Partisan bias in opinion formation on episodes of political controversy : evidence from Great Britain.

TL;DR: This article found that partisan predispositions and tolerance of political misconduct are both important in shaping voter opinions and that partisanship has the strongest influence among the more knowledgeable and interested voters on political controversies.

Abstract: Voters form judgements about political controversies through a process of motivated reasoning driven by two goals: the desire to reach an objectively accurate conclusion (accuracy) and the desire to reach a conclusion congruent with pre-existing views (direction). The impact of directional goals may depend on political sophistication. We test our hypotheses using data from a 2011 British survey that measured voters' opinions on three specific real-life political controversies. We use voters' underlying tolerance of political misconduct as an indicator of accuracy goals and party identification as a measure of directional goals. We find that partisan predispositions and tolerance of political misconduct are both important in shaping voter opinions and that partisanship has the strongest influence among the more knowledgeable and interested voters. These findings further our understanding of how voters react to political controversies and how they process new political information.

Summary

Introduction

• Controversies about the behaviour of a politician are common in any democracy.
• But while media focus on political controversy undoubtedly puts instances of political misconduct on the public agenda, the consequences of such an episode depend on the opinions that voters form about just how wrong and punishable a politician’s behaviour was.
• The two goals highlighted by political psychologists are accuracy and directional goals (Lodge and Taber 2000).

The role of political sophistication

• Because voters are heterogeneous, the influence of partisan affect on opinions about scandals may not be the same for all individuals.
• On the one hand, there are arguments that the least sophisticated would be the most biased.
• Such voters might therefore also be more inclined towards using heuristics in determining their political opinions.
• Partisan bias affects decisions more among people with higher levels of political knowledge because they are also more likely to recognize the partisan aspects of an issue in the first place and are able to consciously judge it in relation to their pre-existing partisan loyalties (Zaller, 1992: 121).
• The theory and evidence on the moderating influence of political sophistication is therefore mixed and inconclusive.

Data

• Sanders et al. (2007) provide evidence for the reliability of such internet-based surveys.
• This shows that opinions on episodes of political controversy are not knee-jerk responses: voters do distinguish between the politicians involved.
• In their sample, 79 per cent said Laws should have resigned, 62 per cent said the same for Chris Huhne, and only 34 per cent believed that Ken Clarke should step down.
• The authors then create a composite indicator based on all answers to these questions; the Cronbach’s alpha for the scale is a satisfactory 0.73.
• The authors therefore decided to measure political knowledge by providing respondents with a list of six politicians and asked which of these were currently in the UK Cabinet; only three of the politicians actually were.

Results

• The authors begin their discussion of the results by examining voter opinions on the politician’s actions.
• The authors then consider views on whether the politician should resign before examining the moderating influence of political sophistication.

Voter opinions on a politician’s actions

• The authors first results concern the influence of partisanship and general attitudes on right and wrong on how wrong voters think a politician’s behaviour was.
• First, Models 2, 4 and 6 show that their control variables do not always have the same effects.
• The effect of the ‘no identification’ indicator is estimated at between 0.47 and 0.68, with all p-values below 0.1.
• The authors also present the expected change in the wrongness scale for various shifts in party identification, either from identifying with the politician’s party or from identifying with no party.
• In sum, the effect of partisanship on specific opinions about episodes of political controversy is strong and consistent, as is the effect of general intolerance towards political misconduct.

Resignation

• When it comes to political controversies, the debate is just as much about the appropriate consequences of the particular wrongdoing as about the gravity of the case itself.
• To answer this question, the authors consider the estimated impact of partisanship and general intolerance of political misconduct on right and wrong on whether a respondent thinks a politician should resign.
• Again, the authors show the results from two sets of models: first with only the key independent variables and then controlling for the same variables in Models 2, 4 and 6.
• All predicted probabilities are calculated using the SPost package in Stata (Long and Freese, 2005).
• Accuracy and directional goals both clearly influence calls for resignation, and do so strongly.

Political sophistication

• Finally, the authors hypothesised that political sophistication may moderate the impact of partisan identification on opinions on episodes of political controversy.
• The authors present results graphically following the recommendations of Brambor et al. (2006).
• Figure 4 shows how the effect of party identification depends on a voter’s level of political sophistication.
• The results are substantively similar for the Huhne and Clarke controversies.
• When comparing Conservative to Liberal Democrat identifiers in the cases of Huhne and Laws, there is a stronger effect of partisanship as sophistication increases only for the Huhne episode.

Conclusion

• This research has suggested that accuracy and directional goals are both important to understanding how voters respond to reports of misbehaviour by politicians.
• The authors overall findings can be interpreted in two ways.
• In addition, existing scholarship argues that the influence and importance of party identification in Britain is similar to that in other European countries and lower than in the United States (Shiveley 1979; Westholm and Niemi 1992; Dalton 2008; Milazzo et al. 2012).
• This also appears to have an important influence on how partisanship influences opinion formation.

Figures

Figure 4 Effect of party identification conditional on political sophistication

Figure 3 Predicted change in the predicted probability of saying the politician should resign

Table 1 Laws, Huhne, Clarke: Wrongness (OLS regression models)

Table 2 Laws, Huhne, Clarke: Resignation (Binary logistic regression models)

Figure 2 Predicted change in perceived wrongness of a politician’s actions

Full text

Durham Research Online
Deposited in DRO:
30 November 2015
Version of attached le:
Accepted Version
Peer-review status of attached le:
Peer-reviewed
Citation for published item:
Wagner, Markus and Tarlov, Jessie and Vivyan, Nick (2014) 'Partisan bias in opinion formation on episodes of
political controversy : evidence from Great Britain.', Political studies., 62 (1). pp. 136-158.
Further information on publisher's website:
http://dx.doi.org/10.1111/j.1467-9248.2012.01002.x
Publisher's copyright statement:
This is the accepted version of the following article: Wagner, M., Tarlov, J. and Vivyan, N. (2014), Partisan Bias in
Opinion Formation on Episodes of Political Controversy: Evidence from Great Britain. Political Studies, 62(1):
136-158, which has been published in nal form at http://dx.doi.org/10.1111/j.1467-9248.2012.01002.x. This article
may be used for non-commercial purposes in accordance With Wiley Terms and Conditions for self-archiving.
Additional information:
Use policy
The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for
personal research or study, educational, or not-for-prot purposes provided that:
•
a full bibliographic reference is made to the original source
•
a link is made to the metadata record in DRO
•
the full-text is not changed in any way
The full-text must not be sold in any format or medium without the formal permission of the copyright holders.
Please consult the full DRO policy for further details.
Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom
Tel : +44 (0)191 334 3042 | Fax : +44 (0)191 334 2971
https://dro.dur.ac.uk

1
Partisan bias in opinion formation on episodes of political controversy: evidence from
Great Britain
Abstract
Voters form judgments about political controversies through a process of motivated reasoning
driven by two goals: the desire to reach an objectively accurate conclusion (accuracy) and the
desire to reach a conclusion congruent with pre-existing views (direction). The impact of
directional goals may depend on political sophistication. We test our hypotheses using data
from a 2011 British survey that measured voters’ opinions on three specific real-life political
controversies. We use voters’ underlying tolerance of political misconduct as an indicator of
accuracy goals and party identification as a measure of directional goals. We find that partisan
predispositions and tolerance of political misconduct are both important in shaping voter
opinions and that partisanship has the strongest influence among the more knowledgeable and
interested voters. These findings further our understanding of how voters react to political
controversies and how they process new political information.

2
Introduction
Controversies about the behaviour of a politician are common in any democracy. Such
controversies are often characterised by intense media attention and an incremental
uncovering of the facts of the case. Whether politicians can survive a political controversy
depends on a variety of factors. For example, the so-called ‘Alistair Campbell rule’, named
after British Prime Minister Tony Blair’s former director of communications, states that ‘no
Minister can survive beyond a two-week feeding frenzy in the press’.
1
But while media focus
on political controversy undoubtedly puts instances of political misconduct on the public
agenda, the consequences of such an episode depend on the opinions that voters form about
just how wrong and punishable a politician’s behaviour was. For one, nowadays almost
instant polls ask voters for their opinion regarding the events, and the results of these polls can
influence both media coverage and politicians’ strategies. Moreover, political controversy can
also have electoral consequences, for example when voters punish misbehaving politicians at
subsequent elections (Ahuja, et al. 1994, Banducci and Karp, 1994, Clarke, et al. 1999,
Dimock and Jacobsen 1995, Farrell et al. 1998, Pattie and Johnston forthcoming, Peters and
Welch 1980, Vivyan et al. forthcoming, Welch and Hibbing 1997).
Voters form their opinions on the moral acceptability and appropriate consequences of
a politician’s actions in the same way they form opinions about other political events: through
a process of motivated reasoning (Kunda 1990; Lodge and Taber 2000). According to this
view, voters form opinions by incorporating and evaluating new incoming information, but
they do so in ways that are guided by specific goals. The two goals highlighted by political
psychologists are accuracy and directional goals (Lodge and Taber 2000). Accuracy goals
exist due to the need to reach a correct, truthful conclusion. Directional goals exist due to the
need to reach a specific conclusion, so one that is in accordance with prior opinions and
1
This is the version of the Alistair Campbell rule cited by Polly Toynbee in the Guardian, 16
December 2004 (http://www.guardian.co.uk/politics/2004/dec/16/schools.davidblunkett).

3
predispositions. These two goals are always present, though one of the two is of often stronger
than the other (Lodge and Taber 2000).
In this paper, we assess the influence of these two motivations in opinion formation on
political controversy. We do so by considering the effect of two voter attitudes: (1) a voter’s
underlying tolerance of political misconduct and (2) his or her partisanship. If a voter’s
tolerance of political misconduct in general predicts reactions to political controversy, then
voter opinions are shaped by accuracy goals: their view on the specific episode is consistent
with the toughness of their underlying normative stance towards political misbehaviour.
In contrast, if partisanship predicts reactions to the episode, then a voter’s opinion is
shaped by directional goals. It is well-known that partisanship, also known as party
identification or party affect, has an important impact on how voters react to and interpret
political events (e.g. Campbell et al. 1960; Zaller 1992; Bartels 2002; Johnston 2006). This
has been elaborated on in more recent research on motivated reasoning and selective
processing (Lodge and Taber 2000), also in relation to political scandals (Fischle 2000). In the
case of political controversy, we would expect partisanship to exert a directional effect on
voter reasoning: if the politician is from the party the voter is attached to, then the voter’s
judgment may be more lenient, whereas if the politician is from a rival party, then the voter’s
judgment may be tougher (Chang and Kerr 2009).
Assessing whether voter opinions are shaped more by general normative standards or
by pre-existing sympathies is important for two reasons. First, from a normative point of view
we want to know what the sources of public disapproval of politician’s behaviour are. Ideally,
voters would base their reactions not on partisan or other types of affect, but on careful
consideration of the facts about the politician’s conduct and by benchmarking this conduct
against their personal moral or ethical expectations of politicians. Second, knowing the extent
to which the two goals shape opinions may help us understand the course of political
controversy. If public reactions depend on the extent to which generally held ethical norms

4
were violated, the politicians should be more likely to face calls for resignation depending on
how normatively wrong their actions were. However, if public reactions depend largely on
affect, then the popularity of the politician’s party (and his or her own popularity) will
become important.
Finally, we also consider whether the influence of directional goals depends on a
voter’s political awareness and knowledge (Shani 2006; Blais et al. 2010). On the one hand, it
may be that low levels of political sophistication mean that the reliance on partisanship as a
guide may be stronger; on the other hand, it may be precisely the more sophisticated voters
who have the motivation and ability to ‘refuse to internalise messages that they recognize as
inconsistent with their underlying predispositions’ (Zaller 1992: 121).
We carry out our empirical analysis using data from a specially conducted nationally
representative internet survey conducted in the UK in May 2011 that measured voter reactions
to three recent or on-going controversial episodes involving prominent politicians: the
Conservative Ken Clarke and the Liberal Democrats Chris Huhne and David Laws. Our
research design has a number of advantages for investigating voter reaction to political
controversy. Even though the three episodes of political controversy that we study are
different in nature, they all were marked by widespread public disapproval of the event and
ran the risk that the revelations would damage the participant’s reputation. They are thus
comparable episodes and fit Thompson’s (2000:13-14) admittedly broad definition of a
political scandal. Secondly, we use actual episodes involving real-life politicians as opposed
to hypothetical scenarios, which should increase the likelihood that responses will reflect
genuine feelings about the event as opposed to reactions to an ‘imagined’ scenario. We
therefore do not have to temper our findings with consideration of the impact of hypothetical
scenarios on voter response. Finally, each episode of political controversy took place very
near to the time of the survey or was currently on-going when the survey was fielded. This

Frequently asked questions

Q1. What are the contributions mentioned in the paper "Partisan bias in opinion formation on episodes of political controversy: evidence from great britain abstract voters form judgments about political controversies through a process of motivated reasoning" ?

The authors use voters ’ underlying tolerance of political misconduct as an indicator of accuracy goals and party identification as a measure of directional goals. The authors find that partisan predispositions and tolerance of political misconduct are both important in shaping voter opinions and that partisanship has the strongest influence among the more knowledgeable and interested voters. These findings further their understanding of how voters react to political controversies and how they process new political information. 

Q2. What are the future works mentioned in the paper "Partisan bias in opinion formation on episodes of political controversy: evidence from great britain abstract voters form judgments about political controversies through a process of motivated reasoning" ?

Future research could investigate further how partisanship influences how voters react to new information, especially in more complex situations characterised by multiparty systems and coalition governments. Overall, their findings have highlighted both the relevance of general normative standards and the impact of partisanship on voter reactions to political controversy, and future work should consider when, how and why the influence of partisanship can vary.