Download this article
Don’t miss our trial consultnt responses at the end of this article: Tara Trask, Charlotte A. Morris, and read the authors’ reply here.
Background
Most people assume they have insight into their goals and motivations, believing they know who they are, what they do, and why they do it. This fundamental assumption – that we are aware of our skills, talents, and even our knowledge – is the basis upon which we navigate our lives. We make decisions every day that rest on a belief that we know our abilities. Students choose one major over another because they believe they might be unable to handle certain prerequisites; teachers believe they can predict their student evaluations at the end of the year based on their experience in doing so; jurors believe they can be impartial during a case; and lawyers believe they can sway a jury or judge with their persuasive abilities.
While it is indeed the case that people can be accurate in their self-perceptions, much research suggests this is not always the case; we often do not have accurate self-perceptions. While people often believe they can predict how they performed on an exam or how they will be evaluated by their superiors, or even their students, research often shows that these beliefs are only modestly related to actual performance in those domains (Mabe & West, 1982; Zell & Krizan, 2013). In some instances, this error in self-perception is only an annoyance for the perceiver, such as a student who believes they are so naturally gifted in an area that they fail to spend enough time studying and thus perform poorly on an exam. It can also, however, be devastating, as in a case in which a medical doctor performs a risky procedure they believed they could do and yet were not skilled enough to complete.
In this paper, we explore a domain of self-knowledge as of yet relatively unexamined and one for which we believe most people would be certain that they have accurate self-knowledge – political orientation. Political orientation is a particularly important self-aspect for many people, particularly in today’s partisan culture. Our relationships, where we live and work, the news we watch, and how we see the world more generally are influenced in part by our self-perceived political orientation. Given that such orientations have been shown to influence a host of behaviors, including even how we process information, it should be particularly important that we be accurate in our self-perceptions. Yet we believe that such a standard of accuracy is not necessarily being met. We begin by outlining prior research on self-knowledge generally and then focus on our predictions for political self-knowledge.
So, are we really that bad?
As stated previously, there are many studies showing that people’s perceptions of their performance is often only weakly related to their actual performance outcomes. For example, people’s views of their own intelligence are only slightly correlated with their performance on academic tests and intelligence assessments (Hansford & Hattie, 1982) and student’s self-assessments of performance in the classroom are often only moderately related to the grades they receive from instructors (Chemers, Hu, & Garcia, 2001). Research showed that surgical residents performance on a standardized test assessing surgical knowledge was completely unrelated to their belief in their knowledge (Risucci, Torolani, & Ward, 1989). In another vein, people often believe they are very good at detecting lies, and yet there is virtually no correlation with such a belief and their actual ability to detect lies (DePaulo, Charlton, Cooper, Lindsay, & Muhlenbruck, 1997). Peers are often better able to predict how long our romantic relationships will last than we are (MacDonald & Ross, 1999). CIA analysts believe their predictions about world events are better than they are (Cambridge & Shrekengost, 1980), and students who performed in the bottom 25% of a class on an exam left the exam believing they outperformed their peers (Dunning, Johnson, Ehrlingr, & Kruger, 2003).
Why are we so bad at this?
One may wonder why it is that we are so bad at accurately knowing ourselves, especially given how seemingly important it is to do so. Part of our poor insight may be due to a lack of enough information against which to judge ourselves – people simply do not know that which they do not know (Dunning, Heath, & Suls, 2004). Imagine a teacher entering the class for the first time and then trying to evaluate how he or she performed at the end of the class; they may not have accurate insight into the cues that would help them assess their performance, and what is worse, they may instead rely on other, less valuable cues. As a result, this lack of knowledge prevents their accurate self-assessment (Caputo & Dunning, 2005). In another example, when asking someone to generate as many words as he or she could using the word “television,” (e.g., vision, not, tons), a person may list 20 and believe this to be a reasonable number, even though there are more than 350 potential words that could be created. Knowing the total number of possible words would certainly help a person more accurately assess their performance, but not knowing the correct amount makes the task difficult at best.
People are also often motivated to see themselves in a positive light and overestimate their abilities, which can lead to these self-perception inaccuracies (e.g., Guenther & Alicke, 2010). Individuals who believe, for example, that they are more likely than their peers to live past 80 and less likely to have a heart attack (Weinstein, 1980), show an optimistic bias, as do the 60% of students in one study who rated themselves as being in the top 10% of students in their ability to get along with others (College Board, 1976-1977). People also expect they will finish tasks more quickly than they actually do, and often fail to make their deadlines even when they predict a completion time they are “certain” they will meet (Buehler, Griffin, & Ross, 2002). These are all examples of how motivation to see oneself positively can also play a role in shaping our self-perceptions.
Political self-knowledge
Given the spate of research in recent years documenting people’s inaccuracy at self-perception, we wanted to know whether this extended to people’s perceptions of how liberal or conservative they are. Are people who identify as conservative really as conservative as they believe and, in turn, are liberals as liberal as they attest to be? We believe there is quite a bit of evidence supporting our assertion that people may not have insight into such identifications. To begin, rather than desiring less distribution of wealth, Americans favor greater distribution than is currently observed in our society (Norton & Ariely, 2011); greater wealth distribution is a far more liberal perspective than a conservative one, even though the majority of the country identifies along the conservative spectrum (Gallup, 2012). This is particularly interesting in light of the fact that the majority of Americans support more liberal policy issues concerning topics such as poverty, environmental regulation, and education practices (Free & Cantrill, 1967; Stimson, 2004). On such issues as immigration, gay marriage, and marijuana legalization, Americans seem to be taking more liberal stances (Plaue, 2012) all the while identifying more as conservatives (Florida, 2011).
So why might people identify as being more conservative while supporting more liberal issues? Well, conservatives tend to have a greater focus on loyalty to their groups (e.g., family, country, religion) than do liberals (Graham, Haidt, & Nosek, 2009). This is consistent with evidence suggesting people who identify as conservatives value obedience to authority and group loyalty more so than do liberals (who show greater concern for harm and fairness; see Haidt & Graham, 2007 for review). It is possible that this sense of loyalty leads individuals to identify more strongly with conservatism than their support of issues may actually reveal. Along a similar vein, conservatism is associated with self-enhancement insofar as research has shown individuals rating themselves as conservative (as compared to liberals) had more distorted, overly favorable self-perceptions (Jost, Liviatan, van der Toorn, Ledgerwood, Mandisodza, & Nosek, 2010). This tendency toward self-enhancement, along with their desire for group loyalty, may lead conservatives to perceive themselves as typical or “true” members of the Republican Party when their attitudes on specific issues may not reflect that. Thus, it is possible that individuals may identify as strongly conservative when they may indeed only be moderately or even weakly conservative.
The current work
We hypothesized that individuals would underestimate how liberal they in fact are. In each of three studies (using different populations and including slightly different measures), participants were asked to rate the strength of their own liberal or conservative orientations. They were then given a more objective measure of their political attitudes via a set of 12 items concerning various political issues, which could then be compared to norms for the population at large. Based on the previous literature described above, we predicted that, generally, people’s political attitudes would be more liberal than their self-assessment of their political orientation, but that this would be particularly true among those who self-identified as politically conservative.
Study 1
We had one-hundred and ninety-nine college students (138 Female; 55% Caucasian, 25% African American; Average Age of 20.34 years) complete a survey for partial course credit. Participants first indicated their political views on a scale including the following categories (1=Liberal Democrat, 2=Average Democrat, 3=Moderate Democrat, 4=Independent, 5=Moderate Republican, 6=Average Republican, 7=Conservative Republican; see Table 1 for frequencies of each category across all three studies).
Participants next completed a political attitudes quiz developed by the Pew Research Center (PBS NewsHour, 2012). Participants responded to 12 attitude statements about issues in American politics that are strongly tied to the political spectrum including topics such as gay rights, abortion, welfare, government regulation of business and the environment, and others. Respondents had to rate their agreement with each statement on a “1 Strongly Disagree” to “4 Strongly Agree” scale. Example items include, “Poor people have become too dependent on government assistance programs,” “Business corporations make too much profit,” and “There need to be stricter laws and regulations to protect the environment.”
Researchers at the Pew Research Center (2012) gave this questionnaire to a representative sample of Americans and developed a coding scheme whereby respondents could be placed into one of the same seven political categories mentioned above. The coding scheme estimates the degree to which participant’s responses match those who are typical of each political group and places them into a category of best fit (for additional details, see Zell & Bernstein, in press). Following the 12 items, participants answered some demographic questions, were thanked, and debriefed.
Study 1 Results
Having made ratings of their political attitudes, we could use the coding scheme developed by the Pew Research Center and then make a more objective assignment to one of the seven categories of political orientation. We could then compare the objective measure with the self-assessment each participant made on the political orientation scale. Thus, each participant had a category to which they assigned themselves as well as a category to which they were assigned via the Pew Research Center scale.
While self-assessment scores and the objective measures did correlate with each other (p<.001), our hypothesis was supported; the objective political orientation scores were significantly less conservative than were the self-assessment scores ( p<.001). As we predicted, as individuals self-identified more with conservatism, the bias to underestimate liberalism increased, p<.001 (see Figure 1). When looking at the political orientations individually, Liberal Democrats significantly overestimated their liberalism (p=.02) and Average Democrats had relatively unbiased self-perceptions (p=.64). However, Moderate Democrats (p=.001), Independents (p<.001), and Republicans (p<.001) significantly underestimated their liberalism.
Figure 1.
Self-perception and objective scores as a function of between self-perceptions and objective scores as a function of political identity (Study 1).
Study 2
We wished to replicate our findings from Study 1 using a different population. In this study, we collected data using Mechanical Turk (e.g., Bernstein & Benfield, 2013; see for review Buhrmester, Kwang, & Gosling, 2011). We collected 360 respondents (233 Female, 74% Caucasian, Average Age = 28.46 years). Participants were all US residents and 44 states were represented in the survey.
The procedure for Study 2 was identical to Study 1 except we counterbalanced the order of the self-assessment survey and the attitude quiz to ensure that the order of the surveys did not affect the results. We also measured education and income to determine if these influenced our results.
Study 2 Results
Neither the order of the surveys nor people’s sex, race, age, income, education level, or region of residence influenced the results. We again replicated our finding from Study 1, such that self-assessed political orientation was more conservative than were the objective assessments (p<.001), though the two measures were correlated (p<.001). Also as in Study 1, the regression analysis (coding remained the same as in Study 1) revealed that the more conservative respondents showed a greater underestimation of their liberalism (i.e., they rated themselves as more conservative than their attitudes suggest) than did those who identified as more liberal, p<.001 (see Figure 2).
Figure 2.
Self-perception and objective scores as a function of between self-perceptions and objective scores as a function of political identity (Study 2).
Study 3
While Studies 1 and 2 were supportive of our hypothesis, they are not without their limitations. First, our sample was largely skewed towards more Democratic respondents. While this is not uncommon for young people to be more affiliated with the Democratic party, we need to show whether our effect occurs among each Moderate, Average, and Conservative Republican groups and thus Study 3 aimed to sample more equally among the political spectrum. Study 3 also examined whether this bias among self-perceived political orientation affected voting behavior in the 2012 Presidential Election.
College student samples from two universities (one in the Southeast and one in the Mid-Atlantic) were collected in January of 2013. We used a screening procedure to ensure we got equal sample sizes with respect to political categories. One hundred and fifty-four participants (110 Female, 66% Caucasian, 22% African American, Average Age = 20.31 years) were collected in total with 22 participants in each of the seven political categories. The procedure was identical to Study 1 with the exception that respondents were also asked to report on whom they voted for in the 2012 Presidential Election (Barack Obama, Mitt Romney, another candidate, or did not vote).
Study 3 Results
We again examined whether self-rated political orientation differed from the more objective measure, and as in the two previous studies, we found the same significant effect, p<.001. As shown previously, self-assessed political ratings were more conservative than the objective measure, although the two were again correlated (p<.001). Further, a regression showed the same pattern as in the prior two studies, namely that this effect was strongest among more conservative individuals, p<.001 (see Figure 3).
Figure 3.
Self-perception and objective scores as a function of between self-perceptions and objective scores as a function of political identity (Study 3).
We also wanted to see if our self-assessed measure of political orientation and our more objective measure could predict voting behavior. We performed a logistic regression for each President Obama and for Governor Romney and found that, in each case, the self-reported scale as well as the objective measure both acted as significant predictors of voting behavior. Interestingly, self-assessed reports of political orientation still predicted who a person voted for even after controlling for the more objective measure, and this was true whether people voted for Obama or for Romney.
Discussion
In three studies, we showed that young adults tend to think they are more conservative than they really are in terms of their support for important political issues and that this was particularly true for young adults who self-identify along the conservative side of the political spectrum. We think it is important to note that there was a reasonably strong correlation between self-assessed political orientation and our more objective measure, indicating some degree of accuracy. While it may be convenient to say that, because of this correlation (especially compared to some of the examples provided in the introduction), this bias is not important, we believe that would be premature.
In the political domain, these results have important implications. Much work shows that self-assessed political orientation predicts voting behavior and support for Presidential candidates who share the same political orientation (Jost, 2006). Nonetheless, if individuals are more liberal than they believe they are in terms of self-identification, it is possible that they could end up voting for candidates who do not actually represent their political views in regards to particular issues. Further, in states in which primaries are closed (i.e., individuals may only vote in the primary for which they are registered as being a party member), individuals may be relegated to voting for primary candidates who also do not represent their personal stance on issues.
In terms of voting, the relevance of this attitude was clear. When examining the results of Study 3, we found that no one who identified themselves as a Democrat was actually more closely aligned with Republicans based on their objective scores. However, 19 individuals who identified themselves as a Republican were actually Democrats based on the objective measure. Of these, six voted for Obama, five for Romney, one for another candidate, while seven did not vote at all. Thus, some people who voted for Romney had a liberal identification according to the objective measure. This was unlike those who voted for Obama, of whom none would be identified as conservative based on their attitudes on issues (the objective measure). This suggests that Republicans who misclassify themselves may vote for a Republican candidate even if their attitudes better align with the Democratic Party, yet the reverse is not necessarily the case.
In terms of the law, this has equally important implications. Beyond adding to the abundance of evidence showing that individuals often do not have accurate perceptions of themselves, this research suggests the importance of asking questions concerning particular issues rather than simply asking ones’ political identification when questioning witnesses and potential jurors. Along these lines, the results of Study 3 showed that objective political orientation scores predicted voting after accounting for self-ratings of political identity, and were as important as self-ratings in predicting voting. Therefore, researchers and legal professionals who only utilize self-ratings of political orientation may be neglecting a key source of data that predicts behavior and potentially jury decision-making.
Another important aspect of this is that asking people (e.g., jurors) to identify which political party they support may not reveal their views on specific political issues. It is commonly assumed that people within the Democratic and Republican parties hold views on issues like immigration, gay rights, and abortion, and that these views are highly polarized. When someone explicitly identifies themselves as a member of a political party, we may stereotype them according to their party and assume they hold views that are consistent with our stereotypes of the party. The present findings suggest that these political stereotypes may be somewhat inaccurate, and that using them to infer other people’s attitudes may be counterproductive in some instances.
More generally, because political conservatism predicts many important behaviors and is correlated with a number of other important personality factors (e.g., prejudice, see Terrizzi, Shook, & Ventis, 2010; social dominance orientation, see Pratto, Felicia, Sidanius, Stallworth, & Malle, 1994; disgust sensitivity, see Inbar, Pizarro, & Bloom, 2008), it may be worthwhile to assess political orientation in terms of issues in addition to self-reported identification.
Self-knowledge is something we all assume we have and yet, its accuracy is often called into question. Findings ways to assess our knowledge while avoiding the biases in our thinking is of the utmost importance when trying to understand peoples’ motivations and behaviors. This quest for greater accuracy is of value for both basic psychological research as well as in the realm of more applied domains.
Michael J. Bernstein, PhD is an Assistant Professor of Psychology at Penn State University in Abington, PA. His primary areas of research are in interpersonal and intergroup relations, examining face memory and processing, social exclusion (e.g., bullying), judgment and decision making, stereotyping, prejudice, and discrimination, explicit and implicit attitudes, and the application of Social Psychology to jury decision making and consumer behavior. You can review his research and contact information here.
Ethan Zell, PhD is an Assistant Professor of psychology at the University of North Carolina at Greensboro. His research examines self-evaluation of ability; accuracy of self and social perception; and the effects of age, culture, and gender on the self. More information can be obtained from his website.
References
Bernstein, M. J., & Benfield, J. A. (2013). Past Perspective is related to Present relationships: Past-Positive and negative time Perspectives differentially predict rejection sensitivity. The Psychological Record, 63, 615-628.
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high quality, data? Perspectives on Psychological Science, 6, 3–5. doi:10. 1177/1745691610393980
Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the ‘‘planning fallacy’’: Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67, 366–381.
Cambridge, R.M., & Shreckengost, R.C. (1980). Are you sure? The subjective probability assessment test. Unpublished manuscript, Office of Training, Central Intelligence Agency, Langley, VA.
Caputo, D., & Dunning, D. (2005). What you don’t know: The role played by errors of omission in imperfect self-assessments. Journal of Experimental Social Psychology, 41, 488–505. doi:10. 1016/j.jesp.2004.09.006
Chemers, M.M., Hu, L., & Garcia, B.F. (2001). Academic self-efficacy and first-year college student performance and adjustment. Journal of Educational Psychology, 93, 55–64.
College Board. (1976–1977). Student descriptive questionnaire. Princeton, NJ: Educational Testing Service.
DePaulo, B.M., Charlton, K., Cooper, H., Lindsay, J.J., & Muhlenbruck, L. (1997). The accuracy-confidence correlation in the detection of deception. Personality and Social Psychology Review, 1, 346–357.
Dunning, D., Heath, C., & Suls, J. M. (2004). Flawed self-assessment: Implications for health, education, and the workplace. Psychological Science in the Public Interest, 5, 69–106. doi:10.1111/j.1529- 1006.2004.00018
Dunning, D., Johnson, K., Ehrlinger, J., & Kruger, J. (2003). Why people fail to recognize their own incompetence. Current Directions in Psychological Science, 12, 83–86.
Florida, R. (2011, March 29). The conservative states of America. Retrieved from http://www.theatlantic.com/politics/archive/2011/ 03/the-conservative-states-of-america/71827/
Free, L. A., & Cantril, H. (1967). The political beliefs of Americans. New Brunswick, NJ: Rutgers University Press.
Gallup. (2012, May 12). In presidential elections, age is a factor only among whites. Retrieved from http://www.gallup.com/poll/ 154712/presidential-election-age-factor-among-whites.aspx
Graham, J., Haidt, J., & Nosek, B. A. (2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96, 1029–1046. doi:10.1037/ a0015141
Guenther, C. L., & Alicke, M. D. (2010). Deconstructing the better- than-average effect. Journal of Personality and Social Psychology, 99, 755–770. doi:10.1037/a0020959
Haidt, J., & Graham, J. (2007). When morality opposes justice: Conservatives have moral intuitions that liberals may not recognize. Social Justice Research, 20, 98–116, doi:10.1007/s11211-007-0034-z
Hansford, B.C., & Hattie, J.A. (1982). The relationship between self and achievement/performance measures. Review of Educational Research, 52, 123–142.
Inbar, Y., Pizarro, D. A., & Bloom, P. (2009). Conservatives are more easily disgusted than liberals. Cognition and Emotion, 23, 714-725.
Jost, J. T. (2006). The end of the end of ideology. American Psychologist, 61, 651–670. doi:10.1037/0003-066X.61.7.651
Jost, J. T., Liviatan, I., van der Toorn, J., Ledgerwood, A., Mandi- sodza, A., & Nosek, B. A. (2010). System justification: How do we know it’s motivated? In D. Bobocel, A. C. Kay, M. P. Zanna, & J. M. Olson (Eds.), The psychology of justice and legitimacy (pp. 173–203). New York, NY: Psychology Press.
Mabe, P. A., & West, S. G. (1982). Validity of self-evaluation of abilty: A review and meta-analysis. Journal of Applied Psychology, 67, 280–296. doi:0021 -9010/82/6703-0280
MacDonald, T.K., & Ross, M. (1999). Assessing the accuracy of predictions about dating relationships: How and why do lovers’ predictions differ from those made by observers? Personality and Social Psychology Bulletin, 25, 1417–1429.
Norton, M. I., & Ariely, D. (2011). Building a better America: One wealth quintile at a time. Perspectives on Psychological Science, 6, 9–12. doi:10.1177/1745691610393524
PBS Newshour. (2012). Where do you fit? The political party test. Retrieved from http://www.pbs.org/newshour/vote2012/quiz/
Pew Research Center. (2012, August 24). Political party quiz: How we placed you. Retrieved from http://www.people-press.org/2012/08/ 24/political-party-quiz-how-we-placed-you/
Plaue, N. (2012, July 10). Americans are becoming more liberal on some issues, more conservative on others. Retrieved from http:// www.businessinsider.com/liberal-and-conservative-trends-2012-7
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741.
Risucci, D.A., Torolani, A.J., & Ward, R.J. (1989). Ratings of surgical residents by self, supervisors and peers. Surgical Gynecology and Obstetrics, 169, 519–526.
Stimson, J. A. (2004). Tides of consent. New York, NY: Cambridge University Press.
Terrizzi Jr, J. A., Shook, N. J., & Ventis, W. L. (2010). Disgust: A predictor of social conservatism and prejudicial attitudes toward homosexuals. Personality and Individual Differences, 49, 587-592.
Weinstein, N.D. (1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 58, 806–820.
Zell, E., & Bernstein, M. J. (in press). You may think you’re right…Young adults are more
liberal than they realize. Social Psychological and Personality Science.
Zell, E., & Krizan, Z. (2013). Do people have insight into their abilities? A meta-synthesis. Revised manuscript under review.
Distribution of Self-Perceived and Actual Political Orientation
| Study 1 | Self-Perception | Objective Score | Difference |
|---|---|---|---|
| Liberal Democrat | 13.1 | 39.2 | -26.1 |
| Average Democrat | 25.6 | 12.6 | 13.0 |
| Moderate Democrat | 11.6 | 17.6 | -6.0 |
| Independent | 28.1 | 14.6 | 13.5 |
| Moderate Republican | 12.6 | 10.1 | 2.5 |
| Average Republican | 6.0 | 4.0 | 2.0 |
| Conservative Republican | 3.0 | 2.0 | 1.0 |
| Study 2 | |||
| Liberal Democrat | 27.8 | 63.3 | -35.5 |
| Average Democrat | 15.6 | 10.8 | 4.8 |
| Moderate Democrat | 15.8 | 7.8 | 8.0 |
| Independent | 28.3 | 8.1 | 20.2 |
| Moderate Republican | 7.8 | 3.9 | 3.9 |
| Average Republican | 3.6 | 3.6 | 0.0 |
| Conservative Republican | 1.1 | 2.5 | -1.4 |
| Study 3 | |||
| Liberal Democrat | 14.3 | 30.5 | -16.2 |
| Average Democrat | 14.3 | 13.0 | 1.3 |
| Moderate Democrat | 14.3 | 14.9 | -0.6 |
| Independent | 14.3 | 13.6 | 0.7 |
| Moderate Republican | 14.3 | 11.7 | 2.6 |
| Average Republican | 14.3 | 6.5 | 7.8 |
| Conservative Republican | 14.3 | 9.7 | 4.6 |
We asked two trial consultants to respond to this article.
Tara Trask is President and Founder of Tara Trask and Associates, a full service Trial Consulting, Jury Research and Litigation Strategy firm with offices in San Francisco and Dallas. She focuses her work on intellectual property litigation, antitrust, securities, products liability and other complex commercial litigation.
Tara responds:
Authors Bernstein and Zell seek to address the accuracy of self-reporting with regard to political attitudes and party affiliation. The authors hypothesize that people are generally poor at self-reporting their levels of conservatismor liberalism. The authors further hypothesize that self-described conservatives, in particular tend to define themselves as more conservative than they are when compared to objectiveassessment.
This research is particularly applicable for those of us who focus on juries. When making determinations about peremptory strikes in the cases I work on, I am often looking for any clues I can gather as to the values and beliefs of the prospective juror I am observing. With regard to whether a juror is more likely to identify with a patent holder, a company making a product, or one side or another of a contract dispute, that juror’s values are of the utmost importance to me.
What this research sheds light on is the fact that knowing political party alone, and making decisions based on that metric alone will likely result in mistakes in jury selection. I never make decisions about a juror based on one characteristic. But this research lends depth and texture to something I was already considering and aware of. I often tell attorneys that I make decisions based on the “constellation of characteristics” I am able to glean about a person. This research informs that perspective.
For example, to assume that someone who self-identifies as a Republican will be unwilling to award damages in a products case would be a faulty assumption based on stereotypes according to this research. I would suggest that other metrics are equally important to consider in the “constellation” of the juror. What do they do for a living? How do they describe their parenting style? What groups or organizations do they belong to? And of course, how are they dressed? Are they carrying a particular book or periodical? All these questions and others combine to assist me in determining whether the story my client will be putting forward will resonate with that particular juror.
On a broad note, I would say that this research underscores something those of us in this field have long known; reliance on over generalization and stereotypes is a recipe for disaster. And, more importantly, while some of the decisions that are made by trial consultants every day may look like they are simple and stereotype-driven, most are not. This research underscores that point.
Charli Morris has 20 years of trial consulting experience and holds a Master’s degree in Litigation Science from The University of Kansas. She is co-author of The Persuasive Edge and can be reached directly at [email protected].
Charli responds:
Bernstein and Zell reach two related conclusions that are consistent with my own experience:
1) “…asking people to identify which political party they support may not reveal their views on specific political issues.”
2) “…this research suggests the importance of asking questions concerning particular issues rather than just asking ones’ political identification.”
Stereotypes have always been rooted in generalities and jury research has always been interested in moving beyond the superficial to find specific attitudes and beliefs that will be meaningful given the facts and law that apply to a case (or case type).
Labels are Loaded
The problem with labels of any kind is that we don’t control their meaning. Words like Latino, Black and Asian are technically nothing more than demographic indicators of race, but consider the wide variety of impressions (indeed prejudices) that are generated by the words alone. Even gender comes with baggage. “Man up.” “You throw like a girl.” Indeed, the ultimate name-calling for males starts with the letter P and is a word used to describe female anatomy.
We add “right-wing” to “Republican” to make it an insult, and curse the “left-wing Liberal” media for being sympathetic to the President when he’s a Democrat. Members of both parties are characterized as “extremist” or “radical” when their political beliefs are strongly held. Our political system has consistently denied third-party candidates even when “Libertarian” and “Independent” sound about as non-threatening as they can get. I dare say many of us are still not entirely sure what it means when people tell us they are members of the Tea Party.
No wonder we are so bad at knowing (or accepting) what the labels mean, even as we apply them to ourselves.
It’s been a long while since I asked focus group participants or potential jurors to tell me their political party affiliation. I’ve gotten the feeling that the question is regarded as intrusive by judges and jurors alike, despite the fact that voter registration is a matter of public record. If I did so, given today’s political climate my list would certainly include more than the two major political parties and an ill-defined third (Republican, Democrat or Independent). But to the extent that we may still see value in asking it, I recommend asking follow-up questions to find out how politically active they are as a measure of how strongly they identify with a particular party.
(This is similar to differences I have observed between identification with a particular religious group [e.g., Catholic] compared to how often someone attends services or practices his or her faith [i.e., religiosity].)
Q: On a scale of 1 to 10, how active are you in support of the political party you selected? (circle one)
Not at all Extremely
1 2 3 4 5 6 7 8 9 10
Q: How often do you donate time or money to the party you support (check one):
____Yearly
____Quarterly
____Monthly
____Rarely
____Never
Q: Describe the ways you participate in your political party’s efforts or activities:
Follow-up questions like these can be a direct and effective way to find out – in a manner of speaking – if a person puts his money where his mouth is. And presumably, the more active a person is in his affiliation, the more likely he is to be influenced by the values and messages of a particular group. In the case of politics, the follow-up questions may also improve the accuracy of a person’s self-assessment. Once a person realizes that his time or money are not actually spent on the support of a political party, he or she could go back and change the initial answer on the Bernstein/Zell scale from Conservative or Liberal to Moderate. In fact, consider what might happen to the results if the “follow-up” questions were asked before the party affiliation question?
Moving from General to Specific
Despite the problems with self-assessment and self-report, when it’s time to design pre-trial research and Supplemental Juror Questionnaires (SJQs) I do feel compelled to include some version of the question about political self-perception because I believe it can tell us something about prospective jurors’ views on issues that are central to litigation. Bernstein and Zell do remind us, “political conservatism predicts many important behaviors and is correlated with a number of other personality factors.”
For example, when working with plaintiff’s counsel I may reject someone who I believe to be too conservative on liability even if he seems moderately liberal on questions regarding damages. To defense counsel, I may recommend a strike when a person seems too liberal on damages even if I believe she may be moderately conservative on liability or causation. And, of course, it is the combination of these beliefs within a panel that can determine our use of individual strikes.
I routinely use a 10-point scaled-response item to measure whether a person considers him- or herself as Conservative or Liberal. I prefer a numbered scale that does not attach any additional labels as Bernstein and Zell do: I see no meaningful difference between their choice of the words “Average” and “Moderate” (as modifiers to Democrat and Republican).
I would also like to know more about whether research has clearly shown that Independents are situated equidistant between Republicans and Democrats as shown in the Bernstein/Zell scale or whether, in fact, Independents consider themselves outside the two-party system as do Libertarian and Tea Party members.
On the last SJQ I developed I was uncomfortable letting the Liberal and Conservative labels stand alone, so I added the words “politically or socially” to the question. I think this is precisely what the researchers have demonstrated: that self-reported party affiliation is belied by our views on issues that are both social and political. In fact, some hotly-contested political issues of today were once thought of as strictly social, moral, religious or personal issues (e.g., reproductive rights and gay marriage).
One prospective juror (out of 81) answered our SJQ this way:
His answer confirms my hunch and from now on I will include at least two questions using the same Liberal/Conservative scale: one for political and one for social.
Goldilocks Gets it Right
When attorneys try their hand at drafting SJQs or voir dire, I sometimes see the attitudinal questions become too specific, as in too case-specific. If we go too far, there is a risk that a judge will reject our questionnaires (or sustain objections to voir dire) because the questions seek commitments or require jurors to prejudge evidence. Jurors and mock jurors also have a tendency to retreat when they believe they are being asked to make up their minds about an issue that is clearly related to evidence in the case before they hear the evidence.
Consider the difference between the following questions for a case against a company responsible for the design, testing and manufacture of safety devices used in electrical line work:
The federal government has too much influence on the way machine guards and protective devices are designed and manufactured to protect against known hazards and foreseeable risks of harm and danger in the workplace.
or
Our government has a legitimate role in establishing safety standards in the workplace to protect against known hazards and foreseeable risks of harm.
In the first, potential jurors would likely need specific experience with “machine guards and protective devices” in order to have a firmly held belief that could make a difference in the case. In the second (better) question, a potential juror needs only work experience of any kind to have developed a belief about the role government plays in establishing safety standards. Our questions designed to uncover liberal or conservative bias must likewise be just right not too general and not too specific.
Final Thoughts about the Studies
The most serious limitation of the research in its application to litigation is the age of participants in two of the three studies. When clients and I are reviewing data from pre-trial research and SJQs, I regularly caution against putting too much stock in the political preferences of people under the age of 25 (e.g., college students), particularly those who are still closely tied to their parents (e.g., financially dependent, living at home, etc.). Just as medical research has demonstrated that the human brain is not yet fully developed in young adults, I would argue our political beliefs are more fully developed and stabilize over time as we age and live more independently.
The business of measuring attitudes and beliefs that are relevant to jury decision-making is both art and social science. Experienced trial consultants can help attorneys apply the principles of research provided in studies like these conducted by Bernstein and Zell to the specific facts and law of a case.
Bernstein and Zell Reply
There are of course limitations to this work that are important avenues for future study. First, an astute reader may wonder whether asking people their political affiliation and then having them report on their ideologies could in fact change people’s initial attitude about their political orientation; in other words, would a participant who states they identify as conservative have a change in their own identification after realizing that they seem to support relatively liberal views? While we believe that revealing a person’s inconsistent views with their self can indeed change attitudes about the self, our data from the second study suggests this is not occurring here; in Study 2, we varied the order of the orientation question and the ideology questions and found no effect. If the order was important, than we should have expected participant’s self-described political orientation to be more liberal following their responses to their ideology questions. We did not find that, however.
There are also valid concerns about using a single item measure to assess political orientation. The scale, for example, does not differentiate between social conservatism and fiscal conservatism and it is not clear whether independent is truly in the “middle” of liberal and conservatism or if it is in fact an orthogonal category. These concerns are not only valid but, in part, support one of our primary claims about the value of our research findings. The single item scale we used is so common in part because it does predict behavior (e.g., voting behavior, policy support). Thus, even a single item scale that does not differentiate between more nuanced understandings of political orientations still predicts well. Our argument, however, is that not only is it not perfect, but it also underestimates liberalism. We wholeheartedly believe that if those in the legal profession only had access to the single item “self-description” measure, that it would be better to add additional questions that assessed other aspects of orientation (e.g., social vs. fiscal conservatism; how important is one’s political orientation to their sense of self; how often they are involved/donate money to their political groups). Future research would do well to examine how these and other additional questions relate to the political ideology scale we incorporated in this research.
Finally, while Study 2 used a non-college sample and revealed the same results as Studies 1 and 3, those latter studies did use only college sample students. There is reason to be cautious about the political beliefs of younger adults simply because they may be strongly based on their level of dependence with their parents. However, that seems more of an issue of consistency over time rather than accuracy or predictive value; for example, age of participant did not differentially predict the effect we found, indicating that young adults and older adults in our sample showed the same bias. Further, while the political orientations of young adults may be more greatly influenced by parents and guardians than the political orientations of older adults, that does not mean they are less important in predicting behavior. For example, a 20-year-old man or woman sitting in a jury pool may strongly identify as conservative now, precisely because his or her parents identify as conservative. While that orientation may chance over time, that does not negate the impact that such identification has now. Nonetheless, future research should continue to examine this effect among older and more diverse populations.