Stealing Thunder in Negative Political Advertising: The Persuasive Impact of One-sided and Two-sided Negative Messages on Partisan Individuals

Abstract

Although two-sided persuasion is a common technique in political speeches, few research has reported the relative persuasiveness of a one-versus two-sided message in the context of negative political advertising. Participants (N = 116) viewed a negative political ad, varying in message sidedness (one-sided vs. two-sided) and partisan match (partisan-match vs. partisan-mismatch vs. unaffiliated). Multivariate results revealed the relative advantage of two-sided persuasion in the partisan-match condition where participants viewed a negative ad targeting their supporting party candidate. One-sided persuasion was more effective in the partisan-mismatch condition where participants viewed a negative ad criticizing the opposition party candidate. Implications for the practice of strategic negative political campaigns were discussed.

Keywords

Message sidedness negative political advertising party bias persuasion two-sided persuasion

Introduction

Voters are bombarded with attack ads during every election. The 2016 US presidential election was no different from any other election; no less than 92 per cent of about 70,000 political TV ads addressed negative information about the opponent during the last few weeks of the election (Wallace, 2016). As such, negative political advertising (NPA) has been an indispensable tool for political office (Cassese & Holman, 2018; Fridkin & Kenney, 2008). Research suggests that a well-crafted negative ad helps strengthen the sponsor’s support base, provoking negative feelings towards the opponent (Kaid & Boydston, 1987; Perloff, 2002; Pinkleton, Um, & Austin, 2002). However, the consequences of NPA go beyond the intended effects; negative ads are likely to create a backlash from voters (Ansolabehere & Iyengar, 1995; Cappella & Jamieson, 1997; Schenck-Hamlin, Procter, & Rumsey, 2000). For instance, a recent study presents evidence that a candidate’s support withers over time as the candidate’s attacks on the opponent increase over the course of the campaign, suggesting that candidates who enjoy massive attacks may undermine themselves after all (Banda & Windett, 2016). Partisan voters, especially in a two-party system, are likely to depreciate negative information about their favourite candidate, raising counterarguments (Faber, Tims, & Schmit, 1993; Kaid & Boydston, 1987; Merritt, 1984). The present study focussed on the strategic use of message sidedness (i.e., one-versus two-sided messages) in negative advertising and its persuasive impact on partisan individuals. Although message sidedness, a classic approach to persuasion (Hovland, Janis, & Kelly, 1953), has been examined in product advertising extensively (see Eisend, 2006), few research has reported the relative persuasiveness of a one-versus two-sided message in the context of NPA. This dearth comes as a surprise, considering two-sided arguments are a common persuasive technique in political speeches (Allen, 1991; O’Keefe, 2013; Perloff, 2003); politicians often strategically address their own mistakes or weakness to neutralize future counterattacks and show honesty (see O’Keefe, 1999; Pfau & Burgoon, 1988).

If applied to NPA, a two-sided negative ad deliberately addresses some positive as well as negative aspects of the targeted candidate to neutralize the negative tone, prepare for potential counterarguments, and add trustworthiness to the argument overall. So, in a sense, a two-sided negative ad can be a kind of novel and creative message technique for campaign practitioners, as it takes a risk of highlighting a strength of the targeted candidate, albeit strategically. On the other hand, a one-sided negative ad is deemed as the typical form of NPA and addresses only negative information about the target, the focus and goal of the message considered. In light of this, the present study sought to examine whether the relative persuasiveness of a one-versus two-sided negative ad varies as a function of voter–target partisan match (e.g., voter–target partisan match vs. voter–target partisan mismatch vs. unaffiliated voters). Partisanship or partisan posture serves as the primary bias in political decisions (Druckman, 2001) when partisans evaluate political messages (Henderson & Theodoridis, 2017; Kennamer, 1994). Relatively few research has reported how one-sided vs. two-sided messages strategically placed in negative political ads work with partisanship. Hence, the findings of this study may provide a guideline for developing effective negative political ad messages by illuminating how political practitioners can utilize and adapt one-sided and two-sided arguments across partisan voters in a polarized party system.

Message Sidedness as Strategic Persuasive Message Tactic

A one-sided persuasive message addresses only one-sided aspects or attributes (i.e., positive) of a person, product, or service. A two-sided message, on the other hand, is considered a kind of strong argument, where negative as well as positive information about a target or offering is strategically highlighted to mitigate the future surprise beforehand (see Dolnik, Case, & Williams, 2003; Pizzutti, Basso, & Albornoz, 2016). Research suggests that the relative effectiveness of a one-versus two-sided message differs as a function of initial attitude or prior perception (see Kao, 2011; O’Keefe, 1999; Sanaktekin & Sunar, 2008). An early study reported that one-sided messages were more persuasive than two-sided messages for people who had a positive initial attitude towards the message position or source (Hovland et al., 1953). A group of scholars in consumer behaviour further reported the relative advantage of a one-sided message (vs. a two-sided message) when an offering was a low risk-choice (e.g., low-priced products or familiar, experienced, in-group political figures) (Lang, Lee, & Zwick, 1999). Relatively recent work suggests that consumers may feel that they are less informed when presented with only one-sided attributes, and thus are less convinced about their opinions formed compared to when presented with two-sided attributes (Rucker, Petty, & Briñol, 2008). This suggests that using two-sided persuasion may enhance argument quality and credibility of persuasive messages among consumers who care about how they reach to a conclusion or judgment (see Chebat & Picard, 1988; Rucker et al., 2008).

Consumers tend to infer that a message is trustworthy when a shortcoming of an offering is addressed by the message source (see Golden & Alpert, 1987; Hoyer & Maclnnis, 2004). Research suggests that a two-sided message can be more effective than a one-sided message in some situations: when people are initially opposed to the message position (Cowley & Hoyer, 1994; Eisend, 2007), when a judgment problem involves risky elements (e.g., high-priced products, novice/out-group political candidates) (Etgar & Goodwin, 1982), or when people may encounter numerous counterarguments (e.g., last few days before the election day) (Chebat & Picard, 1988).

The Interaction of Message Sidedness and Partisan Match in Negative Political Advertising

Work in social identity suggests that partisan individuals are motivated to assess negative political messages in light of their partisan posture (Stevens, Sullivan, Allen, & Alger, 2008; Westen, Blagov, Harenski, Kilts, & Hamann, 2006). Research suggests that partisan voters perceive negative ads from the opposition party to be less persuasive and less agreeable than negative ads from their side (Chang, 2003). As such, partisan posture or party bias can serve as a strong heuristic for voters to rely on when they evaluate negative political messages especially in a two-party system (see Feldman, 2011; Henderson & Theodoridis, 2017). An early study revealed that after exposure to a negative ad, Republicans evaluated the Democratic candidate more unfavourably than the Republican candidate, Democrats reported less attitude changes than Republicans when encountering criticisms about the Democratic candidate, and unaffiliated voters were relatively less affected by the negative ad than partisan voters (Kaid & Boydston, 1987). Relatively recent research further suggests that partisans tend to dismiss political claims fairly and objectively made by opposition parties or news media as biased, untrustworthy, or of secondary importance (see Feldman, 2011; Huge & Glynn, 2010; Matheson & Dursun, 2001). In this regard, the present study highlights the importance of strategic negative appeals to determine the pattern of the effectiveness of NPA for partisan individuals.

NPA involves implicit or explicit intergroup comparisons. From an information-processing perspective, intergroup competition facilitates motivated reasoning when partisan individuals evaluate group-related information (Mackie & Ahn, 1998). Specifically, partisans tend to be more responsive and open to criticisms about the opponent, whereas they are likely to have defensive motives when a political message counters their expectations or prior attitudes (see Wiest, Raymond, & Clawson, 2015). Hence, when partisans encounter a negative ad criticizing their supporting candidate, they are likely to question the legitimacy of the argument, raising counterarguments (Faber et al., 1993; Kaid & Boydston, 1987). The present study suggests that a two-sided negative ad (vs. a one-sided negative ad) may have a greater impact on partisans when the ad criticizes their supporting candidate. In the context of NPA, a one-sided negative ad would address only negative aspects of the targeted candidate (i.e., the message goal), whereas a two-sided negative ad would highlight some positive as well as negative aspects of the target. As such, a two-sided negative ad deliberately admits some good points (e.g., some past achievements) about the target in the hope of undermining potential counterarguments and earning some credibility from voters; however, the positive points highlighted should be of secondary importance to the voting decision (e.g., ‘He used to be one of the most promising human right lawyers before he entered politics. People now call him the most corrupt politician in America.’) (see Hoyer & Maclnnis, 2004). Partisan individuals encountering a negative ad targeting their supporting candidate could have a defensive mindset as the message goal was against their initial preference. Research suggests that defensive individuals are likely to counter-argue negative information incongruent with their prior attitudes (see Agrawal & Maheswaran, 2005). Considering that voters are likely to encounter or raise numerous counterarguments over the course of the campaign especially in a two-party race, it could be argued that a two-sided strategy (vs. one-sided) might be more effective when partisans encounter negative ads criticizing their supporting candidate (i.e., the partisan-match condition in this study). Research further suggests that a two-sided message can be more effective than a one-sided message for people with no initial attitude or preference (e.g., unaffiliated voters) (see Etgar & Goodwin, 1982). On the other hand, a one-sided negative ad targeting the opposition party candidate (i.e., the partisan-mismatch condition in this study) would be preference-consistent information for partisans; the message goal stands for their partisan posture. The above theorizing suggests that a one-sided message is expected to be more effective than a two-sided message when the persuasive goal is congruent with prior attitude or expectation (see Hovland et al., 1953). Thus, the following hypotheses were proposed.

H1: There will be an interaction between message sidedness and partisan match on attitudes towards the targeted candidate. Specifically, a two-sided negative ad (versus a one-sided negative ad) will generate less favourable attitudes towards the target in the partisan-match condition (H1a), a one-sided negative ad (versus a two-sided negative ad) will generate less favourable attitudes towards the target in the partisan-mismatch condition (H1b), and a two-sided negative ad (versus a one-sided negative ad) will generate less favourable attitudes towards the target in the unaffiliated condition (H1c).

H2: There will be an interaction between message sidedness and partisan match on voting intention for the targeted candidate. Specifically, a two-sided negative ad (versus a one-sided negative ad) will induce less voting intention for the target in the partisan-match condition (H2a), a one-sided negative ad (versus a two-sided negative ad) will induce less voting intention for the target in the partisan-mismatch condition (H2b), and a two-sided negative ad (versus a one-sided negative ad) will induce less voting intention for the target in the unaffiliated condition (H2c).

Method

Sample and Procedure

Participants were recruited from introductory journalism classes at a southwestern state university in the US. A total of 116 college students (72 women and 44 men) participated in return for course credit. The average age was 20.16 (SD = 1.23). The majority (86.2 per cent) were whites, followed by African Americans (4.3 per cent), American Indians (4.3 per cent), Hispanics (2.6 per cent), and Asian Americans (2.6 per cent). More than half of the participants (N = 116) were Republicans (53.4 per cent), followed by unaffiliated voters (24.1 per cent) and Democrats (22.4 per cent). The present study developed a 2 (message sidedness: one-sided vs. two-sided) × 3 (partisan match: partisan-match versus partisan-mismatch versus unaffiliated) between-subjects design. Sessions were conducted in a laboratory. Based on partisanship, participants were randomly assigned to one of the six experimental conditions. After exposure to the stimulus, participants completed dependent measures and provided demographic information. It took approximately 20 minutes to complete the instrument. Upon completion of the instrument, participants were thanked and debriefed.

Stimulus Materials: One-versus Two-sided Negative Ad

Participants read a plain printed advertising copy without any source information. To develop stimuli for the experiment, factual elements from existing negative political advertisements were employed (see Etgar & Goodwin, 1982). In the ad, a Congressman John Sutherland (fictitious character) was being criticized as ‘big oil company’s friend’. The negative ad messages highlight one of two message sidedness conditions at the beginning: a one-sided negative ad including only unfavourable arguments that the targeted candidate campaigned on economy cure and oil price stabilization but he didn’t keep up to his promise; and a two-sided negative ad, including both positive and negative information, that the target had been a reliable worker before he entered the politics, and many people supported his promise of economy cure and oil price stabilization, but he didn’t keep up to his promise (see Table 1 for a full description of the negative ads).

Table 1.

One-versus Two-sided Negative Ads

Sidedness	Negative Ad
One-sided	Democrat (Republican) Congressman John Sutherland campaigned on economy cure and oil price stabilization from the very beginning and he didn’t keep up to his promise. He took that opportunity and he blew it. Congressman John Sutherland has spent the last three years in Washington becoming big oil’s trusted friend. He has taken over ten million dollars from other states’ special interests and fat cats. His campaign has taken more than a million dollars from those big oil and gas companies already. Check your oil in Oklahoma because it’s time for a change.
Two-sided	Democrat (Republican) John Sutherland had been a reliable worker for our economy before he became a congressman. Many people supported his promise of economy cure and oil price stabilization from the very beginning. But he didn’t keep up to his promise. He took that opportunity and he blew it. Congressman John Sutherland has spent the last three years in Washington becoming big oil’s trusted friend. He has taken over ten million dollars from other states’ special interests and fat cats. His campaign has taken more than a million dollars from those big oil and gas companies already. Check your oil in Oklahoma because it’s time for a change.

Source: The author.

Partisan Match

The partisan match conditions were created based on participant-target partisanship (i.e., partisan-match vs. partisan-mismatch vs unaffiliated). Participants answered a filter question about their political party affiliation first, and then were assigned to one of the six experimental conditions before they viewed the stimulus. For example, Republican participants viewed a negative ad criticizing the Republican (Democratic) party candidate for the partisan-match condition (partisan-mismatch condition). Considering the two-party politics of the US (see Henderson & Theodoridis, 2017; Houston, Doan, & Roskos-Ewoldsen, 1999), the targeted candidate was portrayed as either a Democrat or Republican party candidate in the ad. Participants with no party preference viewed an ad targeting either the Republican or Democratic party candidate.

Dependent Measures

This study measured attitudes towards the targeted candidate and intention to vote for the target as the dependent measures. Attitudes were measured with a feeling thermometer scale generally used by the National Election survey and political advertising research to assess the degree of favourability towards a political candidate (see Kaid, 2004). Participants rated the targeted candidate on a scale from 0 (cool) to 100 (warm; M = 32.39, SD = 18.18). Voting Intention for the target was measured with an item on a 7-point Likert scale with 1 representing very unlikely to vote and 7 representing very likely to vote (M = 2.41, SD = 1.39).

Control Variables

Work in political communication suggests that the effects of political messages can be influenced by pre-experimental factors such as political knowledge, political cynicism, and involvement with politics (see Schenck-Hamlin et al., 2000; Shah, Kwak, Schmierbach, & Zubric, 2004). Hence, despite the randomization of participants, this study attempted to control the potential influence of some confounding factors on the results (see Shah et al., 2004). Two individual-level variables were included in the analysis as control variables: issue involvement and perceived credibility of NPA. The involvement was measured with two items; ‘how important is the issue addressed in the ad to you personally’, and ‘how relevant is the issue addressed in the ad to you personally’. Each item was rated on a 7-point Likert scale from not at all to very much (M = 4.79, SD = 1.47, α = 0.79). Perceived credibility of NPA was measured with four items; ‘NPA is untrustworthy’, ‘NPA is biased’, ‘NPA is unbelievable’, and ‘NPA is not reliable’. Each item was rated on a 7-point Likert scale from strongly disagree to strongly agree (M = 3.49, SD = 1.13, α = 0.82).

Results

The hypotheses were developed to test the interaction effects of message sidedness and partisan match on attitudes and voting intention for the targeted candidate. Because a moderate correlation was expected among the two dependent variables, a multivariate analysis of covariance (MANCOVA) was used as the main method of hypotheses testing. The two dependent measures were submitted to a 2 (message sidedness: one-sided vs. two-sided) × 3 (partisan-match versus partisan-mismatch versus unaffiliated) between-subjects MANCOVA with issue involvement and perceived credibility of NPA as covariates. As Table 2 shows, the MANCOVA revealed that the message sidedness × partisan match interaction was significant [F(4, 214) = 4.02, p = 0.004, η² = 0.07] although the two covariates were controlled for. This indicated that the effects of one-sided and two-sided negative ads on the dependent variables significantly varied across the partisan match conditions and that the two covariates did not account for the observed interaction effect. The analysis overall indicated that a two-sided negative ad (vs. a one-sided negative ad) was more persuasive in the partisan-match condition, whereas a one-sided negative (vs. a two-sided negative ad) was more effective in the partisan-mismatch condition. To further explore how the interaction unfolded, analyses of simple effects with the covariates were conducted for each dependent variable.

Table 2.

MANCOVA Results of the Two Dependent Variables

Independent Variables	Wilks’ Λ	F	H df	Error df	P	η²
Message sidedness (one-vs. two-sided)	0.99	0.43	2	107	0.65	0.01
Partisan match (partisan-match vs. partisan-mismatch vs. independent)	0.92	2.30	4	214	0.06	0.04
Message sidedness × partisan identification	0.87	4.02	4	214	0.004	0.07
Issue involvement (covariate 1)	0.93	4.05	2	107	0.02	0.07
Perceive credibility of negative political advertising (covariate 2)	0.89	6.51	2	107	0.002	0.11

Source: The author.

Attitudes Towards the Targeted Candidate

Additional analyses indicated that a two-sided negative ad yielded significantly less favourable attitudes towards the target than a one-sided negative ad in the partisan-match condition [F(1, 108) = 9.22, p = 0.003, M_{partisan match-one sided} = 34.59, SD = 18.11, M_{partisan match-two sided} = 24.38, SD = 18.61] (see Figure 1). This strongly supports the idea that two-sided persuasion can be more effective for partisans encountering attacks on their supporting party candidate. Further, Cohen’s effect size value (d = 0.56) suggested moderate practical significance (see Bowman, 2017). Therefore, H1a was supported. On the other hand, a one-sided negative ad generated less favourable attitudes towards the target than a two-sided message in the partisan-mismatch condition, but the difference was marginal [F(1, 108) = 3.27, p = 0.073, M_{partisan mismatch-one sided} = 28.93, SD = 17.14, M_{partisan mismatch-two sided} = 39.87, SD = 17.08]. Thus, H1b was not supported at p < 0.05. However, Cohen’s value (d = −0.64) indicated a moderate to high effect. Meanwhile, when unaffiliated participants viewed a negative ad targeting either Democratic or Republican party candidate, a one-sided negative ad (vs. a two-sided negative ad) generated less favourable attitudes towards the target; however, the result was contrary to the expectation of the study [F(1, 108) = 3.25, p = 0.071, M_{unaffiliated-one sided} = 31.13, SD = 15.46, M_{unaffiliated-two sided} = 44.08, SD = 14.47, d = −0.86]. Therefore, H1c was not supported.

Figure 1.

Source: The author.

Voting Intention for the Targeted Candidate

Additional analyses for voting intention revealed no significant difference between one-sided and two-sided negative ads in the partisan-match condition [F(1, 108) = 0.87, p = 0.354, M_{partisan match-one sided} = 2.42, SD = 1.79, M_{partisan match-two sided} = 2.80, SD = 1.57, d = −0.23]. Similar results were obtained for the partisan-mismatch [F(1, 108) = 2.10, p = 0.15, M_{partisan mismatch-one sided} = 2.26, SD = 1.16, M_{partisan mismatch-two sided} = 1.84, SD = 0.95, d = 0.40] and unaffiliated [F(1, 108) = 1.45, p = 0.232, M_{unaffiliated-one sided} = 2.67, SD = 1.18, M_{unaffiliated-two sided} = 3.38, SD = 1.75, d = −0.48] conditions that none of the message sidedness effect on voting intentions approached statistical significance at p < 0.05. Therefore, H2a~c were not supported.

Discussion

While voter evaluations of candidates emerge through multiple layers of influence, the present study focussed on the strategic use of particular persuasive message tactics in the context of negative advertising. Despite its pervasiveness in political communications, product ads, and legal settings (see Williams, Bourgeois, & Croyle, 1993), message sidedness has received little attention in NPA. This study sought to extend the NPA literature by investigating whether an attack ad could benefit from using two-sided arguments when the ad targeted partisan voters’ supporting candidate. The use of strategic messages such as this is particularly important in the current two-party politics. Political scientists contend that polarized voters tend to selectively accept or reject particular political messages based on their partisan posture (see Henderson & Theodoridis, 2017; Iyengar & Han, 2009). Direct attack ads from the opposition party may not work well for partisan voters because partisan voters seek to defend the legitimacy of their party identity when the position of the arguments counters their political or ideological views. The obtained data support this view that partisan leanings override external information in candidate evaluations.

This study offers an explanation about the pattern of the effectiveness of message sidedness for partisan voters: the relative advantage of two-sided persuasion for partisans who may have defensive motives, raising counterarguments especially when they process unfavourable political information and the relative advantage of one-sided persuasion for partisans who may have impressive motives, making support arguments when they encounter a political message congruent with their partisan view. Future studies may want to investigate whether a two-sided negative ad (versus a one-sided negative ad) elicits less counterarguments and less source derogations from a message processing perspective.

Limitations and Suggestions for Future Research

This study has some important limitations. First, this study failed to detect the expected interaction effects on voting intention. One of the reasons for the null effects might be that participants recognized the hypothetical target candidate. It is plausible to argue that participants might not be interested in voting judgment or voting intention regardless of their attitudes formed as a result of ad exposure. Further, a research study (Keren, 2007) suggests that people tend to perceive two-sided messages as more trustworthy and have positive attitudes towards the messages, but their actual judgments can be asymmetric with these perceptions induced by message exposure. Second, one might raise a question about the ecological validity of the study. Although it may be unusual that a negative political ad addresses positive information about the target, stealing the opponent’s thunder in advance is a quite common strategy in competitive contexts (see Dolnik et al., 2003; Williams et al., 1993). Considering widespread and growing disgust with negative campaigns today, it is important for campaign practitioners to develop attack messages that sound agreeable but critical to candidate evaluations. One could also question about the validity of the operationalization for the two-sided negative ad—is it really two-sided? The development of the two-sided message in the present study was based on some actual two-sided advertisements highlighting past events (positive or negative) as two-sided attributes: for example, ‘Thirty years ago, GM quality was the best in the world. Twenty years ago, it wasn’t. The story of our long journey back’. In this regard, presenting a past positive attribute (e.g., how good she or he used be) of the targeted candidate could be a two-sided negative ad tactic as it may not be common that a one-sided attack ad presents a positive achievement or reputation of the target, albeit a thing of the past. Third, this study is based on a convenient sample of college students at a Southwestern state university. Therefore, the results may not generalize to the general voter population. However, it does provide insight into young voters’ perceptions of one-sided versus two-sided persuasion in NPA. Future research may be conducted to explore how the results may vary across various contexts or the demographic segments especially because different audience segments may have different political orientation. Another potential limitation is that the experimental stimuli were entirely text-based for effective manipulations of the content elements. Even though similar approaches have been used in previous research, it would be meaningful to replicate the findings using more realistic versions of NPA. Finally, the present study did not focus on the impact of ad sponsors, a critical factors for voters to evaluate a political message. Future research may want to examine an interaction of message sidedness, partisan match, and ad source; low credible ad sponsors might benefit from using two-sided persuasion. Despite limitations, the study presents significant findings on negative political message strategies, on the importance of political partisanship, and on these variables in relation to partisan voters. Hopefully, study finding will spur further investigation on NPA and give insight to strategic message design in political advertising campaigns.

Implications

Many have questioned the efficacy of negative advertising in the political process. For example, a group of scholars reported only minimal impact of negative ads (vs. positive ads) on voter recall (Lau, Sigelman, & Rovner, 2007) or no evidence that negative ads were linked to political efficacy and political attitudes formed (Jackson, Mondak, & Huckfeldt, 2009). Despite reservations about the effectiveness of NPA, political candidates are running negative ads more and more in every election, exploiting polarized partisanship to tailor their messages to the target audiences. Thus, the use of various negative strategies and its impact on partisan voters deserve further attention to understand how negative campaigns influence participation in the voting process.

The results of this study have important implications for the development of effective NPA messages across different partisan voters. For instance, a negative campaign plan can be designed in light of the political orientation in a geographic market; a Democratic party candidate could effectively attack a Republican candidate by incorporating two-sided arguments in his or her campaigns in southern states such as Oklahoma and Texas, which contain relatively more Republicans than Democrats. The data also show that a one-sided message was more effective when voters viewed an ad targeting the opposition party candidate, and the same effect occurred for unaffiliated voters. Therefore, to maximize the effectiveness of a negative campaign, practitioners will need to assess the political party orientation in detail across demographic consumer segments such as race, income, and occupation as well as across geographic markets and distribute one-and two-sided ads accordingly.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

References

Agrawal

, & Maheswaran

(2005). Motivated reasoning in outcome-bias effects. Journal of Consumer Research, 31(4), 798–805.

Allen

(1991). Meta-analysis comparing the persuasiveness of one-sided and two-sided messages. Western Journal of Speech Communication, 55, 390–404.

Ansolabehere

, & Iyengar

(1995). Going negative: How political advertisements shrink and polarize the electorate. New York, NY: The Free Press.

Banda

K. K.

, & Windett

J. H.

(2016). Negative advertising and the dynamics of candidate support. Political Behavior, 38(3), 747–766.

Bowman

N. D.

(2017). The importance of effect size reporting in communication research. Communication Research Reports, 34(3), 187–190.

Cassese

E. C.

, & Holman

M. R.

(2018). Party and gender stereotypes in campaign attacks. Political Behavior, 40(3), 785–807.

Cappella

J. N.

, & Jamieson

K. H.

(1997). Spiral of cynicism: The press and the public good. New York, NY: Oxford University Press.

Chang

(2003). Party bias in political-advertising processing. Journal of Advertising, 32(2), 55–67.

Chebat

J.-C.

, & Picard

(1988). Receivers’ self-acceptance and the effectiveness of two-sided messages. The Journal of Social Psychology, 128(3), 353–362.

10.

Cowley

A. E.

, & Hoyer

W. D.

(1994). An integrative framework for understanding two-sided persuasion. Journal of Consumer Research, 20, 561–574.

11.

Dolnik

, Case

T. I.

, & Williams

K. D.

(2003). Stealing thunder as a courtroom tactic revisited: Processes and boundaries. Law and Human Behavior, 27(3), 267–287.

12.

Druckman

J. N.

(2001). The implications of framing effects for citizen competence. Political Behavior, 23, 225–256.

13.

Eisend

(2006). Two-sided advertising: A meta-analysis. International Journal of Research in Marketing, 23, 187–198.

14.

Eisend

(2007). Understanding two-sided persuasion: An empirical assessment of theoretical approaches. Psychology and Marketing, 24, 615–640.

15.

Etgar

, & Goodwin

S. A.

(1982). One-sided versus two-sided message appeals for new brand introduction. Journal of Consumer Research, 8, 460–465.

16.

Faber

J. R.

, Tims

R. A.

, & Schmitt

G. K.

(1993). Negative political advertising and voting intent: The role of involvement and alternative information sources. Journal of Advertising, 22(4), 67–76.

17.

Feldman

(2011). Partisan differences in opinionated news perceptions: A test of the hostile media effect. Political Behavior, 33, 407–432.

18.

Fridkin

K. L.

, & Kenney

P. J.

(2008). The dimensions of negative messages. American Politics Research, 36(5), 694–723.

19.

Golden

L. L.

, & Alpert

M. I.

(1987). Comparative analysis of the relative effectiveness of one-and two-sided communication for contrasting products. Journal of Advertising, 16, 18–25.

20.

Henderson

J. A.

, & Theodoridis

A. G.

(2017). Seeing sports: Partisanship, negativity and the conditional receipt of campaign advertisements. Political Behavior. Advance online publication. doi. org/10.1007/s11109-017-9432–6.

21.

Houston

D. A.

, Doan

, & Roskos-Ewoldsen

(1999). Negative political advertising and choice conflict. Journal of Experimental Psychology: Applied, 5(1), 3–16.

22.

Hovland

C. I.

, Janis

I. L.

, & Kelley

H. H.

(1953). Communication and persuasion. New Haven, CT: University Press.

23.

Hoyer

W. D.

, & Maclnnis

D. J.

(2004). Consumer behavior (3rd ed.). Boston, MA: Houghton Mifflin Company.

24.

Huge

, & Glynn

C. J.

(2010). Hostile media and the campaign trail: Perceived media bias in the race for governor. Journal of Communication, 60(1), 165–181.

25.

Iyengar

, & Hahn

K. S.

(2009). Red media, blue media: Evidence of ideological selectivity in media use. Journal of Communication, 59, 19–39.

26.

Jackson

R. A.

, Mondak

J. J.

, & Huckfeldt

(2009). Examining the possible corrosive impact of negative advertising on citizens’ attitudes toward politics. Political Research Quarterly, 62(1), 55–69.

27.

Kaid

L. L.

(2004). Measuring candidate images with semantic differentials. In Hacker

K. L.

(Eds.), Presidential candidate images (pp. 231–236). Westport, CT: Praeger.

28.

Kaid

L. L.

, & Boydston

(1987). An experimental study of the effectiveness of negative political advertisements. Communication Quarterly, 35, 193–201.

29.

Kao

D. T.

(2011). Message sidedness in advertising: The moderating role of need for cognition and time pressure in persuasion. Scandinavian Journal of Psychology, 52, 329–340.

30.

Kennamer

D. J.

(1994). Public opinion, the press, and public policy. Westport, CT: Greenwood Publishing Group, Inc.

31.

Keren

(2007). Framing, intentions, and trust–choice incompatibility. Organizational Behavior and Human Decision Processes, 103(2), 238–255.

32.

Lang

, Lee

C. K.

, & Zwick

(1999). Message sidedness at the brand and product from levels: Overcoming the shortcomings of two-sided messages. Advances in Consumer Research, 26, 485–490.

33.

Lau

R. R.

, Sigelman

, & Rovner

I. B.

(2007). The effects of negative political campaigns: A meta-analytic reassessment. The Journal of Politics, 69(4), 1176–1209.

34.

Mackie

D. M.

, & Ahn

M. N.

(1998). Ingroup and outgroup inferences: When ingroup bias overwhelms outcome bias. European Journal of Social Psychology, 28(3), 343–360.

35.

Matheson

, & Dursun

(2001). Social identity precursors to the hostile media phenomenon: Partisan perceptions of coverage of the Bosnian conflict. Group Processes & Intergroup Relations, 4(2), 116–126.

36.

Merritt

(1984). Negative political advertising: Some empirical findings. Journal of Advertising, 13, 27–38.

37.

O’Keefe

D. J.

(1999). How to handle opposing arguments in persuasive message: A meta-analytic review of the effects of one-sided and two-sided messages. Annals of the International Communication Association, 22(1), 209–249.

38.

O’Keefe

D. J.

(2013). The relative persuasiveness of different message types does not vary as a function of the persuasive outcome assessed: Evidence from 29 meta-analyses of 2,062 effect sizes for 13 message variations. Annals of the International Communication Association, 37(1), 221–249.

39.

Pfau

, & Burgoon

(1988). Inoculation in political campaign communication. Human Communication Research, 15(1), 91–111.

40.

Perloff

R. M.

(2002). Political campaign persuasion and its discontents: Perspectives from the past and research prescriptions for the future. In Dillard

J. P.

& Pfau

(Eds.), The persuasion handbook: Developments in theory and practice (pp. 605–632). London: Sage.

41.

Perloff

R. M.

(2003). The dynamics of persuasion: Communication and attitudes in the 21st century. Mahwah, NJ: Lawrence Erlbaum Associates.

42.

Pinkleton

B. E.

, Um

N. H.

, & Austin

E. W.

(2002). An exploration of the effects of negative political decision making. Journal of Advertising, 31(1), 13–25.

43.

Pizzutti

, Basso

, & Albornoz

(2016). The effect of the discounted attribute importance in two-sided messages. European Journal of Marketing, 50(9/10), 1703–1725.

44.

Rucker

D. D.

, Petty

R. E.

, & Briñol

(2008). What’s in a frame anyway? A meta-cognitive analysis of the impact of one versus two sided message framing on attitude certainty. Journal of Consumer Psychology, 18(2), 137–149.

45.

Sanaktekin

O. H.

, & Sunar

(2008). Persuasion and relational versus personal bases of self-esteem: Does the message need to be one-or two-sided? Social Behavior and Personality, 36(10), 1315–1332.

46.

Shah

, Kwak

, Schmierbach

, & Zubric

(2004). The interplay of news frames on cognitive complexity. Human Communication Research, 30(1), 102–120.

47.

Schenck-Hamlin

W. J.

, Procter

D. E.

, & Rumsey

D. J.

(2000). The influence of negative advertising frames on political cynicism and politician accountability. Human Communication Research, 26(1), 53–74.

48.

Stevens

, Sullivan

, Allen

, & Alger

(2008). What’s good for the goose is bad for the gander: Negative political advertising, partisanship, and turnout. Journal of Politics, 70(2), 527–541.

49.

Wallace

(2016, November 8). Negative ads dominate in campaign’s final days. CNN. Retrieved from http://www.cnn.com/2016/11/08/politics/negative-ads-hillary-clinton-donald-trump/index.html

50.

Westen

, Blagov

, Harenski

, Kilts

, & Hamann

(2006). Neural bases of motivated reasoning: An FMRI study of emotional constraints on partisan political judgment in the 2004 U.S. Presidential election. Journal of Cognitive Neuroscience, 18(11), 1947–1958.

51.

Wiest

S. L.

, Raymond

, & Clawson

R. A.

(2015). Framing, partisan predispositions, and public opinion on climate change. Global Environmental Change, 31, 187–198.

52.

Williams

K. D.

, Bourgeois

M. J.

, & Croyle

R. T.

(1993). The effects of stealing thunder in criminal and civil trials. Law and Human Behavior, 17(6), 597–609.