HOW VOTERS DECIDE

Information Processing during Election Campaigns

This book attempts to redirect the field of voting behavior research by proposing a paradigm-shifting framework for studying voter decision making. An innovative experimental methodology is presented for getting “inside the heads” of citizens as they confront the overwhelming rush of information from modern presidential election campaigns. Four broad theoretically defined types of decision strategies that voters employ to help decide which candidate to support are described and operationally defined. Individual and campaign-related factors that lead voters to adopt one or another of these strategies are examined. Most importantly, this research proposes a new normative focus for the scientific study of voting behavior: We should care about not just which candidate received the most votes, but also how many citizens voted correctly – that is, in accordance with their own fully informed preferences. Since its inception the field of voting behavior has focused on what leads some citizens to vote Democratic and others to vote Republican; it is now time to ask what leads some citizens to vote correctly and others to vote incorrectly.

Richard R. Lau is Professor of Political Science and Director of the Walt Whitman Center for the Study of Democracy in the Political Science Department at Rutgers University. His research has been supported by the National Institute of Health, the National Science Foundation, the Ford Foundation, and the Carnegie Corporation. He has published in all of the major journals in political science and social psychology and recently wrote (with Gerald Promper) Negative Campaigning: An Analysis of U.S. Senate Elections (2004).

David P. Redlawsk is Associate Professor of Political Science at the University of Iowa. Prior to completing his Ph.D. and arriving at the University of Iowa in 1999, Redlawsk spent nearly ten years in the technology industry, managing information systems for colleges and working as a management consultant. As a political scientist, he has published in the American Political Science Review, the American Journal of Political Science, the Journal of Politics, and Political Psychology, among others. He twice received the Roberta Sigel Best Paper Award from the International Society of Political Psychology. He coedited Hate Speech on Campus: Cases, Commentary, and Case Studies (1997) with Milton Heumann and Thomas Church, and he is currently completing an edited volume on emotion in politics to be published in 2006. His research has been supported by the National Science Foundation.

---

CAMBRIDGE STUDIES IN PUBLIC OPINION AND POLITICAL PSYCHOLOGY

Series Editors

Dennis Chong, Northwestern University
James H. Kuklinksi, University of Illinois, Urbana-Champaign

Cambridge Studies in Public Opinion and Political Psychology publishes innovative research from a variety of theoretical and methodological perspectives on the mass public foundations of politics and society. Research in the series focuses on the origins and influence of mass opinion, the dynamics of information and deliberation, and the emotional, normative, and instrumental bases of political choice. In addition to examining psychological processes, the series explores the organization of groups, the association between individual and collective preferences, and the impact of institutions on beliefs and behavior.

   Cambridge Studies in Public Opinion and Political Psychology is dedicated to furthering theoretical and empirical research on the relationship between the political system and the attitudes and actions of citizens.

Books in the series are listed on the page following the Index.

---

HOW VOTERS DECIDE

INFORMATION PROCESSING DURING ELECTION CAMPAIGNS

RICHARD R. LAU
Rutgers University

DAVID P. REDLAWSK
University of Iowa

---

CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press
40 West 20th Street, New York, NY 10011-4211, USA

www.cambridge.org
Information on this title: www.cambridge.org/9780521848596

© Richard R. Lau and David P. Redlawsk 2006

This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.

First published 2006

Printed in the United States of America

A catalog record for this publication is available from the British Library.

Library of Congress Cataloging in Publication Data

Lau, Richard R.
How voters decide : information processing during election campaigns /
Richard R. Lau, David P. Redlawsk.
p. cm. – (Cambridge studies in public opinion and political psychology)
Includes bibliographical references and index.
ISBN 0-521-84859-8 (hardback) – ISBN 0-521-61306-X (pbk.)
1. Voting research – United States. 2. Elections – United States.
I. Redlawsk, David P. II. Title. III. Series.
JK1967.L38    2006
324.973–dc22    2005021892

ISBN-13 978-0-521-84859-6 hardback
ISBN-10 0-521-84859-8 hardback

ISBN-13 978-0-521-61306-4 paperback
ISBN-10 0-521-61306-X paperback

Cambridge University Press has no responsibility for
the persistence or accuracy of URLs for external or
third-party Internet Web sites referred to in this publication
and does not guarantee that any content on such
Web sites is, or will remain, accurate or appropriate.

---

To Karen, who has meant so much to me that – even if you only consider what I have come to take for granted, or what she thinks I take for granted but do not and in fact appreciate daily, or just the relatively little that I manage to convey that she knows I appreciate – would easily constitute the most important decision of my life. I am so glad, in so many different ways, that I met you.

RRL

To Aletia, Andrew, and Greg, who put up with a husband and father who often couldn’t seem to make up his mind exactly what to do in life, and whose support, encouragement, and sometimes even frustration helped me find the direction I needed. And to my father, who has continued to get smarter as I have gotten older!

DPR

---

In his reflective moments even the most experienced politician senses a nagging curiosity about why people vote as they do. His power and his position depend upon the outcome of the mysterious rites we perform as opposing candidates harangue the multitudes who finally march to the polls to prolong the rule of their champion, to thrust him, ungratefully, back into the void of private life, or to raise to eminence a new tribune of the people....

Scholars, though they have less at stake than do politicians, also have an abiding curiosity about why voters act as they do.

V. O. Key (1966, p. 1)

---

Contents

List of Tables and Figures			page xi
Acknowledgments			xv
I. Theory and Methods
1		Introduction	3
2		A New Theory of Voter Decision Making	21
3		Studying Voting as a Process	47
4		What Is Correct Voting?	72
II. Information Processing
5		What Voters Do – A First Cut	93
6		Individual Differences in Information Processing	119
7		Campaign Effects on Information Processing	135
III. Politics
8		Evaluating Candidates	157
9		Voting	184
10		Voting Correctly	202
11		Political Heuristics	229
IV. Conclusion
12		A Look Back and a Look Forward	255
Appendix A   Detailed Examples of Decision Strategies in Action			265
Appendix B   How the Dynamic Information Board Works			279
Appendix C   Overview of Experimental Procedures			287
Appendix D   Detailed Decision Scripts			299
Appendix E   Calculating the On-line Evaluation Counter			307
References			313
Index			335

---

List of Tables and Figures

TABLES
3.1		Indicators of Political Sophistication	page 67
4.1		Willingness to Change Original Vote as a Function of the Quality of the Original Choice	81
4.2		Effect of New Information on Decision to Change Vote	83
4.3		Correct Voting in American Presidential Elections, 1972–2000	86
8.1		On-line versus Memory-Based Global Evaluations	169
8.2		Liking of Presidential Candidates and the Vote Choice	176
9.1		Vote for the Modal Candidate in the Primary Election	191
9.2		Vote for the Republican Candidate in the General Election	194
9.3		Defection from the Party’s Candidate in the General Election	197
10.1		Baseline Model of Correct Voting	208
10.2		Effect of Decision Strategies on Correct Voting, Primary Election Campaign	214
10.3		Effect of Decision Strategies and Memory on Correct Voting, General Election Campaign	216
11.1		Further Validity Evidence for Measures of Heuristic Use	240
11.2		Effect of Political Sophistication on Use of Political Heuristics	241
C.1		Characteristics of Experimental Subjects	289
E.1		Comparison of Evaluation Integration Rules, In-party Candidates Only	311
FIGURES
1.1		Four models of individual decision making	8
2.1		Process-oriented framework for studying voter decision making	22
2.2		Characteristics of different decision rules	36
2.3		Theoretical assumptions and predictions derived from process-oriented framework for studying the vote decision	45
3.1		Content of media coverage of the 1988 U.S. presidential election campaign	48
3.2		A dynamic information board	55
3.3		Brief description of mock presidential candidates	56
3.4		Images of mock presidential candidates	59
3.5		Defection from party in presidential voting by strength of party identification	69
3.6		Defection from party in presidential voting by vote in party primary	70
5.1		Number of items accessed, primary election	95
5.2		Number of unique items accessed per candidate, primary election	96
5.3		Unique items accessed per candidate, each third of primary campaign	98
5.4		Number of items accessed, general election	100
5.5		Unique items accessed per candidate running in each election campaign	101
5.6		Content of search, primary election campaign	103
5.7		Content of search, general election campaign	103
5.8		Memory for general election candidates	106
5.9		Distribution of different measures of information search, primary election campaign	110
5.10		Characteristics of different types of decision strategies	113
5.11		Prevalence of different decision strategies: a first cut	115
5.12		Prevalence of different decision strategies: revised measure	117
6.1		Distribution of subject political expertise	121
6.2		Effects of background characteristics on content of search	125
6.3		Effects of background characteristics on information processing	127
6.4		Effects of background characteristics on choice of decision strategy in the primary election	130
6.5		Effects of background characteristics on memory	132
7.1		Effect of number of candidates running in the primary on information search	137
7.2		Effect of number of candidates running in the primary on decision strategy	137
7.3		Effect of ideological distinctiveness on information search, general election campaign	139
7.4		Effect of ideological distinctiveness of general election candidates on decision strategy	139
7.5		Effect of stereotypic nature of out-party candidate manipulation on information search	141
7.6		Effect of stereotypic nature of out-party candidate manipulation on decision strategy	142
7.7		Effect of differential campaign resources on amount of information search directed toward primary candidates	146
7.8		The cognitive underbelly of the vote decision	150
8.1		Global candidate evaluations	167
8.2		Global evaluation by decision strategy	174
8.3		Relative probability of defection from on-line candidate evaluation compared to Model 2	180
9.1		Vote choice, primary election campaigns, four-candidate condition	186
9.2		Vote choice when candidate available, general election campaigns	187
9.3		Relative power of party identification and information processing measures on increased probability of a vote for the Republican candidate	195
9.4		Probability of defection from in-party candidate in general election by search strategy	198
9.5		Probability of defection from in-party candidate in general election by memory and search strategy	199
10.1		Estimated levels of correct voting in experiments and recent U.S. presidential elections	204
10.2		Effect of strength of partisanship, political sophistication, task demands, and perceived difficulty on correct voting	206
10.3		Change in probability of a correct vote due to individual and campaign factors, U.S. presidential elections, 1980–2000	210
10.4		Effect of decision strategies on change in probability of a correct vote, primary election	217
10.5		Effect of decision strategies and objective difficulty of decision on change in probability of a correct vote, primary election campaign	218
10.6		Effect of decision strategies on change in probability of a correct vote, general election campaign	219
10.7		Effect of accurate memory on the probability of a correct vote	222
11.1		Heuristic (and nonheuristic) search during election campaigns	235
11.2		Effect of static–dynamic manipulation on heuristic use	243
11.3		Effect of number of candidates running in the primary on heuristic use	244
11.4		Effect of ideological distinctiveness of general election candidates on heuristic use	245
11.5		Decision strategies and use of political heuristics	248
11.6		Effect of heuristic use by novices and experts on probability of a correct vote	251
12.1		Scorecard on theoretical assumptions and predictions	260
A.1		Information board for three-candidate election with all cells exposed, showing utilities and importance weights for a hypothetical voter	266
A.2		Examples of Model 1 rational choice compensatory decision rules	268
A.3		Example of Model 3 fast and frugal compensatory decision rule	271
A.4		Examples of Model 4 intuitive noncompensatory decision rules	273
A.5		Example of Model 2 confirmatory decision making	276
B.1		Dynamic information board information card	280
B.2		Dynamic information board video screen	282
B.3		Dynamic information board scenario page	283
C.1		Instructions to subjects	292
C.2		Outline of experimental procedure	296
D.1		Model 1d: deep intra-attribute search, ideal world	300
D.2		Model 2: moderately deep intra-candidate search, ideal world	302
D.3		Model 3: shallow intra-attribute search, ideal world	303
D.4		Model 4c: relatively shallow intra-candidate search, ideal world	304

---

Acknowledgments

Any project that has gone on for as long as this one will accrue debts (both intellectual and personal) too numerous to mention in the space normally allotted to such a task. With apologies to those we will inevitably overlook, in more or less chronological order, we thank Peter Bentler, Thad Brown, Barry Collins, Bob Jervis, Hal Kelley, John Petrocik, Bernie Weiner, and especially David Sears, as well as all graduate school teachers of the first author, without whose knowledge, training, and subsequent friendship all of this would have been impossible. We also thank Gerry Pomper, whose guidance helped the second author decide to pursue his Ph.D. (for better or for worse) after many years in the “real” world, and Milt Heumann, who, although not exactly in our field, gave great Hanukkah parties we always looked forward to, and whose advice and support were crucial to the second author during those initial disorienting days of graduate school and beyond. Thanks also go to Alan Kornberg, who, as the second author’s mentor during his undergraduate years at Duke, helped him develop an excitement and appreciation for the joys of academe and research.

   John Herstein graciously shared his stimulus materials with the first author at a very early stage in the development of this research; Herstein’s dissertation (or at least the 1981 published version of it) was the first time we had seen a decision board in action. Dana Dunn then used those stimulus materials as the basis for a senior honors project at Carnegie Mellon University, which became the infamous “Rick’s study” that none of Lau’s future students could ever figure out exactly what to do with. Ralph Erber got further than anyone else and helped point the way to a new way of thinking about this type of data.

   The National Science Foundation provided the crucial financial support necessary for the development of the dynamic process-tracing methodology described herein, with awards to both authors (SBR 93-21236 and SBR-9411162). Eric Johnson, one of the original developers of the decision board methodology, provided invaluable advice on using a decision board at an early stage in our development process. The Foundation’s reluctance to fund subsequent projects has kept us more or less focused on this one. Rutgers University and later the University of Iowa provided the laboratory facilities and countless other relatively minor resources necessary for any research project, which of course cumulatively far outweigh any formal grant award. The second author in particular thanks the Obermann Center for Advanced Study at the University of Iowa; its Director, Jay Semel; and administrative assistant, Lorna Olsen, for space and resources provided during a particularly intense period of manuscript revision.

   Racheal Ankrah, Jennifer Holt, Jill Locke, John Manyo, Grace Ann Mumoli, and Jeff Schnug worked as experimenters at various stages in this project. Licia DeVivo was an experimenter who also coded much of the open-ended data. Gail Shirazi, Elizabeth Williams, and Rachelle Brooks worked diligently coding the data that led to our initial conception of the on-line evaluation counter. Jason Humphrey, Andrew Civettini, and Kimberly Briskey all played key roles as research assistants as portions of the project moved to Iowa. At Rutgers, Paul Babbitt and Liz Felter served in the multiple roles of expert judges, project managers, experimenters, and readers of early papers from this research and thus deserve particular thanks.

   One of the most interesting tasks early on was the creation of campaign ads, using technology that seemed quite advanced at the time. We would like to thank all those who lent their voices to us to narrate the ads, but we especially thank George Bruce Morgan, the second author’s father-in-law and a retired radio announcer.

   Larry Bartels and the Center for the Study of Democratic Politics at Princeton University provided a home away from home for the first author and offered the most valuable resources of all, time to think and write, when the book manuscript was beginning to take shape. Numerous colleagues at Columbia, Duke, Iowa, North Carolina, Ohio State, Princeton, Rutgers, Stony Brook, UCLA, UCSD, and Vanderbilt and at the New York Area Political Psychology meeting, have listened to and provided valuable feedback on various aspects of this research. Bartels, Adam Berinsky, John Geer, Jane Junn, Tali Mendelberg, Steve Nicholson, Gerry Pomper, and David Sears have all read and provided feedback on early drafts of several different chapters of the book, as did graduate students in the second author’s Experimental Methods and Political Decision Making seminars. Several anonymous readers and Cambridge editors Dennis Chong and James Kuklinski have each read the whole damn thing twice and have provided trenchant criticisms, valuable suggestions, and unflagging encouragement throughout the publication process.

   And finally we want to thank Mo’s grandmother, who unbeknownst to her (and as far as we know, her grandson) symbolically at least inspired this entire research project.

---

HOW VOTERS DECIDE

Information Processing during Election Campaigns