Do Campaigns Matter?
Publication Year: 1996
Subject: Elections & Political Campaigns
A thorough examination of the impact of campaign politics on presidential elections in the United States is presented in this book. Using actual election results and empirical evidence, the author also incorporates data on additional factors such as media coverage, the impact of nominating conventions on public opinion, presidential debates, and other events such as staff shake-ups, endorsements and scandals. In so doing, Holbrook develops a model for testing campaigns and proves how campaigns play a key role in shaping public opinion and, ultimately, influencing outcomes.
- Front Matter
- Back Matter
- Subject Index
- Chapter 1: Campaigns and Elections
- Postelection Analyses
- The Argument against Campaign Effects
- The Argument for Campaign Effects
- Who's Right?
- The Plan of This Book
- Chapter 2: Evidence against Campaign Effects
- Aggregate Models
- Individual-Level Evidence
- Campaign Effects?
- Chapter 3: A Model of Campaign Effects
- A Model for Finding Campaign Effects
- The Role of Information
- The Campaign Process
- Information Trends in Presidential Campaigns
- Chapter 4: Nominating Conventions
- Nominating Conventions, 1984–1992
- The Electoral Payoff
- Information and Convention Bumps
- Explaining Convention Bumps, 1964–1992
- Chapter 5: The Effect of Debates
- Debates and Information
- Contemporary Debates
- The Aggregate Effect of Debates
- Individual-Level Evidence
- Chapter 6: Campaigns, National Conditions, and U.S. Presidential Elections
- Subordinate Campaign Effects
- National Conditions
- A Model of Candidate Support in Presidential Campaigns
- Analysis of Model
- Chapter 7: The Role of Campaigns in U.S. Presidential Elections
- A Recapitulation of Hypotheses and Supporting Evidence
- The Campaign Process
Contemporary American Politics[Page ii]Series Editors
- Richard G. Niemi, University of Rochester
- Barbara Sinclair, University of California, Riverside
- John Aldrich, Duke University
- Gregory Caldeira, The Ohio State University
- Stanley Feldman, SUNY Stony Brook
- Katherine Tate, The Ohio State University
- Sue Thomas, Georgetown University
The Contemporary American Politics series is intended to assist students and faculty in the field of American politics by bridging the gap between advanced but oft-times impenetrable research on the one hand, but oversimplified presentations on the other. The volumes in this series represent the most exciting work in political science—cutting-edge research that focuses on major unresolved questions, contradicts conventional wisdom, or initiates new areas of investigation. Ideal as supplemental texts for undergraduate courses, these volumes will examine the institutions, processes, and policy questions that make up the American political landscape.
Books in This Series
DO CAMPAIGNS MATTER?
Thomas M. Holbrook
GENDER DYNAMICS IN CONGRESSIONAL ELECTIONS
Richard Logan Fox
Copyright © 1996 Sage Publications, Inc.
All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the publisher.
For information address:
SAGE Publications, Inc.
2455 Teller Road
Thousand Oaks, California 91320
SAGE Publications Ltd.
6 Bonhill Street
London EC2A 4PU
SAGE Publications India Pvt. Ltd.
Greater Kailash I
New Delhi 110 048 India
Printed in the United States of America
Library of Congress Cataloging-in-Publication Data
Holbrook, Thomas M.
Do campaigns matter? / author, Thomas M. Holbrook.
p. cm. — (Contemporary American politics)
Includes bibliographical references and index.
ISBN 0-8039-7344-6 (cloth: acid-free paper). —
ISBN 0-8039-7345-4 (pbk.: acid-free paper)
1. Presidents—United States—Election. 2. Electioneering—United States. 3. Voting—United States. I. Title. II. Series.
JK524.H65 1996 96-4469
This book is printed on acid-free paper.
96 97 98 99 10 9 8 7 6 5 4 3 2 1
Sage Production Editor: Gillian Dickens
Sage Typesetter: Janelle LeMaster
[Page v]This book is dedicated to the memory of my friend, colleague, and mentor, Charley Tidmarch.[Page vi]
Series Editor's Introduction[Page ix]
This new Sage series on Contemporary American Politics is intended to convey the most exciting work in political science—“cutting-edge research focusing on major unresolved questions or research that contradicts conventional wisdom or that initiates entirely new areas of investigation.” In this respect, Thomas Holbrook's Do Campaigns Matter? represents a great beginning. In the past several decades of work on voting and elections, little attention has been paid to the campaigns themselves. Occasionally, explicit statements have suggested that the campaign has had only a negligible effect on the voters; more often, there was simply puzzlement—wonderment at how much attention parties and candidates paid to the campaign when we were unable to demonstrate that it had much effect. Holbrook's book breaks that mold, demonstrating that campaigns have a variety of effects on voters.
[Page x]Do Campaigns Matter? is also an excellent beginning in another sense. The series is intended to convey cutting-edge research in a style that “bridges the gap between advanced, but sometimes impenetrable research and understandable [though] greatly simplified presentations.” Holbrook does not shy away from complex ideas and materials, but his findings are presented in a way that is accessible to readers at various levels of learning in regard to statistical skills.
One important feature of Holbrook's work is his innovative combination of a variety of kinds of data about the campaign, including trial heats, a record of important campaign events, newspaper reports about the campaign, and surveys of voter behavior. By combining these elements, Holbrook paints a more realistic picture of campaigns and of their effects than is possible with only one or two kinds of data.
As yet another long presidential campaign is upon us, we are delighted to inaugurate our series with a volume that shows that all our collective efforts are not without consequence. Presidential campaigns do matter; it is surprising that we had to wait so many years for political scientists to recognize that fact.—
I began thinking about the ideas presented in this book several years ago, when I was a graduate student at the University of Iowa. In fact, I considered writing my dissertation on the effect of campaigns on presidential elections. There was one small problem, however: I didn't have the foggiest idea about how one would go about assessing the effect of campaigns. I think, and I hope the reader agrees, that I have finally figured out at least one way of assessing the effect of campaigns.
Having cut my teeth on economic voting and election forecasting models, I originally approached the topic as a bit of a nonbeliever. When I first began to mull over the issues addressed in this book, I was convinced that campaigns were virtually irrelevant to election outcomes. Given the apparent ability of simple forecast models to accurately predict election outcomes and the impressive performance of individual-level voting models, I did not see any room for campaigns to exert an independent influence on elections. By the time I finally started this project, however, I had come around to a point of view that could best be classified as agnostic: I had been impressed enough by anecdotal evidence of campaign effects that I was at least keeping an open mind to the possibility of campaign effects. By the end of this project, I became persuaded that “something is out there.” I think the evidence in this book clearly shows that although campaigns may not be the most important [Page xii]determinant of presidential election outcomes, they certainly play a key role in shaping public opinion and, ultimately, influencing outcomes.
A number of people have either directly or indirectly influenced this project and, hence, deserve to be thanked. First, I owe a debt of gratitude to those individuals who helped shape my approach to political science, my teachers at the University of Iowa. I especially want to thank Mike Lewis-Beck, Doug Madsen, Peter Snow, and Art Miller for what they taught me. I also need to thank those who commented on some of my earlier work on presidential campaigns: Jim Stimson, Ken Meier, Jim Garand, John Bibby, and Charley Tidmarch. I am also grateful for the comments provided by anonymous reviewers at various journals and publishers that either accepted or rejected my earlier efforts in this area. Almost without exception, I have found their comments useful.
My colleagues at UWM deserve thanks for their support of my research program over the years, as well as for their friendship. Thanks go especially to my colleague, Steve Percy, who always manages to convince me, by example, that I'm really not that busy after all. In addition, I would like to thank Linda Hawkins and the staff of the Social Science Research Facility for their assistance in getting the data onto my mainframe account so I could use it.
I also need to thank the Inter-University Consortium for Political and Social Research (ICPSR) and the Center for Political Studies (CPS) at the University of Michigan. Many of the data analyzed in this book were made available by the ICPSR and the CPS. The support provided by institutions such as the ICPSR and the CPS makes it possible for others in the scholarly community, such as myself, to pursue large-scale independent research projects.
The organizers (The National Election Studies and the Annenberg School of Communication) and participants of the Conference on the Impact of Presidential Campaigns are largely responsible for restarting my intellectual engine at a time when I needed a boost. This book was essentially idle for a three-month period before I attended the conference. I came home from the conference with a fresh perspective and a clearer idea of what issues a book on campaigns should address. Therefore, I thank all who participated in the conference.
[Page xiii]A number of other people have had a more direct effect on the work done in this book. First, Dick Niemi and Barbara Sinclair, coeditors of the Sage series, have provided invaluable input throughout this process. Their thorough evaluation of the manuscript and helpful suggestions have made this a much better book than it would otherwise have been. I also want to thank John Aldrich of Duke University, whose first-rate review of this book provided a number of useful insights and comments, many of which were incorporated into the manuscript. Also, David Farrell of the University of Manchester was generous enough to read the entire manuscript and share his insights with me. I can honestly say that all of the advice I received from these people was useful and led to a better manuscript.
I would like to thank the University of Wisconsin Press for granting permission to use a limited amount of material from “Campaigns, National Conditions, and U.S. Presidential Elections,” an article that I published in the American Journal of Political Science, volume 38, 1994.
The staff at Sage deserves thanks for making the entire production process as smooth and efficient as possible. Although there are no doubt countless individuals working behind the scenes, the people I have dealt with have been extremely helpful. They are: Carrie Mullen, Peter Labella, Renée Piernot, Druann Pagliossotti, Gillian Dickens, and Jennifer Morgan. Through their professionalism, these people have made this a fairly quick and fairly painless process.
Last, but certainly not least, I am grateful to my wife, Kathleen Dolan, who provided support on a number of different levels. Kathy saw and commented on more versions of this manuscript than anyone else. Her input included comments on everything from spelling and grammatical errors, to ideas about the structure of the book and suggestions about how to make the arguments clearer and more intelligible. Kathy was especially good at stopping me from adding more analysis when it really wasn't necessary. Because of this, the book is not only better, it's also done. In addition to academic input, Kathy also provided immeasurable support at home. Although I did not find writing this book to be a particularly arduous task, I did have to work a lot harder than I usually do, which did not make me a very happy person. Kathy put up with my “disposition” and did all that she could to make me comfortable. Those who know me well will understand what an onerous task this must have been. For this and everything else, I thank her.
Appendix A: Calculating Aggregate Candidate Support[Page 159]
The dependent variable for the aggregate candidate support analysis is based on trial-heat poll results for the race between the major-party presidential candidates. For the 1984 election, the data were taken from Goldman, Fuller, et al. (1985, 454). Goldman, Fuller, et al. present daily tracking poll results provided by Richard Wirthlin, pollster for the Reagan campaign. The poll results are based on four-day tracking polls from June 4 through October 4, and two-day tracking polls from October 5 through November 5. These data are unique compared to those used for the 1988 and 1992 analysis because they are provided by a single polling organization and cover virtually every day from early June through election day.
Gathering the data for 1988 and 1992 required a bit more creativity. These opinion estimates were generated on the basis of poll results provided in [Page 160]Public Opinion (November/December, 1988) and Public Perspective (July/August, September/October, and November/December, 1992). Daily candidate support figures were calculated in the following manner. First, poll results for support for the Republican and Democratic candidates were recorded for each day a poll was taken from early June through the last polls taken before the election. To minimize the problem of incompatible poll results, only those polls that sampled registered voters were used in the analysis. Second, the poll results were averaged by day, yielding a single figure for Republican and Democratic candidate support for each day that polling results were available. The difference between the percentage supporting the Republican candidate and the percentage supporting the Democratic candidate (Republican percentage – Democratic percentage) was then calculated and used as the dependent variable. The data in Table A.1 illustrate how the dependent variable was calculated for June 6 to June 8, based on the results of a Gallup poll taken from June 6 to June 8 and a Washington Post/ABC News poll taken from June 6 to June 7.Table A.1 Calculating Candidate Support (in percentages)
The days for which polling data are not available present a special problem. Rather than exclude these days from the analysis, poll values are estimated for these days by interpolating between days with existing data. For example, if the polling data give a Republican advantage of 6 percentage points on one day and 8 percentage points two days later, and there are no data available for the middle day, an interpolated value of 7 percentage points would be assigned to the middle day.
[Page 161]On any given day these trial-heat results represent the relative level of support for the major-party candidates. Although it is difficult to demonstrate that these results represent the actual vote intention of respondents, some indication of the accuracy of these data can be gleaned by comparing the estimated outcome based on the last-day polling results to the actual election outcome. As the data in Table A.2 indicate, the last-day polling results are an accurate reflection of the actual election outcome, especially for the 1984 and 1992 elections. One important implication of this close relationship between the poll results and the actual election outcomes is that any model that generates an accurate prediction of the last-day polling result is also generating a close prediction of the actual election outcome (see Chapter 6).[Page 162]Table A.2 Accuracy of Last-Day Polling Results
Appendix B: Debate Surveys[Page 163]
All surveys used in the individual-level analysis in Chapter 5 were conducted by CBS News and the New York Times and were obtained through the Inter-University Consortium for Political and Social Research (ICPSR). All surveys were conducted by telephone and all samples were drawn using random-digit dialing. Analyses of these surveys included a weighting variable that weighted on the basis of region, race, and gender. The surveys each included a predebate battery of questions on a wide variety of topics related to the election and a postdebate battery of questions focusing on the respondents' reactions to the debates.
The survey used for the first debate of 1984 is CBS News/New York Times National and Local Surveys, Part 11: First Presidential Debate (ICPSR #8399). The dates of the predebate interviews are September 30 through October 4 and the dates for the postdebate interviews are October 7 and October 9. The survey used for the second debate of 1984 is CBS News/New [Page 164]York Times National and Local Surveys, Part 12: Second Presidential Debate (ICPSR #8399). The predebate interviews were conducted between October 14 and October 17 and the postdebate interviews were conducted on October 21, October 24, and October 25. In both postdebate surveys, all respondents—debate viewers and nonviewers—were asked about their impressions of the debates.
The survey used for the first debate of 1988 is CBS News/New York Times First Presidential Debate Panel Survey, September 1988 (ICPSR #9143). The predebate interviews were conducted on September 21 and September 22 and the postdebate interviews were conducted on September 25. The survey used for the second debate in 1988 is CBS News/New York Times Second Presidential Debate Panel Survey, October 1988 (ICPSR #9147). Predebate interviews were conducted between October 8 and October 10 and postdebate interviews were conducted on October 13. Only debate viewers were asked their impression of the debate in the postdebate interview.
The survey used for the first debate of 1992 is CBS News/New York Times Monthly Poll #1, October 1992 (ICPSR #6091). The predebate interviews were conducted between October 2 and October 4 and the postdebate interviews were conducted on October 11. Only debate viewers were asked for their impressions of the debate in the postdebate survey.
Appendix C: Alternative Models[Page 165]
One would be remiss if certain design issues concerning the method of analysis used here were not brought up. The first issue concerns the degree to which this design is truly dynamic in nature—that is, the degree to which one can be confident that the results reflect the degree to which the independent variables influence changes in the dependent variable. Even though the data analyzed here represent differences in the dependent variable over time, the analysis is still open to being described as essentially static in nature because the values do not represent the actual change in candidate support from one day to the next. One of the more stringent tests for dynamic causality involves including a lagged value of the dependent variable as an independent variable. Doing this allows inferences to be made about the effect of the independent variables while controlling for the effect of previous levels of the dependent variable.
[Page 166]When the model in Table 6.4 is modified to include candidate support in t – 1 as an independent variable, the substantive interpretation of the results changes very little (Table C.1). As might be expected, the explanatory power of the model increases substantially and the size of all other coefficients is reduced in the presence of the lagged dependent variable. However, there were very few changes in terms of which variables are significantly related to the dependent variable. The one exception to this is the third debate in 1992, which became significant when the lagged dependent variable was added to the model. More important, the relative effect of each of the independent variables remains essentially intact.
Table C.1 A Dynamic Model of Candidate Support With a Lagged Dependent Variable, 1984–1992 (GLS Results) Variable b t-Score Constant −16.90 −3.05** Support t −1 .67 18.72** Conventions Democratic, 1984 −1.03 −2.10** Republican, 1984 1.64 3.12** Democratic, 1988 −2.02 −3.42** Republican, 1988 4.26 5.36** Democratic, 1992 −5.15 −3.12** Republican, 1992 2.10 3.76** Debates First, 1984 −.63 −.95 Second, 1984 1.84 2.45** First, 1988 .07 .11 Second, 1988 1.51 2.34** First, 1992 −1.04 −.92 Second, 1992 −.75 −.56 Third, 1992 2.22 2.01** Perot 1.24 1.83** Campaign events .25 2.08** Momentum .06 2.43** National variables External events .18 .69 Political and economic climate .24 3.16** Year effects 1984 −.70 −.44 1988 −6.87 −3.94**
NOTE: The data have been corrected for first-order autocorrelation using the Yule-Walker procedure. The estimated value of first-order autocorrelation (prior to correction) is .15.
R2 = .98; SE = 1.72; N = 445.
*p < .10
**p < .05.
One other possible problem with this analysis lies in the potential for endogeneity among the independent variables. Specifically, it is possible that levels of presidential popularity and consumer sentiment, which are used to measure national conditions, could be influenced by campaign events and campaign rhetoric (see Shaw 1995). If this is the case, then some of the effect being attributed to national conditions may actually reflect the effect of campaigns on perceptions of the president and the state of the economy. One simple method for obviating the potential effect of endogeneity is to measure national conditions just before the campaign actually begins and use this as a constant value for national conditions during the campaign period. Events occurring during the campaign cannot possibly influence perceptions of national conditions prior to the campaign. The only problem with using the May national conditions variable is that it creates a state of perfect collinearity between national conditions and the year dummy variables. This problem is easily resolved by dropping the 1984 dummy variable from the analysis. When the model in Table 6.4 is modified to include only the value of national conditions in May of the election year, very few changes occur in the values of the other coefficients (Table C.2). In fact, one of the few changes that does occur is an increase in the size of the coefficient for national conditions. If endogeneity is a problem, it probably serves to dampen the effect of national conditions.[Page 167][Page 168]
Table C.2 Model of Candidate Support With June National Conditions, 1984–1992 (GLS Results) Variable b t-Score Constant −61.86 −10.01** Conventions Democratic, 1984 −1.84 −1.66* Republican, 1984 5.81 5.52** Democratic, 1988 −2.28 −1.99** Republican, 1988 11.40 8.59** Democratic, 1992 −13.06 −9.19** Republican, 1992 6.07 5.66** Debates First, 1984 −1.25 −.95 Second, 1984 5.32 3.64** First, 1988 .01 .01 Second, 1988 3.26 2.45** First, 1992 −2.49 −1.47 Second, 1992 −1.76 −.98 Third, 1992 3.44 2.02** Perot 6.76 6.49** Campaign events .73 3.45** Momentum .16 4.86** National variables External events −.39 −.96 Political and economic climate .83 11.81** Year effects 1988 −17.67 −15.04**NOTE: The data have been corrected for first-order autocorrelation using the Yule-Walker procedure. The estimated value of first-order autocorrelation (prior to correction) is .65.R2 = .85; SE = 1.92; N = 446.
*p < .10
**p < .05.
References[Page 169]1988. An improved model for predicting presidential elections. PS21: 843–47..1990. Change and continuity in the 1988 elections. Washington, DC: CQ Press., , and .1982. Interpreting and using regression. Beverly Hills, CA: Sage..1980. Before the convention: Strategies and choices in presidential nominating campaigns. Chicago: University of Chicago Press.1993. Presidential selection. In Researching the presidency. edited by George C.Edwards III, John H.Kessel, and Bert A.Rockman. Pittsburgh, PA: University of Pittsburgh Press.1988. Measuring change in party identification in an election campaign. American Journal of Political Science32: 996–1017. http://dx.doi.org/10.2307/2111198, and .American Enterprise. 1992. Public opinion and demographic report. The public perspective: A Roper Center review of public opinion and polling. November/December: 100–01.American National Election Studies Cumulative Data File, 1952–1992 (ICPSR #8475). Ann Arbor: University of Michigan, Center for Political Studies.American National Election Studies Cumulative Data File, 1952–1992 (ICPSR #5475). Ann Arbor: University of Michigan, Center for Political Studies.1993. The media game. New York: Macmillan., , and .1992. Behind Bush's mixed abortion signals. New York Times, August 15, A1.1975. The effect of aggregate economic conditions on Congressional elections. American Political Science Review69: 1232–39. http://dx.doi.org/10.2307/1955283, and .1988. Presidential primaries and the dynamics of public choice. Princeton, NJ: Princeton University Press.1992. The impact of electioneering in the United States. In Electioneering: A comparative study of continuity and change. edited by DavidButler and AustinRanney. Oxford, UK: Clarendon.[Page 170]1993. Messages received: The political impact of media exposure. American Political Science Review87: 267–85. http://dx.doi.org/10.2307/29390401975. Voter response to short-run economic conditions: The asymmetric effect of prosperity and recession. American Political Science Review69: 1240–54. http://dx.doi.org/10.2307/1955284, and .1984. Jackson charges he was ignored for the No. 2 spot. New York Times, July 11, A1..1983. State-level economic conditions and the 1980 presidential election. Presidential Studies Quarterly13: 62–9., and .1960. The American voter. New York: John Wiley., , , and .1992. Forecasting the presidential vote in the States. American Journal of Political Science36: 386–407. http://dx.doi.org/10.2307/21114831992. The convention bump. American Politics Quarterly20: 287–307. http://dx.doi.org/10.1177/1532673X9202000302, , and .1990. Trial-heat forecasts of the presidential vote. American Politics Quarterly18: 251–69. http://dx.doi.org/10.1177/1532673X9001800301, and .CBS News/New York Times First Presidential Debate Panel Survey, September 1988 (ICPSR #9143). Ann Arbor: University of Michigan, Center for Political Studies.CBS News/New York Times Monthly Poll #1, October 1992(ICPSR #6091). Ann Arbor: University of Michigan, Center for Political Studies.CBS News/New York Times National and Local Surveys, Part 11: First Presidential Debate (ICPSR #8399). Ann Arbor: University of Michigan, Center for Political Studies.CBS News/New York Times National and Local Surveys, Part 12: Second Presidential Debate (ICPSR #8399). Ann Arbor: University of Michigan, Center for Political Studies.CBS News/New York Times Second Presidential Debate Panel Survey, October 1988 (ICPSR #9147). Ann Arbor: University of Michigan, Center for Political Studies.Congressional Quarterly. 1987. National party conventions. Washington, DC: Congressional Quarterly.1966. The concept of the normal vote. In Elections and the political order. edited by AngusCampbell, Warren E.Miller, and Donald E.Stokes. New York: John Wiley.1994. Economic indicators handbook. Detroit, MI: Gale Research..1993. Logit modeling: Practical applications. London: Sage..1994. A communication model of presidential campaigns: A 1992 overview. In The 1992 presidential campaign: A communication perspective. edited by Robert E.Denton, Jr.Westport, CT: Praeger., and .1957. An economic theory of democracy. New York: Harper & Row..1990. Presidential approval: A sourcebook. Baltimore, MD: Johns Hopkins University Press., and .Facts on file. 1984. New York: Facts on File.Facts on file. 1988. New York: Facts on File.Facts on file. 1992. New York: Facts on File.1978. The effect of economic events on votes for president. Review of Economics and Statistics60: 159–72. http://dx.doi.org/10.2307/19249691993. Reexamining the “minimal effects” model in recent presidential elections. Journal of Politics55: 1–21. http://dx.doi.org/10.2307/2132225.1981. Retrospective voting in American national elections. New Haven, CT: Yale University Press.. [Page 171]1988. The effects of presidential debates on the electorate's preferences for candidates. American Politics Quarterly16: 486–501. http://dx.doi.org/10.1177/0044780880160040051993. Why are American presidential election polls so variable when votes are so predictable?British Journal of Political Science23: 409–51. http://dx.doi.org/10.1017/S0007123400006682, and .1985. Wake us when it's over: Presidential politics in 1984. New York: Macmillan., and .1989. Whose broad stripes and bright stars? The trivial pursuit of the presidency 1988. New York: Warner., and .1994. Quest for the presidency: 1992. College Station: University of Texas Press., , , , and .1985. The quest for the presidency: The 1984 campaign. New York: Bantam., , et al.1989. The quest for the presidency: The 1988 campaign. New York: Touchstone., , et al.1975. Commentary on Arcelus and Meltzer: The effect of aggregate economic conditions on Congressional elections. American Political Science Review69: 1255–65. http://dx.doi.org/10.2307/1955285, and .1977. Statistical methods for social scientists. New York: Academic Press., and .1991. The democratic facade. Pacific Grove, CA: Brooks/Cole., and .1989. The campaign and the media. In The election of 1988: Reports and interpretations. edited by GeraldPomper. Chatham, NJ: Chatham House..1982. President Reagan's mandate from the 1980 elections: A shift to the right?American Politics Quarterly10: 387–420. http://dx.doi.org/10.1177/0044780820100040011991. Presidential elections in space and time. American Journal of Political Science35: 91–109. http://dx.doi.org/10.2307/21114391994. The behavioral consequences of vice-presidential debates: Does the undercard have any punch?American Politics Quarterly22: 469–82. http://dx.doi.org/10.1177/1532673X94022004041992. Packaging the presidency..2nd ed. Oxford, UK: Oxford University Press.1988. Presidential debates: The challenge of creating an informed electorate. New York: Oxford University Press., and .1966. The responsible electorate. Cambridge, MA: Belknap.1985. The economic basis of Reagan's appeal. In The new direction in American politics. edited by John E.Chubb and Paul E.Peterson. Washington, DC: Brookings Institution., and .1989. Economics and politics in the 1984 presidential election. American Political Science Review33: 491–515. http://dx.doi.org/10.2307/2111157, , and .1971. Short-term fluctuations in U.S. voting behavior, 1896–1964. American Political Science Review65: 131–43. http://dx.doi.org/10.2307/19550491981. Political debates. In The handbook of political communication. edited by D.Nimmo and K.Sanders. Beverly Hills, CA: Sage., and .1991. The “turning point”: Viewers' reactions to the second 1988 presidential debate. American Politics Quarterly19: 80–95. http://dx.doi.org/10.1177/1532673X91019001051992. One that made a difference: Cognitive consistency, political knowledge, and the 1980 presidential debate. Public Opinion Quarterly56: 168–84. http://dx.doi.org/10.1086/2693091991. The joint press conference. New York: Greenwood., and . [Page 172]1995. Information search during an election campaign: Introducing a processing-tracing methodology for political scientists. In Political judgement: Structure and process. edited by MiltonLodge and KathleenMcGraw. Ann Arbor: University of Michigan Press.1944. The people's choice. New York: Duell, Sloane, & Pearce., , and .1991. News verdicts, the debates, and presidential campaigns. New York: Praeger., , , , and .1986. Interrupted time series. In New tools for social scientists. edited by William D.Berry and Michael S.Lewis-Beck. Beverly Hills, CA: Sage..1988. Economics and the American voter: Past, present, future. Political Behavior10: 5–21. http://dx.doi.org/10.1007/BF00989377.1992. Forecasting elections. Washington, DC: CQ Press., and .1976. A dynamic comparative analysis of presidential and House elections. American Journal of Political Science20: 671–91. http://dx.doi.org/10.2307/21105661989. One hundred years of economic statistics. New York: Facts on File..1985. The Election: Candidates, Strategies, and Decisions. In The Elections of 1984. edited by MichaelNelson. Washington, DC: CQ Press., and .1989. An impression-driven model of candidate evaluation. American Political Science Review87: 399–419., , and .1995. The responsive voter: Campaign information and the dynamics of candidate evaluation. American Political Science Review89: 309–26. http://dx.doi.org/10.2307/2082427, , and .1993. Inside the mental voting booth: An impression-driven model of candidate evaluation. In Explorations in political psychology. edited by ShantoIyengar and William J.McGuire. Durham, NC: Duke University Press., and .1988. The impact of personal and national economic conditions on the presidential vote: A pooled cross-sectional analysis. American Journal of Political Science32: 137–54. http://dx.doi.org/10.2307/21113141979. A dynamic simultaneous equation model of electoral choice. American Political Science Review73: 617–32. http://dx.doi.org/10.2307/1954960, and .Moore, Jonathan, ed. 1986. Campaign for president: The managers look at ‘84. Dover, MA: Auburn House.1994. Expectations and preferences in British general elections. American Political Science Review88: 371–83. http://dx.doi.org/10.2307/2944710, , and .1988. Jackson, “too mature” for anger, icily offers no embrace for ticket. New York Times, July 13, A1..1991. Media messages in American presidential elections. New York: Greenwood..1979. Reciprocal effects of policy preferences, party loyalties, and the vote. American Political Science Review73: 1071–89. http://dx.doi.org/10.2307/1953990, and .1989. The press and its missed assignment. In The elections of 1988. edited by MichaelNelson. Washington, DC: CQ Press.Pomper, Gerald, ed. 1985. The election of 1984: Reports and Interpretations. Chatham, NJ: Chatham House.Pomper, Gerald, ed. 1989. The election of 1988: Reports and interpretations. Chatham, NJ: Chatham House.1991. The reasoning voter: Communication and persuasion in presidential campaigns. Chicago: University of Chicago Press.. [Page 173]1993. The election: A “new Democrat” and a new kind of presidential campaign. In The elections of 1992. edited by MichaelNelson. Washington, DC: CQ Press., and .1983. Forecasting presidential elections. New Haven, CT: Yale University Press..1985. Why Reagan won. Brookings Review3: 25–32..1989. Campaign for president: The managers look at ‘88. Dover, MA: Auburn House..1989. Candidates, parties, and campaigns., and .2d ed. Washington, DC: CQ Press.1995. Strong persuasion? The effect of campaigns in U.S. presidential elections. Ph.D. dissertation, University of California, Los Angeles..1991. The mass media and public opinion polls in the 1988 presidential election. American Politics Quarterly19: 59–79. http://dx.doi.org/10.1177/1532673X9101900104, and .1992. From differential abstention to conversion: A change in electoral change, 1864–1988. American Journal of Political Science36: 309–30. http://dx.doi.org/10.2307/2111479.1988. Do the polls affect elections? Some 1980 evidence. Political Behavior10: 136–50. http://dx.doi.org/10.1007/BF00991410.1992. Vital statistics on American politics., and .3d ed. Washington, DC: CQ Press.1994. Vital statistics on American politics., and .4th ed. Washington, DC: CQ Press.1985. Regression in space and time: A statistical essay. American Journal of Political Science29: 914–47. http://dx.doi.org/10.2307/2111187.1992. The impact of media polls on the public. In Media polls in American politics. edited by Thomas E.Mann and GaryOrren. Washington, DC: Brookings Institution..1978. Political control of the economy. Princeton, NJ: Princeton University Press.1990. The decline of American political parties, 1952–1988. Cambridge, MA: Harvard University Press.1991. The rise of candidate-centered politics. Cambridge, MA: Harvard University Press.1992. The road to the White House 1992: The politics of presidential elections. New York: St. Martin's.1990. Sources of short-term change in party identification. Paper presented at the annual meeting of the Midwest Political Science Association, Chicago, April., and .1984. G.O.P. worry: Wooing viewers. New York Times, July 22, A1.1993. Air wars: Television advertising in election campaigns, 1952–1992. Washington, DC: CQ Press.
About the Author[Page 178]
Thomas M. Holbrook is Associate Professor of Political Science at the University of Wisconsin—Milwaukee. He has published articles on elections and state politics in the American Political Science Review, American Journal of Political Science, Journal of Politics, and several other political science journals.