Research Article Review


ResearchArticle Review

(MediaInfluences and Third-Person Perceptions in Voting)

Rationaleof the Study

Thepurpose of the study is to find out the influence exerted by media onvoters. The study establishes that ambivalent voters they need toinformation in order to make election decisions. Thus, this group ofvoters is more vulnerable to manipulation compared to univalentvoters. This means that election coverage exerts much influence onambivalent voters, which in turn creates a lesser discrepancies inperception of self-other in relation to coverage susceptibility.Univalent voters make their decisions before the time for theelection and thus political campaigns have little influence on theirelection decisions. To them, perceptual influence in electioncampaign news may increase only slightly, only when they create adesirable influence. This study exposes the extent of vulnerabilityof the voters that is created by campaign news coverage. There existsa superiority bias, where some people perceive they are lessvulnerable than others, and try to boost their self esteem. Thiscreates a discrepancy in perception of self-other.

TheoriesUsed In the Study

Thepaper utilizes two theories. First is the Information ProcessingTheory of Ambivalent as proposed by Rudolph and Popp 2007 (Chang,2014). He asserts that ambivalent voters take longer time to arrivein a voting decision. The theory suggests that the percentage ofambivalent voters decreases as the campaign events unfold. The theoryof cognitive dissonance, as argued by Festinger, is applied in thestudy (Chang, 2014). There is a tendency for people to agree with theinformation that match their earlier attitudes because if theydisagree, it is likely to stimulate cognitive dissonance. This theorydemonstrates that despite the lesser impact of campaign informationon univalent voters, where there is a high desirability of intendedinformation, univalent voters may create positive perception aboutthe intended information, for example information that makes themaccept their susceptibility helps them avoid cognitive dissonance.

ResearchQuestions and Hypothesis Explored

Theresearch question addressed was “Areambivalent voters more or less likely to engage in actions againstbiased news coverage and media than univalent voters?” Varioushypotheses were tested. Hypothesis 1 predicted that ambivalent voterswere alleged to be more subject to influence by election news morethan univalent voters(Chang, 2014). Hypothesis2a and 2b suggested positive coverage would have greater influence onboth univalent and ambivalent supporters of the featured candidatethan the supporters of the opposing candidate (Chang,2014). 2apredicted the influence on Ma’s support while 2b predicted theeffect on Tsai supporters.

Inaddition, hypothesis 3 proposed that univalent and ambivalent votersthat do not support the featured candidate will suppose moreinfluence to their opponents than to themselves. 3a focused onsupporters for Ma, while 3b focused on Tsai supporters. Hypothesis 4predicted that ambivalent voters have a smaller Third PersonPerception (TPP) of the effects of media than in univalent voters(Chang, 2014).

Moreover,hypothesis 5 and 6 predicted that the TPP generated from positive andnegative coverage of a candidate on supporters of a candidate morethan the supporters of the rivaled candidate(Chang, 2014). Hypothesis5b supported a smaller TPP on Ma`s ambivalent voters and supportersthan on Tsai supporters. Hypothesis 5b supported a smaller TPP onTsai ambivalent voters and supporters than on Ma supporters(Chang, 2014). Hypothesis6a held that negative coverage of Ma will generate a smaller TPPamong the supporters and ambivalent voters of Tsai while hypothesis6b held that negative coverage of Tsai will generate a smaller TPPamong the supporters and ambivalent supporters of Ma (Chang,2014).


Thestudy procedure involved the online survey on the 2012 Taiwan’spresidential elections campaign. The survey was conducted throughoutthe campaign period until the election date. The respondents werecategorized in terms of voter ambivalence into univalent voters andambivalent voters. The respondents were required to respond withtheir voting intention between Tsai and Ma. The responses were testedusing the Griffins ambivalence formula, (M + T)/2, where M is thelikelihood for voting Ma, and T is the likelihood for voting Tsai.Responses were also measured in terms of perception on election news.In a scale of 1-5 (1-not at all, 5-a great deal), they rated theinfluence of the different types of media including online,newspaper, television, databases and other related news(Chang, 2014).


Theresults upheld the first hypothesis. Contrary to hypotheses 2a and2b, there was no significant difference between the Ma supports andTsai supporters. There was a greater influence among the ambivalentvoters for than the univalent voters for both Ma and Tsai supporters.Negative coverage of Ma was found to have greater effects onambivalence of voters. However, there was no significant contrastbetween the effects of ambivalent Ma supporters and Tsai supporters(Chang, 2014). Theseresults supported hypothesis 3a. Pertaining to the hypothesis 3b,there was a significant influence of ambivalent Tsai supporters.

Therewas a significant difference between Ma supporters and Tsai univalentvoters. This result disapproved hypothesis 3b. Hypothesis 4prediction was supported. There was no significant contrast betweenthe univalent and opposing voters and ambivalent and opposing voters.The predictions of hypothesis 5a were upheld by the results sincethere was a significant effect on voter ambivalence of TPP, when Ma`scoverage was positive. Also, hypothesis 5b was approved by the resultbut hypothesis 6a was partially supported, but hypothesis 6b wasrejected by the outcome(Chang, 2014).

Critiqueof the Study

Theresults presented are from one particular location, thus it cannot betaken to be true for every other place in the world. Media influenceon voters depends on its status, for example, how much weight peopleattach on media. Where uphold media coverage with ultimate trust,much influence is felt. The study presented a comparison of only twoelection candidates. The outcome may not be the same where there arethree or more close contenders.


Chan,C., Wei, R.,&nbsp&ampLo, V. (2014). Ambivalent Versus Univalent Voters: Perceived MediaInfluences and Third-Person Perceptions. MediaPsychology,14(4). 420-450.