a meta-analysis of the effects of viewing us presidential debates

17
This article was downloaded by: [University of Pennsylvania] On: 15 August 2012, At: 12:17 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Communication Monographs Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rcmm20 A meta-analysis of the effects of viewing U.S. presidential debates William L. Benoit, Glenn J. Hansen & Rebecca M. Verser Version of record first published: 24 Jun 2010 To cite this article: William L. Benoit, Glenn J. Hansen & Rebecca M. Verser (2003): A meta- analysis of the effects of viewing U.S. presidential debates, Communication Monographs, 70:4, 335-350 To link to this article: http://dx.doi.org/10.1080/0363775032000179133 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and- conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Upload: michel-andre-breau

Post on 02-Jan-2016

98 views

Category:

Documents


0 download

DESCRIPTION

A Meta-Analysis of the Effects of Viewing US Presidential DebatesWilliam L. Benoit, Glenn J. Hasen, & Rebecca M. VerserVersion of record first published: 24 June 2010Published in 2003 in Communication Monographs, Volume 70, Issue 4, Pages 335 - 350, by Routledge

TRANSCRIPT

Page 1: A Meta-Analysis of the Effects of Viewing US Presidential Debates

This article was downloaded by: [University of Pennsylvania]On: 15 August 2012, At: 12:17Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Communication MonographsPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/rcmm20

A meta-analysis of the effects ofviewing U.S. presidential debatesWilliam L. Benoit, Glenn J. Hansen & Rebecca M. Verser

Version of record first published: 24 Jun 2010

To cite this article: William L. Benoit, Glenn J. Hansen & Rebecca M. Verser (2003): A meta-analysis of the effects of viewing U.S. presidential debates, Communication Monographs, 70:4,335-350

To link to this article: http://dx.doi.org/10.1080/0363775032000179133

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make anyrepresentation that the contents will be complete or accurate or up to date. Theaccuracy of any instructions, formulae, and drug doses should be independentlyverified with primary sources. The publisher shall not be liable for any loss, actions,claims, proceedings, demand, or costs or damages whatsoever or howsoever causedarising directly or indirectly in connection with or arising out of the use of this material.

Page 2: A Meta-Analysis of the Effects of Viewing US Presidential Debates

AMeta-analysis of the Effects of Viewing U.S.Presidential Debates

William L. Benoit, Glenn J. Hansen, and Rebecca M. Verser

Televised debates are now an expected component of the American presidential election campaign. Ameta-analysis was used to cumulate the research on the effects of watching presidential debates. Generalcampaign debates increase issue knowledge and issue salience (the number of issues a voter uses to evaluatecandidates) and can change preference for candidates’ issue stands. Debates can have an agenda-settingeffect. Debates can alter perceptions of the candidates’ personality, but they do not exert a significant effecton perceptions of the candidates’ competence (leadership ability). Debates can affect vote preference.Primary debates increase issue knowledge, influence perceptions of candidates’ character, and can altervoter preferences (the effect sizes for these variables are larger in primary than general debates). The effectsizes for the dependent variables with significant effects were heterogeneous (except for effects of debatesother than the first on vote preference). No support was found for several possible moderator variables onissue knowledge, character perceptions, candidate competence, and vote preference: nature of subject pool(students, nonstudents), study design (pretest/posttest, viewers/nonviewers), number of days betweendebate and election, or data collection method (public opinion poll or experimenter data). The first debatein a series had a larger effect on vote preference than other debates, but was not a moderator for otherdependent variables. The possibility that other moderator variables are at work cannot be rejected.

Televised presidential debates have become an expected feature of the Americanpolitical landscape (Friedenberg, 1994). They have occurred in eight general

campaigns (1960, 1976–2000) and, although the records are not as clear, it appearsthat debates have been broadcast in the primary phase of the campaign in thirteenAmerican presidential elections, beginning in 1948 with a radio debate featuringThomas Dewey and Harold Stassen (Benoit et al., 2002a). Presidential debates haveoccurred in several other countries including Australia, Canada, Germany, Israel,Korea, and New Zealand (see, e.g., Coleman, 2000).Debates are a key element of a campaign. A citizen’s vote represents a choice

between (or among) competing candidates. Debates present the leading candidatesside by side, discussing more or less the same topics, which helps voters compare theoptions available to them and assists them in making their vote choice. Moreover,the face-to-face confrontation of debates allows candidates an opportunity torespond to opponents and such clashes can yield benefits for voters. Furthermore,the length of presidential debates (90 minutes for general election debates after thefour debates of 1960) provides voters with an extended opportunity to hear thecandidates discuss issues. Although there are more ads than debates, each debate isfar longer than any individual ad (most of which are 30 seconds long in recentcampaigns). Although debate viewership varies from campaign to campaign as wellas from debate to debate within a campaign, debates usually attract larger audiencesthan other campaign events (Carlin, 1994). The huge size of the presidential debate

William L. Benoit (Ph.D., Wayne State University, 1979) is Professor, and Glenn J. Hansen (M.A., BethelCollege, 1999) and Rebecca M. Verser (M.A., University of Arkansas, 2003) are Doctoral students at theUniversity of Missouri, Columbia.

Communication Monographs, Vol. 70, No. 4, December 2003, pp 335–350Copyright 2002, National Communication AssociationDOI: 10.1080/0363775032000179133

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 3: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS336

audience means that capacity for influence is considerable. Finally, the candidatesprepare extensively for the debates, but unexpected questions or statements fromopponents may provide voters with a more spontaneous (and possibly less contrived)impression of the candidates than they can obtain from other messages. The RacineGroup (2002) concluded, “while journalists and scholars display varying degrees ofcynicism about the debates, few deny that viewers find them useful and almost noone doubts that they play an important role in national campaigns” (p. 201).Some have voiced concerns about limitations on the power of debates. Jamieson

and Birdsell (1988, p. 161) asserted, “debates don’t very often convert partisans onone side to the other.” Many voters are not partisans, however, which means thatthey are susceptible to influence from debates. At the start of the 2000 presidentialcampaign, 38% of voters considered themselves independent, 34% identified withthe Democratic Party, and 28% with the GOP (Gallup, 1999). Furthermore, asubstantial proportion of partisans do vote for the nominee of the other party. Nie,Verba, and Petrocik (1999) reported that between 14% and 27% of party membersdefect to the opposing party. Although they do not explain the reasons for these votedefections, and they do not attribute them to debates, many partisans do vote for thecandidate of the other political party. This fact means that a sizeable number ofpartisans could be susceptible to influence from debates.Presidential debates serve other important functions besides conversion of parti-

sans, e.g., they can help undecided voters make a decision. Additionally, debates canincrease viewers’ confidence in their vote choice (Benoit, McKinney, & Holbert,2001), which could firm up the support of partisans and prevent some potentialvoters from defecting. This process also results in increased turnout of thatcandidate’s supporters on election day. Finally, debates may provide voters withmore or better information on which to base their decision without necessarilychanging their vote choice (Holbert, Benoit, & McKinney, 2002b).Quantitative studies of the effects of presidential debates on voters have occurred

in the United States since the first presidential debates between Richard Nixon andJohn Kennedy in 1960 and have been conducted in every campaign since thedebates resumed in 1976. The recent essay by the Racine Group (2002) lays outimportant issues and summarizes that work in a narrative review of the literature.The literature has now accumulated a substantial body of research across a numberof campaigns, which means there is a solid base of data for conducting ameta-analysis on the effects of debates on voters.Functional Theory (Benoit, Blaney, & Pier, 1998a) argues that citizens can base

their vote choices on policy and character, and they explain that two components ofcharacter are personality and competence (or leadership). Using these variables(policy or issues, character, competence) from Functional Theory and drawing onfindings in the existing literature (e.g., Benoit et al., 2001; Racine Group, 2002),seven hypotheses are proposed for this meta-analysis:1

H1. Presidential debate watching increases knowledge of candidates’ issue stands.H2. Presidential debate watching increases issue salience (number of issues used to evaluate

candidates).H3. Presidential debate watching affects preference for one candidate’s issue positions over

another’s.H4. Presidential debate watching has an agenda-setting effect (influences perceptions of the

relative importance of issues).H5. Presidential-debate watching affects perceptions of the candidates’ character (e.g., honesty,

compassion, morality).

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 4: A Meta-Analysis of the Effects of Viewing US Presidential Debates

PRESIDENTIAL DEBATES 337

H6. Presidential-debate watching affects perceptions of the candidates’ competence (leadershipability, experience).

H7. Presidential-debate watching affects vote choice.

Procedures will be described next, followed by results and discussion of theimplications of this meta-analysis.

Method

Procedure

Literature search and coding. Suggestions by Stock, Okum, Haring, Miller, Kinney,and Williams (1982) were employed to locate and code the articles for the study. Thefirst step was to locate studies to include in the meta-analysis. Recent review articleswere consulted first (e.g., Benoit et al., 2001; Racine Group, 2002; Yawn & Beatty,2000). Then various indices (CommSearch, ComIndex, J-Stor) and Louden’s usefulweb-page bibliography (2003) were examined to locate other studies. Each time apertinent study was located its references were examined for other studies. Thisapproach generated an initial pool of 87 publications that reported empirical studiesof presidential debates.The literature indicates that primary debates are likely to have greater influence

on viewers than general debates because voters have less information about thecandidates at that point in the campaign (Benoit, McKinney, & Stephenson, 2002a;Best & Hubbard, 2000). For example, in 2000 voters surely knew less about primarycontenders such as Orrin Hatch or Gary Bauer than about the two major partynominees who participated in the general debates (Governor George W. Bush andVice President Al Gore). In fact, voters probably knew less about Bush and Goreduring the primary season than they knew about these candidates in the Fall. Thus,primary debates are likely to have greater effects than general debates. Accordingly,the results for general and primary debates will be reported separately (studieslocated permitted tests of H1, H5, and H7 for primary debates).2

Second, a coding sheet was developed that listed the citation for the study, theinformation sought about the nature of the study (Which campaign and debate wasbeing studied? Did the study use a student sample? Was the design pretest/posttestor posttest only? Did the study use public opinion poll data?), the dependentvariable(s) tested in the hypotheses (e.g., issue learning, issue salience, preference forcandidate policy stands, agenda setting, candidate character, candidate competence,and vote preference), sample size, and results. The coding sheet and codingprocedures were refined by jointly coding nine studies. Finally, the studies weredivided among the authors and each individually coded one third of the studies.When questions arose about coding a particular article, all authors met to agree onthe coding for that study.Some studies could not be included in the meta-analysis. For example, Kraus and

Smith (1962) investigate differences between Republicans and Democrats. Althoughinteresting, they do not report either pretest/posttest data for debate watchers orcompare debate watchers with nonwatchers. Gallup (1987) reports the percentagessupporting each presidential candidate before and after the 1980 debate, but doesnot provide the sample size for the surveys. Studies that employed regression wereomitted unless they also provided zero-order correlations for the variables of interest.Regression was not used because there is no accurate method of deriving the effectsize from beta-weights (b or �; or from the alpha levels associated with the

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 5: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS338

beta-weights). This decision is consistent with most other meta-analytic literature(Allen & Burrell, 2002; Hunter & Schmidt, 1990). Also, Holbrook (1999) used adependent variable considered to be a measure of knowledge but which is moreaccurately conceptualized as a measure of issue salience (see Brians & Wattenberg,1996; Holbert, Benoit, Hansen, & Wen, 2002a). Accordingly, this study was notincluded in this analysis.The studies used a variety of dependent variables, which were clustered into the

variables embodied in the hypotheses. For example, a single character variable wasemployed. Nevertheless, some studies asked multiple character questions about suchpersonality traits as honesty, compassion, courage, morality, or sociability. When astudy reported data from multiple questions about a concept considered to be asingle dependent variable, like character, results for those related items wereaveraged. For example, if a study reported the effects of watching a debate onhonesty, compassion, courage, morality, and sociability, the data on these fivecharacter traits were averaged to create a single score for the character variable. Hadthe results for each question been included separately, that would have artificiallyinflated the N for the meta-analysis (that is, the n for that study would have beencounted five times, one for each of the character traits). When the questions askedabout the candidates’ competence, experience in government, or leadership ability,those items were considered to operationalize the competence variable. If a studyprovided results for four character variables—like competence, honesty, experience,courage—the data for honesty and courage were averaged (and used to test H5 oncharacter), whereas data for competence and experience were averaged (and used totest H6 on competence). Similarly, some studies reported the results for severalquestions about issue knowledge (e.g., education, taxation, economy). Again, thesequestions were averaged into a single estimate of issue knowledge from that sample.

Statistical analysis. This analysis employed the correlation coefficient r as opposed toother measures of effect size (e.g., Cohen’s d). Hunter and Schmidt (1990)recommended r as the most useful measure of effect size in meta-analysis (they alsoprefer r over r2; see also Beatty, 2002). Formulas for converting the statistics reportedin the coded articles to r were obtained from several sources. Hunter and Schmidt(p. 272) provided the formula for converting t to r. This formula was used to convertF to r after extracting the square root of F (the numerator df must equal 1). Wolf(1986, p. 35) provided similar formulas including the formula for converting �2 to r.Finally, Glass, McGaw, and Smith (1981, p. 139) supplied the Probit transformationtable used to convert differences in proportions to effect sizes.One correction was made to the effect sizes. Sampling error was corrected by

weighting the average overall effect size by the number of subjects in the study.Hunter and Schmidt (1990) noted that if the population correlation is assumed to beconsistent across all studies then “the best estimate of that correlation is not thesimple mean across studies but a weighted average in which each correlation isweighted by the number of persons in that study” (p. 100). Similarly, Wolf (1986)indicated, “it can be argued that not all studies synthesized in a meta-analyticliterature review should be given equal weight” (p. 39). All things being equal, studieswith larger sample sizes provide a better estimate of the population parameter beingmeasured. Hunter and Schmidt supplied the formula used to correct for samplingerror (p. 100).Effect size mean confidence intervals were used to determine whether the average

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 6: A Meta-Analysis of the Effects of Viewing US Presidential Debates

PRESIDENTIAL DEBATES 339

TABLE 1KNOWLEDGE OF CANDIDATE ISSUE POSITIONS IN GENERAL DEBATES

Author(s) Year r N Ss1 Design2 Debate # Days3

Bishop, Oldendick, 1976 0.440 898 N V all n/a& Tuchfarber, 1978Graber, 1978 1976 0.410 21 N V all n/aChaffee, 1978 1976 0.243 164 N P all n/aAbramowitz, 1978 1976 0.381 139 N P 1 41Benoit & Hansen, in press 1976 0.174 458 N V all n/aMulder, 1978 1976 0.235 91 N V 1 41Lang & Lang, 1978 1976 0.146 308 S P 1 41Benoit & Hansen, in press 1980 0.143 842 N V all n/aWall, Golden, & James, 1988 1984 0.089 203 S P 2 9Benoit & Hansen, in press 1984 0.206 588 N V all n/aLanoue, 1991 1988 0.522 142 S P 2 20Zhu, Milavasky, 1992 0.252 53 S V 1 24& Biswas, 1994Jamieson & Adasiewicz, 2000 1996 0.414 1013 N V 1 31Holbert et al., 2002a 1996 0.118 722 N V 1 31Benoit et al., 1998b 1996 0.160 352 S V 1 31Benoit & Hansen, in press 1996 0.204 393 N V all n/aHolbert et al., 2002a 1996 0.101 413 N V 2 21Benoit & Hansen, in press 2000 0.429 402 N V all n/aMean weighted effect size, 0.256 7202total N

Notes. 1Student or nonstudent. 2Pretest/posttest or viewers/nonviewers. 3Number of days between the debate andelection day. None of these studies utilized public opinion poll data.

r differs significantly from zero. If the confidence interval of an effect size includeszero then that effect was considered to be no different from zero.The homogeneity of the variance of each effect size was tested using the formula

in Hunter and Schmidt (1990, p. 110). A nonsignificant �2 indicates that the varianceof the effect size is within the limits of sampling error and that the studies form ofhomogeneous group (so differences in effect size are due to random error rather thana moderating variable). A significant result indicates that the differences in effectsizes are larger than what would be expected from sampling error alone (D’Alessio& Allen, 2000; Hunter & Schmidt, 1990). The average effect size when the varianceis heterogeneous should be viewed cautiously (Allen & Burrell, 2002). When this testindicated heterogeneous variance, potential moderator variables were examined bycalculating correlations between the potential moderator variable and the dependentvariable (Hunter & Schmidt, p. 486).

Results

General Debates

Knowledge of candidate issue positions. Candidate issue knowledge measureswhether debate viewing increases viewers’ knowledge of the candidates’ issuepositions. Table 1 indicates that 18 measurements of this variable were obtainedfrom 13 studies. The total N for this group of studies was 7202.3 The mean weightedeffect size was .256 (SD� .136). The 95% confidence interval for the effect size was.195 – .323. Thus, watching debates has a positive effect on issue knowledge (H1).The test of homogeneity of effect size was statistically significant, �2(15)� 152.62,p� .001, which could indicate the presence of one or more moderator variables.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 7: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS340

TABLE 2ISSUE SALIENCE IN GENERAL DEBATES

Author(s) Year r N Debate # Days1

Benoit & Hansen, in press 1976 0.201 555 all n/aBenoit & Hansen, in press 1984 0.223 623 all n/aBenoit & Hansen, in press 1996 0.183 410 all n/aHolbert et al., 2002a 1996 0.181 413 2 21Holbert et al., 2002a 1996 0.076 722 1 31Benoit & Hansen, in press 2000 0.309 430 all n/aMean weighted effect size, total N 0.187 3153

Notes. 1Number of days between the debate and election day. None of these studies utilized public opinion poll data.All of these studies compared viewers with nonviewers and all studies used nonstudents.

Because the variance in this effect size was heterogeneous, potential moderatorvariables were examined. The effect size was not related to student (coded low)versus nonstudent samples, r� .119, p� .64. Nor was effect size related to use ofviewers/nonviewers (coded low) or pre/posttest design, r� .079, p� .76. There wasno larger effect size for the first debate (coded low) compared with subsequentdebates, r� .246, p� .33.4 Because debates held further away from (or, alternatively,closer to) election day could have had greater effects, a correlation was calculatedbetween the days between the debate and the election and the effect size. Again, thispotential moderator variable did not correlate substantially with the effect size,r� .138, p� .70. Thus, none of the moderator variables tested were significantinfluences on effect size.5

Issue salience. Hypothesis two concerned issue salience. The sample of studies offerssix data points for the effects of watching debates on issue salience from two studieswith a total unique N of 3153 (see Table 2). The mean weighted effect size was .187(SD� .07). The 95% confidence interval was .134 – .258, indicating that watchingdebates has an effect on issue salience. The test for homogeneity of effect size wasstatistically significant, �2(5)� 16.59, p� .01. The characteristics of these six studiesdid not allow additional analyses for potential moderator variables (e.g., none ofthese studies used students as subjects).

Issue preference. Four studies examined the effects of debates on preference forcandidates’ policy positions (see Table 3), the third hypothesis. The total N was 625.The average weighted effect size for this variable was .136 with SD� .10. The 95%confidence interval was .027 – .319, suggesting that this effect is different from zero.The test of homogeneity was not statistically significant, �2(3)� 6.49, p� .09. Nofurther moderator tests were possible with these studies.

Agenda-setting effect. Three studies (total N of 216) examined the agenda-settingeffect of presidential debates (see Table 4). The mean weighted agenda-setting effectsize was .291 (SD� .168). The 95% confidence interval did not include zero (.215 –.441), indicating that watching debates has an agenda-setting effect (H4). Noadditional tests for moderator variables were conducted on these three studies.

Perceptions of candidates’ character. The sample included 10 studies with 17 effect sizesand a combined N of 5426 studied perceptions of candidate character (see Table 5),the fifth hypothesis. The mean weighted effect size was .266 (SD� .156) and the95% confidence interval ranged from .182 to .350, which suggests this effect is

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 8: A Meta-Analysis of the Effects of Viewing US Presidential Debates

PRESIDENTIAL DEBATES 341

TABLE 3ISSUE PREFERENCE IN GENERAL DEBATES

Author(s) Year r N Ss1 Design2 Debate # Days3

Kraus & Smith, 1962 1960 0.206 131 N P 1 37Abramowitz, 1978 1976 0.006 139 N P 1 41Becker, Sobowale, Cobbey, 1976 0.124 297 N V all n/a& Eyal, 1978Benoit et al., 2001 2000 0.355 58 S P 1 29Mean weighted effect size, total N 0.136 625

Notes. 1Student or nonstudent. 2Pretest/posttest or viewers/nonviewers. 3Number of days between the debate andelection day. None of these studies utilized public opinion poll data.

TABLE 4AGENDA SETTING IN GENERAL DEBATES

Author(s) Year r N Ss1 Debate # Days2

McLeod Durall, Ziemke, 1976 0.110 97 N all n/a& Bybee, 1979Swanson & Swanson, 1978 1976 0.490 61 S 1 41Benoit et al., 2001 2000 0.385 58 S 1 29Mean weighted effect size, total N 0.291 216

Notes. 1Student or nonstudent. 2Number of days between the debate and election day. None of these studies utilizedpublic opinion poll data. All of these studies used a pretest/posttest design.

TABLE 5PERCEPTIONS OF CANDIDATES’ CHARACTER IN GENERAL DEBATES

Author(s) Year r N Ss1 Design2 Debate # Days3

Lang & Lang, 1962 1960 0.411 91 N P 1 12Atkin, Hocking, 1976 0.496 466 N P 3 12& McDermott, 1979Carter, 1962 1976 0.481 107 N P all n/aBenoit & Hansen, in press 1976 0.079 512 N V all n/aBecker et al., 1978 1976 0.407 297 N V 1 41Becker et al., 1978 1976 0.182 256 N V all n/aBecker et al., 1978 1976 0.220 233 N V all n/aMulder, 1978 1976 0.070 310 S V 1 41Graber, 1978 1976 0.450 21 N V all n/aBenoit & Hansen, in press 1984 0.162 606 N V all n/aJamieson & Adasiewicz, 2000 1996 0.434 1013 N V 1 31Yawn & Beatty, 2000 1996 0.002 59 N P 2 21Yawn & Beatty, 2000 1996 0.002 215 S P 2 21Benoit et al., 1998b 1996 0.159 355 S V 1 29Benoit & Hansen, in press 1996 0.295 402 N V all n/aBenoit & Hansen, in press 2000 0.225 425 N V all n/aBenoit et al., 2001 2000 0.441 58 S P 1 29Mean weighted effect size, total N 0.266 5426

Notes. 1Student or nonstudent. 2Pretest/posttest or viewers/nonviewers. 3Number of days between the debate andelection day. None of these studies utilized public opinion poll data.

statistically significant. The test of homogeneity of the effect sizes on character wasstatistically significant, �2(16)� 150.8, p� .001.The nature of these studies permitted several tests for possible moderator

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 9: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS342

TABLE 6PERCEPTIONS OF CANDIDATE COMPETENCE IN GENERAL DEBATES

Author(s) Year r N Ss1 Design2 Debate # Days3

Ben-Zeev & White, 1962 1960 0.016 164 N P 3 21Ben-Zeev & White, 1962 1960 0.023 169 N P 4 12Lang & Lang, 1962 1960 0.411 91 N P all n/aBen-Zeev & White, 1962 1960 0.108 175 N P 2 27Mulder, 1978 1976 0.028 310 S V 1 41Benoit et al., 1998b 1996 0.139 355 S V 1 31Mean weighted effect size, total N 0.096 1264

Notes. 1Student or nonstudent. 2Pretest/posttest or viewers/nonviewers. 3Number of days between the debate andelection day. None of these studies utilized public opinion poll data.

variables. There was no relationship between students (coded low) and nonstudents,r� .323, p� .21. There was no moderating effect for those studies that usedviewers/nonviewers (coded low) versus a pre/posttest design, r� .175, p� .50. Norelationship was detected between the first debate (coded low) and all other debates,r� –.041, p� .876. The correlation between days between the debate and electionday was not significant, r� –.134, p� 730. No other potential moderator variablescould be examined.

Perception of candidates’ competence. The sample included four studies (six effect sizes)investigating perceptions of candidates competence (see Table 6). The N for thisdependent variable was 1264 and the mean weighted effect size was .096(SD� .102). The 95% confidence interval included zero (–.003 – .245), so this effectis not significant (H6). The test for homogeneity of effect size was significant,�2(5)� 29.54, p� .001. However, no moderator variable could be identified in thesedata: subjects (students coded low), r� .192, p� .716; design (viewer/nonviewerscoded low), r� .192, p� .716; debate order (first coded low), r� .523, p� .287; anddays before election, r� .283, p� .645.

Candidate vote preference. Hypothesis 7 predicted that debate watching influencesvote preference, the final dependent variable examined. This variable had 25 datapoints from 14 studies of general election debates and included 8876 subjects (seeTable 7). The weighted average mean was .149 (SD� .123). The 95% confidenceinterval ranged from .099 to .211, indicating that the effect was significant. The testof homogeneity was statistically significant, �2(24)� 176.39, p� .001.These studies did allow tests of several possible moderators. There was no

relationship between student (coded low) and nonstudent samples, r� –.290,p� .159. There was no significant relationship for design, viewers/nonviewers(coded low) versus pretest/posttest, r� .171, p� .413. No relationship existedbetween effect size and whether the study used public opinion poll data or localsamples (coded low), r� .093, p� .657. There was no significant relationshipbetween the days between the debates and election day and effect size, r� .112,p� .112. There was, however, a moderator effect between first (coded low) and allother debates, r� –.554, p� .021. The first debate had a larger effect on votepreference compared to subsequent debates.In order to make certain that no moderators were present in these subsamples,

first and other debates were considered separately. The weighted mean effect size forwatching the first debate in a campaign on vote preference was .252 (SD� .112).

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 10: A Meta-Analysis of the Effects of Viewing US Presidential Debates

PRESIDENTIAL DEBATES 343

TABLE 7CANDIDATE VOTE PREFERENCE IN GENERAL DEBATES

Author(s) Year r N Ss1 Design2 Debate # Days3 Poll4

Lang & Lang, 1962 1960 0.349 95 N P all n/a NBen-Zeev & White, 1962 1960 0.000 149 N P 3 21 NDesutschmann, 1962 1960 0.381 159 N P all n/a NBen-Zeev & White, 1962 1960 0.040 165 N P 4 12 NBen-Zeev & White, 1962 1960 0.020 155 N P 2 27 NBenoit & Hansen, in press 1976 0.061 345 N V all n/a NBecker et al., 1978 1976 0.136 277 N V 3 12 NBecker et al., 1978 1976 0.184 168 N V 2 n/a NGeer, 1988 1976 0.215 1137 N P 1 41 PBecker et al., 1978 1976 0.018 190 N V 1 41 NRose, 1976 1976 0.048 1416 N V all n/a NBecker et al., 1978 1976 0.107 162 N V 3 12 NBecker et al., 1978 1976 0.411 387 N V 1 41 NMcLeod et al., 1979 1976 0.120 97 N P all n/a NLang & Lang, 1978 1976 0.303 694 S P 1 41 NJacoby, Troutman, 1980 0.367 30 S V 1 8 N& Whittler, 1986Geer, 1988 1984 0.177 514 N P 2 9 PGeer, 1988 1984 0.177 510 N P 1 23 PWall et al., 1988 1984 0.418 302 S P 1 23 NBenoit & Hansen, in press 1984 0.013 544 N V all n/a NPayne, Golden, 1988 0.065 498 S P 1 38 NMarlier, & Ratzan, 1989Payne et al., 1989 1988 0.129 182 S P 2 20 NLanoue, 1991 1988 0.070 98 S P 2 20 NBenoit & Hansen, in press 1996 0.024 285 N V all n/a NBenoit & Hansen, in press 2000 0.054 317 N V all n/a NMean weighted effect size, 0.149 8876total N

Notes. 1Student or nonstudent. 2Pretest/posttest or viewers/nonviewers. 3Number of days between the debate andelection day. 4Public opinion poll data/nonpublic public opinion poll data.

The 95% confidence interval for this effect size is .137 – .355, which means it shouldbe considered statistically significant. These effect sizes were not homogeneous,�2(7)� 102.56, p� .001, which could signal the presence of one or more additionalmoderator variables. None of the potential moderators tested were significant,however: subject (students coded low), r� –.289, p� .488; design (viewers/nonview-ers coded low), r� –.100, p� .813; days before election, r� –.390, p� .340; and poll(nonpoll data coded low), r� –.204, p� .628.The weighted mean effect size for other debates (excluding studies that examined

all debates as a group) was .116 (SD� .06). The 95% confidence interval for thiseffect size was .052 – .140, which indicates that it is unlikely to be different fromzero. These effects were homogeneous, �2(8)� 7.69, p� .45, which means that it isunlikely that a moderator variable affects these results.

Primary Debates

Knowledge of candidate issue positions. Two studies examined candidate issue knowl-edge gained from viewing primary debates (see Table 8). These studies had five datapoints and 187 subjects. The weighted mean was .833 (SD� .101). There is a 95%probability that the mean ranges between .697 and .897, indicating that this is asignificant effect (H1). The test of homogeneity was statistically significant,

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 11: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS344

TABLE 8KNOWLEDGE OF CANDIDATE ISSUE POSITIONS IN PRIMARY DEBATES

Author(s) Year r N Design1

Pfau, 1988 1984 0.688 24 PPfau, 1988 1984 0.673 26 PPfau, 1988 1984 0.792 25 PBenoit et al., 2002a 2000 0.902 78 VBenoit et al., 2002a 2000 0.931 34 VMean weighted effect size, total N 0.833 187

Notes. 1Pretest/posttest or viewers/nonviewers. All studies used student subjects; none used public opinion poll data.

�2(4)� 19.67, p� .001. Given the lack of variability across the moderator variablesno tests for potential moderator variables could be conducted.

Perceptions of candidates’ character. Two studies with three unique findings examinedperceptions of the candidates’ character in primary debates (see Table 9). A total of169 subjects were studied with a mean weighted effect size of .799 (SD� .082). The95% confidence interval for the mean effect size was between .707 and .959. Thissupports hypothesis five in primary debates. The test of homogeneity of variance wasstatistically significant, �2(2)� 8.69, p� .001, but given the small number of studiesno tests for moderator variables were conducted.

Candidate vote preference. Vote preference in primary debates was studied by fourdifferent researchers (five unique data points) using 386 subjects. The mean weightedeffect size was .541 (SD� .133) with a 95% confidence interval of .432 – .756. Thiseffect indicates that watching primary debates influences vote preference (H7).Finally, the test of homogeneity of variance was statistically significant, �2(4)� 25.46,p� .001, but no moderator tests could be conducted for this variable.

TABLE 9PERCEPTIONS OF CANDIDATES’ CHARACTER IN PRIMARY DEBATES

Author(s) Year r N

Benoit et al., 2002a 2000 0.946 34Benoit et al., 2002a 2000 0.730 78Benoit & Stephenson, in press 2000 0.769 57Mean weighted effect size, total N 0.799 169

Notes. All studies used student subjects and a pretest/posttest design; none used public opinion poll data.

TABLE 10CANDIDATE VOTE PREFERENCE IN PRIMARY DEBATES

Author(s) Year r N

Best & Hubbard, 1999 1996 0.481 126Benoit et al., 2002a 2000 0.840 34Benoit et al., 2002a 2000 0.434 78Benoit & Stephenson, in press 2000 0.735 57McKinney, Kaid, & Robertson, 2001 2000 0.481 91Mean weighted effect size, total N 0.541 386

Notes. All of these studies used student subjects and a pretest/posttest design; none used public opinion poll data.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 12: A Meta-Analysis of the Effects of Viewing US Presidential Debates

PRESIDENTIAL DEBATES 345

Discussion

The first four hypotheses pertain to issues or policy. The sample of issue learningstudies provides 18 effect sizes encompassing more than 7000 subjects. The datasupport the prediction that debates engender learning about the issue positions ofcandidates with a significant mean effect size of .275. Presidential debates offer anextended opportunity (90 minutes in each debate after 1960) for candidates topresent their views. Benoit et al. (2002b) summarized content analyses of everygeneral presidential debate, reporting that 75% of utterances in general debatesdiscuss policy. This outcome means that viewers have an extended opportunity tohear the candidates discuss the pros and cons of their policy stands. It is quitereasonable, therefore, to find that citizens’ issue knowledge increases from watchingdebates.

Six studies with more than 3000 subjects produced a significant mean effect sizeof .196 for issue salience. Those who watch debates tend to use more issues toevaluate presidential candidates. The fact that presidential candidates emphasizepolicy in these debates can explain why a greater number of issues are used toevaluate these candidates by those who watch debates. Of course, it is possible thatthose who have higher levels of issue salience are more likely to watch debates.

Four studies including more than 600 participants supported the prediction thatwatching a presidential debate can influence issue preference, or which candidate’sissue position appears superior to voters (or closer to voters’ own issue positions). Themean effect size for this variable was .138, which is significant. Debates not onlyengender knowledge about the candidates’ policy proposals, but they also influencethe policy preferences of viewers.

Three studies involving more than 200 subjects yielded a significant mean effectsize of .291 for agenda setting in general debates. The issues discussed in debates areperceived as more important by viewers.

Two hypotheses addressed character. Seventeen studies of almost 5500 subjectsproduced a mean effect size for candidate personality perceptions of .270, which wassignificant. Although content analysis reveals that candidates emphasize policy(Benoit et al. 2002b), presidential candidates did discuss character in 25% of generaldebate utterances. Thus, there is a clear opportunity for debate viewing to influenceperceptions of the candidates’ personality (of course, character perceptions could beinfluenced by nonverbal factors as well as verbal behavior).

Six studies with more than 1200 subjects investigated the effects of debates onperceptions of candidate competence (leadership ability). The mean effect size of.096 was not statistically significant. Content analysis of general debates (Benoit etal., 2002b) reveals that discussion of leadership ability accounts for only 6% of theremarks in these encounters. Therefore, it is not surprising that no effect was foundfor perceptions of candidate competence.

Finally, what is surely the bottom-line in studies of effects of watching presidentialdebates concerns vote preference. Twenty-five studies with more than 8500 subjectsyielded a mean effect size for debate watching on vote preference of .149, which wassignificant.

It is reasonable to place more confidence in some of these findings than othersbecause of the number of studies and subjects involved. Three of these conclusionsare based on stronger support: vote preference (25 studies, 8876 Ss), issue knowledge(18 studies, 7202 Ss), and candidate character (17 studies, 5426 Ss). The remaining

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 13: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS346

conclusions should be considered more tentative: issue salience (6 studies, 3153 Ss),candidate competence (6 studies, 1264 Ss), issue preference (4 studies, 625 Ss), andagenda setting (3 studies, 216 Ss).Primary debates have significant effects on issue knowledge, candidate character,

and vote preference (the effect sizes for primary debates were larger than for generaldebates, as was expected from the literature review, as indicated in footnote 2).Because much less research has been conducted on debates in this phase of thecampaign, however, these conclusions must also be considered somewhat tentative:vote preference (5 studies, 386 Ss), issue knowledge (5 studies, 187 Ss), and candidatecharacter (3 studies, 169 Ss).It is important to note that the effect sizes for these dependent variables were not

homogeneous. This result could indicate the presence of moderator variables.Several potential moderator variables were examined for one or more of thedependent variables investigated: nature of subject pool (students, nonstudents),6

study design (pretest/posttest, viewers/nonviewers), order of debate (first, other),number of days between debate and election, or data collection method (publicopinion poll or experimenter data). The only moderator detected was that the firstgeneral debate has a larger effect on vote preference than subsequent debates. Ofcourse, the fact that effect sizes were heterogeneous could mean that othermoderator variables are operating, although they could not be tested. It could alsomean that the quality of these studies (e.g., wording of questionnaire items) varies,increasing variance in effect sizes. Thus, the mean effect sizes reported should beinterpreted cautiously.This analysis suggests that research on debates has become more sophisticated

over time, which could mean that the quality of research has increased.7 First, earlierstudies of debates were less likely to conduct statistical analysis (to show that thecontrasts observed are probably significant differences rather than chancefluctuations). For example, some early studies report percentages or frequencieswithout attempting any statistical analyses. Second, when statistical analyses areconducted, the complete results are more likely to be reported in more recentpublications. Some early studies indicate which contrasts are significant (and at whatalpha level) but do not report the actual t or F values. Of the 87 potential studieslocated initially, only 33 reported data that could be included in the meta-analysis.The fact that only a few studies provided data on the reliability of their measures islamentable. It also means that the estimates reported here for effect sizes may beconservative. Third, political scientists appear to have a proclivity for regressionanalysis. Although regression is a perfectly acceptable form of statistical analysis,beta-weights cannot be cumulated across studies. So, using regression analysiswithout providing zero order correlations impedes the cumulation of data throughmeta-analysis.The advice offered on research into presidential debates by Zhu et al. (1994)

merits consideration. After reviewing the literature they offered recommendationsfor future research on debate watching:

The basic design should be a between-subjects design contrasting viewers and non-viewers, thedependent variables should be measured both pre- and post-debate, the results should be verifiedwith content analysis of the debate to ascertain content-specific effects, and the data should becollected from both the forced viewing condition (experiment, to assure that exposure does takeplace) and the natural viewing condition (survey, to offset laboratory artifacts). (pp. 311–312)

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 14: A Meta-Analysis of the Effects of Viewing US Presidential Debates

PRESIDENTIAL DEBATES 347

Research into the effects of debates must consider the content of debates (e.g., itwould be a mistake to attempt to measure issue learning from a debate withquestionnaire items concerning the candidates’ stands on abortion if abortion wasnot discussed in that debate). The search for potential moderator variables indicatesthat some of this advice is questionable. Design choice (pretest/posttest, viewer/non-viewer) is not a moderating variable for effect size for issue knowledge or candidatepersonality, the two dependent variables these authors discuss.

Finally, this meta-analysis indicates where more research is needed. Research hasclearly established that debates are capable of increasing issue knowledge. Althoughthere are undoubtedly more specific questions concerning this variable that could beinvestigated (e.g., do Republicans, Democrats, and other voters learn equally fromdebates), the basic question of whether debates enhance knowledge can be answeredaffirmatively. Similar points could be made about two other variables, votepreference and character.

On the other hand, other variables, like issue preference or agenda setting (fourand three studies in this sample, respectively), deserve to be prioritized in futureresearch. Closer examination of the Tables reported here reveals that, of the sixstudies using competence (leadership ability) as a dependent variable, four of thoseinvestigated the 1960 debates and one the 1976 debates. Future research mightprofitably study this variable (perhaps more sophisticated research would find aneffect). Finally, relatively few studies have been conducted into the primary debates.Thus, this study might be used both to assess the present status of the study ofpresidential debates, as well as directions for future research.

Conclusion

This study employed meta-analysis to cumulate the quantitative research on theeffects of watching televised presidential debates. Results indicate that generaldebates have several significant effects (issue knowledge, issue salience, issuepreference, agenda setting, candidate character, and vote preference), although theydo not appear to influence perceptions of candidate competence. Primary debateshad significant effects on the three variables with multiple studies: issue knowledge,candidate character, and vote preference (and these effects were larger than ingeneral debates).

Debates potentially can have both direct influence (on voters who choose to watchdebates) and indirect influence (on voters who learn about debates from the news).A meta-analysis (D’Alessio & Allen, 2002) confirmed Festinger’s (1957) predictionthat people are more likely to expose themselves to consonant than dissonantinformation. Thus, prior attitudes and expectations about debates could influencewho watches debates and are thereby susceptible to these effects. Given the lack ofmedia bias (D’Alessio & Allen, 2000), those influenced by a more indirect path (fromnews) are likely to receive a balanced report.

The variance in effect size for all statistically significant results was heterogeneous.This outcome could mean that other moderator variables are at work or that thereare problems with the quality of some studies. For one or more dependent variables,analyses found no support for several potential moderator variables: nature ofsample (students versus nonstudents), design (pretest/posttest versus a posttest-onlydesign of viewers/nonviewers), and whether the data were from public opinion polls

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 15: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS348

or not. The data do reveal that the first debate tends to have a larger effect on votepreference than subsequent debates.Of course, debate watching is self-selected. It is possible that those who choose to

watch debates also are more likely to expose themselves to other sources of campaigninformation, which means we cannot be certain that debates are responsible forthese results. However, the three times that it was possible to test design as apotential moderator variable revealed that it was not a moderator. The fact thatthere was no difference in effect size between the pre/posttest design (which ofcourse controls for knowledge levels before debates) and the viewer/nonviewerdesign, it seems unlikely that self-selection explains the results.

Footnotes1Readers might notice that one potential hypothesis is missing. Although some research investigates the

question of who won a debate, it was decided to exclude that question from this analysis for two reasons. First,we are not certain what this variable means. Elections are ultimately about vote choice (H7). Approaches topolitical communication (e.g., Benoit, 1999; Kaid & Johnston, 2000) typically divide message content into thetopics of issue or policy and character or image (H1–H5). Winning a debate, however, does not seem topertain directly to any of these concepts; instead, it seems to measure ability to perform in an artificialcampaign communication event (and one that seems to us to have little to do with being president). Second,given that Ross Perot was considered by viewers to be the winner of the first and third debates of 1992 withoutwinning a single electoral vote (Benoit & Wells, 1996), there are serious questions about the utility of thisvariable.

2T-tests were conducted to verify that the effect size for primary debates was significantly larger than forgeneral debates on each variable with multiple studies using primary debates. All were statistically significant:knowledge, t(df� 21)� 7.83, p� .001, primary m� 0.797, SD� 0.119, general m� 0.273, SD� 0.136;character, t(df� 18)� 5.40, p� .001, primary m� 0.833, SD� 0.108, general m� 0.268, SD� 0.173; votepreference, t(df� 28)� 6.17, p� .001, primary m� 0.594, SD� 0.181, general m� 0.155, SD� 0.138.

3All reported ns are the number of unique subjects for a particular dependent variable. For example,Holbert et al. (2002a) employed two different groups of subjects and effect sizes were obtainable for each, sothis study appears twice in Table 1, once for each distinct group.

4Many studies (see Table 1) examined all debates for a given election in aggregate, so those studies could notbe included in this analysis.

5Other potential moderators would have been tested (e.g., town hall versus other debate formats, partisansversus nonpartisans) if possible. The nature of the studies in the sample did not permit these tests, however(e.g., no study of issue knowledge in this sample concerned a town hall debate).

6Analysis of presidential television spots also found no significant difference between studies using studentsand nonstudents (Lau, Sigelman, Heldman, & Babbitt, 1999).

7Sophisticated research should not be confused with complex research (Boster, 2002).

References

*Abramowitz, A. I. (1978). The impact of a presidential debate on voter rationality. American Journal of PoliticalScience, 22, 680–690.

Allen, M., & Burrell, N. (2002). The negativity effect in political advertising: A meta-analysis. In J. P. Dillard& M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. 83–96). Thousand Oaks, CA:Sage Publications.

*Atkin, C., Hocking, J., & McDermott, S. (1979). Home state voter response and secondary media coverage.In S. Kraus (Ed.), The great debates: Carter vs. Ford, 1976 (pp. 429–436). Bloomington, IN: Indiana UniversityPress.

Beatty, M. J. (2002). Do we know a vector from a scalar? Why measures of association (not their squares) areappropriate indices of effect. Human Communication Research, 28, 605–611.

*Becker, L. B., Sobowale, I. A., Cobbey, R. E., & Eyal, C. H. (1978). Debates’ effects on voters’ understandingof candidates and issues. In G. F. Bishop, R. G. Meadow, & M. Jackson-Beeck (Eds.), The presidential debates:Media, electoral, and policy perspectives (pp. 126–139). New York: Praeger.

Benoit, W. L. (1999). Seeing spots: A functional analysis of presidential television advertisements from 1952–1996. NewYork: Praeger.

Benoit, W. L., Blaney, J. R., & Pier, P. M. (1998a). Campaign ‘96: A functional analysis of acclaiming, attacking, anddefending. New York: Praeger.

* Study contributed data for the meta-analysis.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 16: A Meta-Analysis of the Effects of Viewing US Presidential Debates

PRESIDENTIAL DEBATES 349

*Benoit, W. L., & Hansen, G. J. (in press). Presidential debate watching, issue knowledge, characterevaluation, and vote choice. Human Communication Research.

*Benoit, W. L., McKinney, M. S., & Holbert, R. L. (2001). Beyond learning and persona: Extending the scopeof presidential debate effects. Communication Monographs, 68, 259–273.

*Benoit, W. L., McKinney, M. S., & Stephenson, M. T. (2002a). Effects of watching campaign 2000presidential primary debates. Journal of Communication, 52, 316–331.

Benoit, W. L., Pier, P. M., Brazeal, L. M., McHale, J. P., Klyukovksi, A., & Airne, D. (2002b). The primarydecision: A functional analysis of debates in presidential primaries. Westport, CT: Praeger.

*Benoit, W. L., & Stephenson, M. T. (in press). Effects of watching a presidential primary debate. ContemporaryArgumentation and Debate.

*Benoit, W. L., Webber, D. J., & Berman, J. (1998b). Effects of presidential debate watching and ideology onattitudes and knowledge. Argumentation and Advocacy, 34, 163–172.

Benoit, W. L., & Wells, W. T. (1996). Candidates in conflict: Persuasive attack and defense in the 1992 presidential debates.Tuscaloosa, AL: University of Alabama Press.

*Ben-Zeev, S., & White, L. (1962). Effects and implications. In S. Kraus (Ed.), The great debates: Background,perspective, effects (pp. 331–337). Bloomington, IN: Indiana University Press.

*Best, S. J., & Hubbard, C. (1999). Maximizing “minimal effects”: The impact of early primary season debateson voter preferences. American Politics Quarterly, 27, 450–487.

Best, S. J., & Hubbard, C. (2000). The role of televised debates in the presidential nominating process. In W.G. Mayer (Ed.), In pursuit of the White House 2000: How we choose our presidential nominees (pp. 255–284). NewYork: Chatham House.

*Bishop, G. F., Oldendick, R., & Tuchfarber, A. (1978). Debate watching and the acquisition of politicalknowledge. Journal of Communication, 28, 99–113.

Boster, F. J. (2002). On making progress in communication science. Human Communication Research, 28,473–490.

Brians, C. L., & Wattenberg, M. P. (1996). Campaign issue knowledge and salience: Comparing receptionfrom TV commercials, TV news, and newspapers. American Journal of Political Science, 40, 172–193.

Carlin, D. P. (1994). A rationale for a focus group study. In D. B. Carlin & M. S. McKinney (Eds.), The 1992presidential debates in focus (pp. 3–19). Westport, CT: Praeger.

*Carter, R. (1962). Some effects of the debates. In S. Kraus (Ed.), The great debates: Background, perspective, effects(pp. 253–270). Bloomington, IN: Indiana University Press.

*Chaffee, S. H. (1978). Presidential debates—Are the helpful to voters? Communication Monographs, 45,330–346.

Coleman, S. (2000). Televised election debates: International perspectives. New York: St. Martin’s.D’Alessio, D., & Allen, M. (2000). Media bias in presidential elections: A meta-analysis. Journal of

Communication, 50, 133–156.D’Alessio, D., & Allen, M. (2002). Selective exposure and dissonance after decisions. Psychological Reports, 91,527–532.

*Desutschmann, P. (1962). Viewing, conversation, and voting intentions. In S. Kraus (Ed.), The great debates:Background, perspective, effects (pp. 232–252). Bloomington: Indiana University Press.

Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press.Friedenberg, R. V. (1994). Patterns and trends in national political debates: 1960–1992. In R. F. Friedenberg(Ed.), Rhetorical studies of national political debates, 1960–1992 (2nd ed., pp. 235–259). Westport, CT: Praeger.

Gallup, G. (1987). The impact of presidential debates on the vote and turnout. In J. L. Swerdlow (Ed.),Presidential debates: 1988 and beyond (pp. 34–42). Washington, DC: Congressional Quarterly.

Gallup Poll. (1999, April 9). Independents rank as largest U.S. political group. Retrieved June 20, 2001, fromhttp://www.gallup.com/poll/releases/pr990409c.asp

*Geer, J. G. (1988). The effects of presidential debates on the electorate’s preference for candidates. AmericanPolitics Quarterly, 16, 486–501.

Glass, G. V., McGaw, B., & Smith, M. L. (1981). Meta-analysis in social research. Beverly Hills, CA: SagePublications.

*Graber, D. A. (1978). Problems in measuring audience effects of the 1976 debates. In G. F. Bishop, R. G.Meadow, & M. Jackson-Beeck (Eds.), The presidential debates: Media, electoral, and policy perspective (pp. 105–125).New York: Praeger.

*Holbert, R. L., Benoit, W. L., Hansen, G. J., & Wen, W.-C. (2002a). The role of political ad recall, news use,political discussion, and debate viewing in campaign issue knowledge and salience. Communication Mono-graphs, 69, 296–310.

Holbert, R. L, Benoit, W. L., & McKinney, M. S. (2002b). The role of debate viewing in establishing “enlightenedpreference” in the 2000 presidential election. Seoul, Korea: ICA.

Holbrook, T. M. (1999). Political learning from presidential debates. Political Behavior, 21, 67–89.Hunter, J. E., & Schmidt, F. L. (1990).Methods of meta-analysis: Correcting error and bias in research findings. NewburyPark, CA: Sage Publications.

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012

Page 17: A Meta-Analysis of the Effects of Viewing US Presidential Debates

COMMUNICATION MONOGRAPHS350

*Jacoby, J., Troutman, T. R., & Whittler, T. E. (1986). Viewer miscomprehension of the 1980 presidentialdebate: A research note. Political Psychology, 7, 297–308.

*Jamieson, K. H., & Adasiewicz, C. (2000). What can voters learn from election debates? In S. Coleman (Ed.),Televised election debates: International perspectives (pp. 25–42). New York: St. Martin’s Press.

Jamieson, K. H., & Birdsell, D. S. (1988). Presidential debates: The challenge of creating an informed electorate. NewYork: Oxford University Press.

Kaid, L. L., & Johnston, A. (2000). Videostyle in presidential campaigns: Style and content of televised political advertising.Westport, CT: Praeger.

*Kraus, S., & Smith, R. G. (1962). Issues and images. In S. Kraus (Ed.), The great debates: Background, perspective,effects (pp. 289–313). Bloomington, IN: Indiana University Press.

*Lang, K., & Lang, G. E. (1962). Reaction of viewers. In S. Kraus (Ed.), The great debates: Background, perspective,effects (pp. 313–330). Bloomington, IN: Indiana University Press.

*Lang, G. E., & Lang, K. (1978). Immediate and delayed responses to a Carter–Ford debate: Assessing publicopinion. Public Opinion Quarterly, 42, 322–341.

*Lanoue, D. J. (1991). The turning point. American Politics Quarterly, 19, 80–95.Lau, R. R., Sigelman, L., Heldman, C., & Babbitt, P. (1999). The effects of negative political advertisements:A meta-analytic assessment. American Political Science Review, 93, 851–875.

Louden, A. (2003). Political debates: Selected bibliography. Retrieved May 12, 2003, from http://www.wfu.edu/%7Elouden/Political%20Communication/Bibs/DEBATES.html

*McKinney, M. S., Kaid, L. L., & Robertson, T. A. (2001). The front-runner, contenders, and also-rans:Effects of watching a 2000 Republican primary debate. American Behavioral Scientist, 44, 2232–2251.

*McLeod, J. M., Durall, J. A., Ziemke, D. A., & Bybee, C. R. (1979). Reactions of young and older voters:Expanding the context of effects. In S. Kraus (Ed.), The great debates: Carter vs. Ford, 1976 (pp. 348–367).Bloomington, IN: Indiana University Press.

*Mulder, R. D. (1978). The political effects of the Carter–Ford debate: An experimental analysis. SociologicalFocus, 11, 33–45.

Nie, N. H., Verba, S., & Petrocik, J. R. (1999). The changing American voter (enlarged ed.). San Jose, CA:toExcel/Harvard University Press.

*Payne, J. G., Golden, J. L., Marlier, J., & Ratzan, S. C. (1989). Perceptions of the 1988 presidential andvice-presidential debates. American Behavioral Scientist, 32, 425–435.

*Pfau, M. (1988). Intra-party political debates and issue learning. Journal of Applied Communication Research, 16,99–112.

Racine Group. (2002). White paper on televised political campaign debates. Argumentation and Advocacy, 38,199–218.

*Rose, D. D. (1976). Citizen uses of the Ford–Carter debates. Journal of Politics, 41, 214–221.Stock, W., Okum, M., Haring, M., Miller, W., Kinney, C., & Williams, R. M. (1982). Rigor in data synthesis:A case study of reliability in meta-analysis. Educational Researcher, 11(6), 10–20.

*Swanson, L. L., & Swanson, D. L. (1978). The agenda-setting functions of the first Ford–Carter debate.Communication Monographs, 45, 347–353.

*Wall, V., Golden, J. L., & James, H. (1988). Perceptions of the 1984 presidential debates and a select 1988presidential primary debate. Presidential Studies Quarterly, 18, 541–563.

Wolf, F. M. (1986). Meta-analysis: Quantitative methods for research synthesis. Beverly Hills, CA: Sage Publications.*Yawn, M., & Beatty, B. (2000). Debate-induced opinion change: What matters? American Politics Quarterly, 28,270–285.

*Zhu, J.-H., Milavasky, J. R., & Biswas, R. (1994). Do televised debates affect image perceptions more thanissue knowledge? A study of the first 1992 presidential debate. Human Communication Research, 20, 302–333.

Received: January 5, 2003Revised: July 7, 2003Accepted: October 25, 2003

Dow

nloa

ded

by [

Uni

vers

ity o

f Pe

nnsy

lvan

ia]

at 1

2:17

15

Aug

ust 2

012