Schrott, Peter R., ‘Electoral Consequences of “Winning” Televised Campaign Debates’, Public Opinion Quarterly, 54 (1990), 567–85.CrossRefGoogle Scholar

⁶ Finkel, Steven E., ‘Reexamining the “Minimal Effects” Model in Recent Presidential Campaigns’, Journal of Politics, 55 (1993), 1–21CrossRefGoogle Scholar; Lazarsfeld, , Berelson, , and Gaudet, , The People's Choice, pp. 73–104Google Scholar; Berelson, , Lazarsfeld, and McPhee, , Voting, pp. 280–96.Google Scholar

⁷ Lazersfeld, , Berelson, and Gaudet, , The People's Choice, pp. 83–4.Google Scholar

⁸ Gelman, Andrew and King, Gary, ‘Why Are American Campaign Polls So Variable When Votes Are So Predictable?’ British Journal of Political Science, 23 (1993), 409–31, p. 433.CrossRefGoogle Scholar

⁹ Schönbach, Klaus, ‘Mass Media in German Election Campaigns’, in Fletcher, Frederick J., ed., Media, Elections and Democracy (Toronto: Dundurn Press, 1991), pp. 63–86.Google Scholar

¹⁰ Conradt, David P., ‘The Christian Democrats in 1990: Saved by Unification?’ in Dalton, Russell J., ed., The New Germany Votes (Providence, RI: Berg, 1993), pp. 59–76Google Scholar; Hancock, M. Donald, ‘The SPD Seeks a New Identity: Party Modernization and the Prospects in the 1990s’Google Scholar, in Dalton, , ed., The New Germany Votes, pp. 77–98Google Scholar; Kuechler, Manfred, ‘Framing Unification: Issue Salience and Mass Sentiment 1989–1991’Google Scholar, in Dalton, , ed., The New Germany Votes, pp. 29–55Google Scholar; Norpoth, Helmut and Roth, Dieter, ‘Unification and Electoral Choice’Google Scholar, in Dalton, , ed., The New Germany Votes, pp. 209–30Google Scholar; Semetko, Holli and Schönbach, Klaus, ‘Media Use and Electors' Opinions in the German 1990 National Elections’ (paper presented at the Annual Meeting of the American Political Science Association, Washington, DC, 1991).Google Scholar

¹¹ Kaase, Max, ‘The Election to the German Bundestag of December 1990’Google Scholar, in Kaltenhaler, Karl Christian and Luthardt, Wolfgang, eds, The Politics of German Unification (forthcoming)Google Scholar; Kuechler, Manfred, ‘Anschluss Ueber Alles? What Did the Voter Decide in the 1990 German Elections?’ (paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago, Ill., 1991).Google Scholar

¹² Schönbach, , ‘Mass Media in German Election Campaigns’, p. 69.Google Scholar

¹³ Kaase, Max, ‘Massenkommunikation und politischer Prozess’, in Kaase, Max, ed., Politische Wissenschaft und Politische Ordnung (Opladen: Westdeutscher Verlag, 1986), pp. 357–74CrossRefGoogle Scholar; Schönbach, , ‘Mass Media in German Election Campaigns’.Google Scholar

¹⁴ Schönbach, , ‘Mass Media in German Election Campaigns’, p. 77Google Scholar. On the role of candidate images, see Kepplinger, Hans Mathias, Dahlem, Stefan and Brosius, Hans-Bernd, ‘Helmut Kohl und Oskar Lafontaine im Fernsehen: Ein Experiment mil Teilnehmern in den alten und neuen Bundesländern’, in Holtz-Bacha, Christina and Kaid, Lynda Lee, eds. Die Massenmedien im Wahlkampf: Untersuchungen aus dem Wahljahr 1990 (Opladen: Westdeutscher Verlag, 1992), pp. 144–84Google Scholar; Schrott, Peter R., ‘Wahlkampfdebatten im Fernsehen von 1972 bis 1987; Politikerstrategien und Wählerreaktion’, in Kaase, Max and Klingemann, Hans-Dieter, eds, Wahlen und Wähler: Analysen aus Anlass der Bundestagswahl 1987 (Opladen: Westdeutscher Verlag, 1990), pp. 647–74CrossRefGoogle Scholar; Schulz, Winfried and Kindelmann, Klaus, ‘Die Entwicklung der Images von Kohl und Lafontaine im Wahljahr: Ein Vergleich der Waehlerurteile mit den Urteilen ausgewaehlter Leitmedien’Google Scholar, in Holtz-Bacha, and Kaid, , eds, Die Massenmedien im Wahlkampf, pp. 10–45Google Scholar. On the role of issue awareness and changes in the public's issue agenda, see Kepplinger, Hans Mathias and Brosius, Hans-Bernd, ‘Der Einfluss der Parteibindung und der Fernsehberichterstattung auf die Wahlabsichten der Bevölkerung’Google Scholar, in Kaase, and Klingemann, , eds, Wahlen und Wähler 1987, pp. 675–86Google Scholar; Semetko, Holli A. and Schönbach, Klaus, Germany's ‘Unity Election’: Voters and the Media (Cresskill, NJ: Hampton, 1994)Google Scholar. On the influence of a campaign's opinion climate, see, e.g., Noelle-Neumann, Elisabeth, ‘Massenmedien und Meinungsklima im Wahlkampf’, in Schulz, Winfried and Schönbach, Klaus, eds, Massenmedien und Wahlen (Munich: Oelschlaeger, 1983), pp. 337–405.Google Scholar

¹⁵ Two panel studies from the 1990 campaign do analyse short-term changes in voter evaluations and preferences at the individual level. See Semetko, and Schönbach, , Germany's ‘Unity’ Election: Voters and the MediaGoogle Scholar, and Schmitt-Beck, Rüdiger and Schrott, Peter R., ‘Dealignment durch Massenmedien? Zur These der Abschwächung von Parteibindungen als Folge der Medienexpansion’, in Klingemann, Hans-Dieter and Kaase, Max, eds, Wahlen und Wähler: Analysen aus Anlass der Bundestagaswahl 1990 (Opladen: Westdeutscher Verlag, 1994), pp. 543–72CrossRefGoogle Scholar. Neither study, however, analyses the overall effect of the campaign on voter choice.

¹⁶ Gibowski, Wolfgang G. and Kaase, Max, ‘Auf dem Weg zum politischen Alltag: Eine Analyse der ersten gesamtdeutschen Bundestagswahl 1990’, Aus Politik und Zeitgeschichte, 11–12 (1991), 3–20Google Scholar; Abramson, Paul R., Aldrich, John and Rhode, David, Change and Continuity in the 1988 Elections (Washington, DC: Congressional Quarterly Press, 1990)Google Scholar. Our empirical analyses also confirm that such traditionally important demographic factors as religion, church attendance, age and education are statistically insignificant predictors of the 1990 German vote, once the political attitudes described below are entered in the model.

¹⁷ Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald, The American Voter (New York: Wiley, 1960).Google Scholar

¹⁸ Falter, and Rattinger, , ‘Parties, Candidates and Issues in the German Federal Election of 1980’, p. 68Google Scholar. See also Baker, Kendall L., Dalton, Russell J. and Hildebrandt, Kai, Germany Transformed: Political Culture and the New Politics (Cambridge, Mass.: Harvard University Press, 1981)Google Scholar; Norpoth, Helmut, ‘Party Identification in West Germany: Tracing an Elusive Concept’, Comparative Political Studies, II (1978), 36–61.CrossRefGoogle Scholar

¹⁹ Dalton, Russell, ‘The German Voter’, in Smith, Gordon, Paterson, William E., and Merkl, Peter H., eds. Developments in West German Politics (Durham, NC: Duke University Press, 1989), pp. 99–121CrossRefGoogle Scholar; Klingemann, and Wattenberg, , ‘Decaying Versus Developing Party Systems: Comparison of Party Images in the United States and West Germany’, British Journal of Political Science, 22 (1992), 131–49CrossRefGoogle Scholar; Richardson, Bradley, ‘European Party Loyalties Revisited’, American Political Science Review, 85 (1991), 751–75.CrossRefGoogle Scholar

²⁰ Fuchs, Dieter and Kühnel, Steffen M., ‘Die evaluative Bedeutung ideologischer Selbstidentifikation’, in Kaase and Klingemann, eds, Wahlen und Wähler, 1987, pp. 217–53Google Scholar; Fuchs, Dieter and Kühnel, Steffen M., ‘Wählen als rationales Handeln: Anmergungen zum Nutzen des Rational-Choice-Ansatzes in der empirischen Wahlforschung’Google Scholar, in Klingemann, and Kaase, , eds, Wahlen und Wähler, 1990, pp. 305–63.Google Scholar

²¹ Fiorina, Morris, Retrospective Voting in American National Elections (New Haven, Conn.: Yale University Press, 1981)Google Scholar; Fuchs, and Kühnel, , ‘Wählen als rationales Handeln’Google Scholar; Kirchgässner, Gebhard, ‘Der Einfluss wirtschaftlicher Variabler auf die Popularität der Parteien’, in Falter, Jürgen W., Rattinger, Hans and Troitzsch, Klaus G., eds, Wahlen und politische Einstellungen in der Bundesrepublik Deutschland (Frankfurt: Verlag Peter Lang, 1989)Google Scholar; Norpoth, Helmut and Yantek, Thorn, ‘Von Adenauer bis Schmidt: Wirtschaftslage und Kanzlerpopularität’, in Kaase, Max und Klingemann, Hans-Dieter, eds, Wahlen und politisches System: Analysen aus Anlass der Bundestagswahl 1980 (Opladen: Westdeutscher Verlag, 1983).Google Scholar

²² Finkel, , ‘Reexamining the “Minimal Effects” Model in Recent Presidential Campaigns’Google Scholar; Gelman, and King, , ‘Why Are American Campaign Polls So Variable When Votes Are So Predictable?’Google Scholar; Lewis-Beck, Michael S., Economics and Elections (Ann Arbor: University of Michigan Press, 1989)Google Scholar; Markus, Gregory B., ‘The Impact of Personal and National Economic Conditions on the Presidential Vote: A Pooled Cross-Sectional Analysis’, American Journal of Political Science, 32 (1988), 137–54.CrossRefGoogle Scholar

²³ Klingemann, and Wattenberg, , ‘Decaying Versus Developing Party Systems’Google Scholar; Falter, and Rattinger, , ‘Parties, Candidates and Issues in the German Federal Election of 1980’.Google Scholar

²⁴ Radunski, Peter, Wahlkämpfe: Moderne Wahlkampfführung als Politische Kommunikation (Munich: Olzog, 1980)Google Scholar; Schulz, and Kindelmann, , ‘Die Entwicklung der Images von Kohl und Lafontaine im Wahljahr: Ein Vergleich der Wählerurteile mit den Urteilen ausgewählter Leitmedien’Google Scholar; Semetko, and Schönbach, , Germany's Unity Election.Google Scholar

²⁵ Strictly speaking, the CDU and CSU are separate parties, but it is customary to treat them as one, since the CDU does not compete in the state of Bavaria and the CSU does not compete in the rest of the country.

²⁶ Dalton, , ‘The German Voter’Google Scholar; Salmore, Barbara G. and Salmore, Stephen A., Candidates, Parties, and Campaigns (Washington, DC: Congressional Quarterly Press, 1989).Google Scholar

²⁷ Finkel, , ‘Reexamining the “Minimal Effects” Model in Recent Presidential Campaigns’Google Scholar; Lazarsfeld, , Berelson, and Gaudet, , The People's ChoiceGoogle Scholar; Berelson, , Lazarsfeld, and McPhee, , Voting.Google Scholar

²⁸ Berelson, , Lazarsfeld, and McPhee, , Voting, p. 282.Google Scholar

²⁹ Lazarsfeld, , Berelson, and Gaudet, , The People's Choice, p. 69Google Scholar; Converse, , ‘Information Flow and the Stability of Partisan Attitudes’Google Scholar; Norpoth, Helmut and Baker, Kendall L., ‘Mass Media Use and Electoral Choice in West Germany’, Comparative Politics, 13 (1980), 1–14CrossRefGoogle Scholar; Zaller, , The Nature and Origins of Mass Opinion.Google Scholar

³⁰ Zaller, , The Nature and Origins of Mass OpinionGoogle Scholar; Zaller, John R., ‘Bringing Converse Back In: Information Flow in Political Campaigns’, Political Analysis, I (1989), 181–234.CrossRefGoogle Scholar

³¹ Zaller, , The Nature and Origins of Mass Opinion, pp. 250–1; see also Hypothesis 2 above.Google Scholar

³² Norpoth, and Baker, , ‘Mass Media Use and Electoral Choice in West Germany’Google Scholar, confirm the ‘floating voter’ hypothesis in German elections by showing that lower levels of media exposure are associated with changes in individual vote choice between elections. Our analysis examines similar processes within campaigns.

³³ This study was conducted by the Forschungsgruppe Wahlen, together with Max Kaase (Mannheim), Hans Dieter Klingemann (Berlin), Manfred Kuechler (New York), Franz Urban Pappi (Mannheim) and Holli Semetko (Michigan). The data were prepared and made publicly available by the Zentralachriv für Empirische Sozialforschung at the Universität zu Köln.

³⁴ Kuechler, , ‘Anschluss Ueber Alles?’, pp. 31–6.Google Scholar

³⁵ The official vote returns for both West and East Gennany may be found in Forschungsgruppe Wahlen, Bundestagswahl 1990: Eine Analyse der ersten gesamtdeutschen Bundestagswahl am 2. Dezember 1990 (Mannheim: Berichte der Forschungsgruppe Wahlen e.V., No. 61, 1990), p. 8.Google Scholar

³⁶ This result for the Greens meant that, according to the rules of the German electoral system, they did not clear the requisite 5 per cent hurdle to win seats in the Bundestag.

³⁷ Aldrich, John and Nelson, Forrest, Linear Probability, Logit and Probit Models (Beverly Hills, Calif.: Sage Publications, 1984).CrossRefGoogle Scholar

³⁸ Achen, Christopher, Interpreting and Using Regression (Beverly Hills, Calif.: Sage Publications, 1982), pp. 71–3CrossRefGoogle Scholar; Denk, Charles E. and Finkel, Steven E., ‘The Aggregate Impact of Explanatory Variables in Logit and Linear Probability Models’, American Journal of Political Science, 36 (1992), 785–804CrossRefGoogle Scholar. As in Finkel's previous article, these ‘impact analyses’ produce the net effect of the independent variables on the level of the dependent variable compared to a hypothetical ‘neutral’ electorate; i.e., if all individuals were at ‘0’ on the particular independent variable. In the case of the change scores, this means that the ‘impact’ of the variables is how much the actual change exhibited over time in the sample contributed to the net aggregate outcome, in comparison to a hypothetical electorate which would have shown no campaign period changes whatsoever. See Finkel, , ‘Reexamining the “Minimal Effects” Model in Recent Presidential Campaigns’, pp. 9–12.Google Scholar

³⁹ Kuechler, , ‘Framing Unification’Google Scholar; Norpoth, and Roth, , ‘Unification and Electoral Choice’Google Scholar; Schulz, and Kindelman, , ‘Die Entwicklung der Images von Kohl und Lafontaine im Wahljahr: Ein Vergleich der Wählerurteile mit den Urteilen ausgewählter Leitmedien’.Google Scholar

⁴⁰ Kuechler, , ‘Framing Unification’, pp. 46–7.Google Scholar

⁴¹ Kitschelt, Herbert, ‘The 1990 German Federal Election and the National Unification’, West European Politics, 14 (1991), 121–48, p. 131CrossRefGoogle Scholar; on the relative campaign efforts of the contending parties, see also Conradt, , ‘The Christian Democrats in 1990’Google Scholar; and Hancock, , ‘The SPD Seeks a New Identity’Google Scholar. See Norpoth, and Roth, , ‘Unification and Electoral Choice’Google Scholar, for an analysis of the ‘rally effect’ in public opinion as a result of Kohl's successful unification policies.

⁴² This procedure actually generates a predicted vote probability for each individual, and those with scores greater than 0.5 were predicted to vote for the government coalition, and those with scores less than 0.5 were predicted to vote for the opposition.

⁴³ Finkel, , ‘Reexamining the “Minimal Effects” Model in Recent Presidential Campaigns’, p. 11.Google Scholar

⁴⁴ A predicted vote probability was calculated for each individual from the full equation of Table 1, including all change-score variables. The correlation between the pre- and post-campaign probability estimates was 0.93.

⁴⁵ Interestingly, further analyses suggests that there may be interaction effects between the media variables and June levels of party identification, ideology and candidate evaluations, such that the June variables influence attitude change to a greater extent among highly attentive than among inattentive respondents. This pattern would confirm the findings of greater polarization of partisan attitudes among more aware individuals during US campaigns in Zaller, The Nature and Origins of Mass Opinion. Unfortunately, the high multicollinearity between the interaction terms necessary to test these hypotheses leads to difficulties in the estimation of the model and some uninterpretable results.

⁴⁶ Finkel, , ‘Reexamining the “Minimal Effects” Model’, p. 11Google Scholar; Zaller, , The Nature and Origins of Mass Opinion.Google Scholar

⁴⁷ The full model's GLS coefficients produce an estimated vote for the government parties of 52.7 per cent, of which 2.2 percentage points are directly attributable to campaign-period attitudinal changes. The actual vote for the government parties was 54.3 per cent, and thus it is possible that the 1.6 percentage point difference between the predicted and actual votes is also the result of unspecified campaign factors, bringing the total ‘campaign effect’ to 3.8 per cent. It is also possible, however, that the difference arises from the weighting process associated with GLS estimation, which may induce some slight mathematical discrepancies in accounting for the mean of the dependent variable from the levels of the independent variable multiplied by their respective unstandardized coefficients, i.e., in the equation .

⁴⁸ Kaase, , ‘The Election to the German Bundestag of December 1990’Google Scholar; Kuechler, Manfred, ‘Politische Streitfragen und Wahlentscheidung: Veriningung als “neue” Frage?’Google Scholar in Klingemann, and Kaase, (eds), Wahlen und Wähler 1990, pp. 422–51.Google Scholar

⁴⁹ Granberg, and Holmberg, , The Political System Matters, chap. 8.Google Scholar

⁵⁰ Finkel, , ‘Reexamining the “Minimal Effects” Model’, p. 15.Google Scholar

⁵¹ The ‘strength of initial dispositions’ variable was created by folding the June predicted probability of voting for the incumbent coalition around 0.5, so that values that were either very large (i.e., pro-government) or very small (i.e., pro-opposition) become large values on the strength variable, and values close to 0.5 on either side become small values on the strength variable.

⁵² Lazarsfeld, , Berelson, and Gaudet, , The People's ChoiceGoogle Scholar; Berelson, , Lazarsfeld, and McPhee, , VotingGoogle Scholar; Zaller, , The Nature and Origins of Mass Opinion.Google Scholar