PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Share a file Manage my documents Convert Recover PDF Search Help Contact

23044920 .pdf

Original filename: 23044920.pdf
Title: Managing Monikers: The Role of Name Presentation in the 2008 Presidential Election

This PDF 1.4 document has been generated by / iText® 5.5.10 ©2000-2015 iText Group NV (Ithaka Harbors Inc.; licensed version); modified using iText® 5.5.10 ©2000-2015 iText Group NV (Ithaka Harbors Inc.; licensed version), and has been sent on pdf-archive.com on 17/01/2017 at 21:32, from IP address 47.17.x.x. The current document download page has been viewed 636 times.
File size: 2 MB (19 pages).
Privacy: public file

Download original PDF file

Document preview

Managing Monikers: The Role of Name Presentation in the 2008 Presidential Election
Source: Presidential Studies Quarterly, Vol. 40, No. 3, The 2008 Presidential Election, Part
II (September 2010), pp. 464-481
Published by: Wiley on behalf of the Center for the Study of the Presidency and
Stable URL: http://www.jstor.org/stable/23044920
Accessed: 17-01-2017 20:31 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

Center for the Study of the Presidency and Congress, Wiley are collaborating with JSTOR to
digitize, preserve and extend access to Presidential Studies Quarterly

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms

Managing Monikers: The Role of Name
Presentation in the 2008 Presidential Election

University of Wisconsin-La Crosse

Florida State University

Given America's widespread contempt for Islamic extremists, Obama's Muslim-s

moniker could have cost him electoral support. Consequently, anecdotal evidence sugge

Obama played a "name game" in which he deflected suspicions about his religious ba

by avoiding use of his middle name (Hussein) and minimizing the frequency with w

opponents used it. Did the presentation of Obama's name affect how voters evalua

Results from a web-based experiment suggest that the answer varies by political orien

Among Republicans and conservatives, Obama's favorability ratings are generally lo

his middle name is present. Name presentation had little effect on Democrats and libe
moderates and independents rated the president more favorably when his middle name

Regardless of party identification or political ideology, name presentation had no effe
probability of voting for Obama.

Monikers matter. People consider them when choosing mates (Fos

1998; Scheuble, Klingemann, and Johnson 2000), evaluating studen

Seraydarian 1977, 1978, 1979; Figlio 2004), purchasing products (Bloch
1983; Heibing and Cooper 2003; Simon and Sullivan 1993), having child
and Scheuble 2002; Mehrabian and Piercy 1993), hiring and compensat
(Bertrand and Mullainathan 2004; Mason 2004; but see Levitt and Frye
ing political leaders (Box-Steffensmeier et al. 2003; Goldenberg and Tr
and so on. Names can reveal, among other things, a person's cultural he
ality, and economic standing (Brennen 2000; Lawson 1996; Mehrabi
2001). In this sense, a name can be a blessing or a burden, for one's li

Ray Block Jr., is an assistant professor of American politics at the University of Wisconsin-La Cr

American politics, political participation, public opinion, and minority group politics.

Chinonye Onwunli is a graduate student in the political science department at Florida State
research interests include public policy and minority group politics.

AUTHORS' NOTE: We thank Jason Barabas, Bill Berry, Jens Grosser, Christina Haynes,

Jennifer Jerit, Patrick Mason, Cherie Maestas, Will Moore, Mark Souva, the anonymous reviewers,
the Spring 2008 Honors Seminar on Racial Attitudes for their helpful comments.

Presidential Studies Quarterly 40, no. 3 (September) 464
© 2010 Center for the Study of the Presidency

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms

Block and Onwunli / MANAGING MONIKERS


partly determined by the background clues conveyed by it. As a result, people use
pseudonyms to manage the information their monikers impart to lovers, educators,
consumers, employers, and constituents.1

For example, the importance of "moniker management" was evident during the
2008 presidential campaign. Specifically, Barack Obama's staff sought to minimize the
likelihood of citizens associating him with Islamic extremists by diverting attention away

from his Muslim-sounding middle name (Hussein), while his opponents, aware of
Americans' current antipathy toward Muslims (Tapper 2007), used this moniker in an
attempt to weaken his electoral support. Although the strategic use of public image is

widely documented in the study of presidential politics (Nimmo and Savage 1976; for
reviews, see Hacker 1995, 2004), to our knowledge, there are no empirical studies on the
effect of name presentation on voters' support for Obama.

We address this research gap in the pages that follow. Drawing insights from
studies of self-presentation (sometimes called image or impression management) and
political advertising, and using an experiment embedded within the September 2008
wave of the Cooperative Campaign Analysis Project (CCAP), we investigate whether
subjects appraise Obama differently depending on whether his full name (Barack Hussein

Obama) is present, or only his first and last name. We argue that the impact of name

presentation varies according to a voter's preexisting beliefs about politics. These pre
dispositions constitute a person's "political orientation," and we use party identification

and political ideology as crude but reliable proxies for this orientation.

We find that Republicans and political conservatives (who are generally the least
favorable toward the president) tend to lower their feeling thermometer ratings of Obama
when his middle name is present. However, name presentation does not increase or decrease

the probalility that these subjects will prefer Obama to McCain. Democrats and liberals,

while typically the most loyal of Obama supporters, are relatively unresponsive to the
experiment: their candidate evaluations tend not to change, regardless of how the president's

name appears. Consistent with past research (particularly Ansolabehere and Iyengar 1995),
presenting Obama's middle name causes independents and moderates—those who are most
likely to see the name game as "dirty politics"—to rate the president more favorably than
they otherwise would. The differential impact of the experiment suggests that the name
game succeeded in reinforcing the anti-Obama sentiments of rightward-leaning subjects,
but it failed to convert subjects on the left, and it backfired among subjects with a more

centrist orientation to politics. Despite its influence on candidate perceptions, name
presentation had no effect whatsoever on a subject's likelihood of voting for Obama.

Obama Muslim Rumor

A "name game" took place during the 2008 presidential campaign. The offensiv

players in this name game were the party leaders, intellectuals, analysts, ideologues

1. Aliases have always been popular with "elites"—particularly entertainers, writers, and activists
(see Griffin 1999, 2003)—and the proliferation of Internet networks has stimulated the use of pseudonyms

among the "rank and file" (Ludlow 1996; Rosen 2000).

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms



activists, opinion leaders, journalists, and prognosticators who worked on behalf the
McCain-Palin ticket. Conversely, the defensive players were the Obama-Biden campaign
and its surrogates. Spectators were the potential voters whom players sought to persuade
(and ultimately mobilize), and the political news and "infotainment" media provided the

arena. The offensive goal of the name game was to spread what Nyhan et al. (2009) call
the "Obama [is a] Muslim rumor," and the defenders sought to keep this from happening.

The primary rule of this game was that players could advance or block the Obama
Muslim rumor by controlling the frequency with which voters were exposed to "name
plays" (presentations of the then—Illinois senator's middle moniker).
On its face, there is nothing malicious about referring to Obama by his full, given

name (in fact, Rush Limbaugh echoed this point throughout the campaign; see, e.g.,
Limbaugh 2008). However, underlying this seemingly harmless act is a deliberately
strategic motive: pointing out his middle name made Obama seem "different," and—
considering that many Americans associate practitioners of Islam with the terrorist
attacks of September 11, 2001—dangerous. Of course, some view the name game as a
veiled (or perhaps not so subtle) racial appeal. For example, Swift and Richman (2009)
would argue that opponents dislike Obama because he is biracial, not because he might
follow the teachings of the Holy Qur'an, and former President Jimmy Carter makes a
similar remark during an interview with NBC's Brian Williams on September 16, 2009-2
Conclusions are mixed regarding whether racial biases continue to influence support for

candidates of color (see Berenski et al. 2009), so the name game need not necessarily
reflect racial intent (or animus). Obama's opponents, to borrow Mendelberg's (2001)
phrasing, may have been pulling the "race card" when they made name plays. However,
the name game draws more on voters' concerns about the War on Terror than on their
deep-seated beliefs about race relations, and one can assume that the defense would have

been just as tempted to make name plays if Obama were a "white candidate."
Admittedly, the offense has an advantage in that the Obama Muslim rumor is easier
to stoke than it is to stop—mainly because Obama's first and last names are "ethnic," and

scholars have demonstrated that distinctively ethnic names tend to inspire negative

perceptions (see Lawson 1996; Mehrabian 2001). Nevertheless, few will deny that
"Barack Hussein Obama" sounds more Muslim than "Barack Obama" does, and the more
Islamic voters believe Obama is, the greater the chance they will view him unfavorably

and, consequently, the less likely voters are to support Obama.3 In this name game,
perceptions trump reality: Obama embraces the Christian faith, but the truth of his
religious background is immaterial; the mere belief that Obama is a Muslim could have
2. In fact, the special issue of the 2007 Journal of Black Studies, the October 2008 PS: Political Science
and Politics symposium on the 2008 election, and the spring 2009 edition of the Du Bois Review are devoted
to reflecting upon the racial implications of Obama's candidacy and resulting presidency.

3. Obama acknowledged the negative imagery that his moniker could conjure up in his 1995 and 2006
memoirs. He even joked about it on October 16,2008, at the 63rd annual Alfred E. Smith Memorial Foundation
Dinner in New York when he admitted that he had received his middle name "from someone who obviously
didn't think [he] would ever run for president." Later that evening, Obama quipped that his name was actually

Barack Steve Obama (for a transcript of Obama's dinner speech, see Kurtzman 2008). The then-senator's

light-hearted tone masks what had been an ongoing concern. While campaigning for her husband in Canton,
Ohio, Michelle Obama recalls past attempts to use innuendos about her husband's religious heritage to play on
the fears of voters (a practice she describes as "dropping the fear bomb"; see Silva 2008).

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms

Block and Onwunli / MANAGING MONIKERS


spoiled the former senator's chances of winning if it had spread throughout the electorate.

Clearly, the stakes in this name game were high, for the candidate who established and

maintained the most positive public image would win the nation's top political office.
Two of the more visible offensive players were radio personality Bill Cunningham

and self-described polemicist Ann Coulter. For example, Cunningham used Obama's ful

name repeatedly at a John McCain rally in Cincinnati, Ohio (much to McCain's displea
sure), and on the January 27 airing of Fox News's Hannity's America when discussing the

results of the Iowa caucuses and New Hampshire primary. Intriguingly, Cunningham
often referred to the then-senator as "Barack Mohammed Hussein Obama," even though

"Mohammed" is not part of Obama's name. Similarly, in an interview with Neil Cavuto
on the February 13, 2009, edition of Fox's Your World, Coulter admitted that she finds it

amusing to address the president as "B. Hussein Obama" or "President Hussein." The
defenders fought hard to minimize these and related instances, but it was the offensive

team that ultimately delayed the name game. In response to a news release featuring

controversial image of Obama wearing traditional Kenyan attire, former Republican

National Committee chair Mike Duncan formally denounced the use of Obama's middl
moniker in February 2008. The fact that Duncan "called foul" lent further credence

to the idea that name plays had become taboo (Thornburgh 2008). However, despite
the chairman's denouncement, the use of Obama's full name continued throughout
the campaign, mainly on local radio and online discussion forums (see Media Matters for
America 2008 [http://mediamatters.org] for a detailed list of such incidents).4

The Obama name game is but one example of a broader social process in which
people seek to manage the images (or impressions) that others have of them. The

examination of the dynamics of self-presentation has a rich tradition in the field of social

psychology (see Goffman 1959; Nimmo and Savage 1976), and from this foundation
grew the field of impression management.5 Borrowing Leary's (1995, 41) definition,

impressions are subjective evaluations of presidential candidates, for instance, and it i

useful to gauge the strength of impressions by how "favorable" or "supportive" voters are
in their evaluations. In this particular case, the defensive (offensive) players in the name
game were trying to convey the impression that Obama is a Christian (Muslim). Research

on impression management suggests that the candidates can control—and manipulate-—
public perceptions directly (e.g., by emphasizing or suppressing the idea that Obama i

an Islamic extremist) or indirectly, by implying or denying that Obama's professional and

personal acquaintances are extremists (Cialdini and Richardson 1980; Richardson and
Cialdini 1981; Schlenker 1980, 2003). The Obama name game is an example of the latter
technique, for an easily recognizable name can garner widespread support or opposition

4. To be fair, the name game was not the only image challenge that Obama faced. For example, there
were also attempts to portray him as inexperienced (e.g., Obama is a community organizer), Marxist (e.g
Obama is a secret socialist), elitist (e.g., Obama eats arugula—a presumably exotic and relatively pricey leafy
green vegetable), militant (e.g., Obama shares the sectarian views of Louis Farrakhan or Reverend Jeremiah
Wright), arrogant (e.g., Obama is a mega-celebrity like Britney Spears or Paris Hilton), ineffectual (e.g.
Obama is all talk and no action), or, worse yet, diabolical (Obama is the Antichrist).

5. Goffman (1959) extends metaphors of theatrical performance in his seminal treatment of impres
sion management, and the fruits of Goffman's intuitions appear in business (see Giacalone and Rosenfeld
1989; Morrison and Bies 1991) and social science (e.g., Leary 1995; Leary and Kowalski 1990) research.

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms



depending on whether voters associate candidates with famous or infamous political
figures (Nimmo 2001, 42). Unfortunately for Obama, including "Hussein" in his already
suspicious-sounding moniker can prompt voters to liken the candidate to militants such

as Osama bin Laden or Saddam Hussein (see Bombardieri 2007). How successful were the

campaigners at playing the name game? We argue that name plays should affect some
citizens differently than others.

Voter Support, Name Presentation, and Political Orientation:
Our Hypotheses
Polls (e.g., Gallup and Newport 2008) show that conservatives and Republicans
prefer McCain to Obama. Unlike their independently partisan and politically moderate
colleagues—from whom Obama enjoyed considerable popularity but whose votes he was
not guaranteed (Todd and Gawiser 2009, 111)—liberals and Democrats were particularly

loyal to Obama (Healy 2009 describes this devotion as bordering on cultish). These
findings are not surprising, and they should hold true no matter how Obama's name
appears. Therefore, regardless of name presentation, support for Obama will differ by
political orientation (HO: it will be strongest among "left-leaning" voters (i.e., liberals
and Democrats), weaker among "centrists" (i.e., moderates and independents), and
weakest among voters on the "right" (e.g., conservatives and Republicans).
In addition to its "overall effect," we also expect political orientation to moderate
the impact of name presentation on Obama support (H2). The name game happened on
a variety of fronts; it appeared in print, on television, and across the World Wide Web.
For present purposes, it is useful to think of the name game as an advertisement campaign

in which the players used the media to "sell" ideas to voters about Obama's religion.6
Political advertisements vary in numerous ways, but studies usually distinguish positive

ads from negative ones (e.g., Goldstein and Freedman 2002; Gronbeck 1985; Hill 1989;
Johnson-Cartee and Copeland 1991; Lau and Sigelman 2000). Attempts to link Obama
with Islamic terrorists can represent "anti-Obama ads," and efforts to prevent such
linkages can signify "counter advertisements."7 There is a wealth of evidence showing
partisan and ideological differences in how voters respond to negative campaigning (for

reviews of these findings, see Benoit 2004; Lau et al. 1999), so we, too, expect group
differences in the effect of name presentation on voters' impressions of Obama. If
6. While there is an obvious link between the Obama name game and racial priming (see Hutchings
and Jardina 2009), we view the advertising and priming literatures as complementing one another (although
the latter is clearly the larger of the two lines of research). Intuitions from the priming literature are useful
for explaining why voters on the left and right respond differently to the name game. In a general sense,

"confirmation bias" suggests that predispositions affect how readily subjects will accept certain primes.
However, as you will see, the research on negative campaign advertisement offers specific guidelines with
respect to our expectations for the effect of name presentation on voters in the center.

7. Perceptions of what constitutes a "negative" advertisement can vary widely across citizens (see
Geer 2006; Sigelman and Kugler 2003). Nevertheless, even if voters do not find the name game particularly
uncivil, it is safe to assume that most Americans view figures such as Saddam Hussein and Osama bin Laden
in a negative light. In other words, it is less important if the use of Obama's middle name is considered a "bad
thing" so long as there is a agreement among voters that Obama's name reminds voters of "bad people."

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms

Block and Onwunli / MANAGING MONIKERS

emphasizing the middle name is comparable to attacking Obama, and if partisanship and
ideology shape a voter's sensitivity to attack advertisements, then the impact of the name

game should vary by political orientation as well.

Emphasizing the middle moniker may weaken Obama's support among voters on
the right—that is, those who describe themselves as politically conservative and/or are

members of the GOP. Intuitions from the research on priming (see, e.g., McNamara
2005) provide the rationale for this claim: attempts to associate Obama with Islam would
reinforce the suspicions of right-wing voters, and this would result in those voters viewing

Obama less favorably. Although suspicions about Obama's religion persist throughout
the electorate (Burke 2008; Nyhan et al. 2009; Salmon 2009), conservatives and Repub
licans are more likely get political information from media sources that advance the
viewpoint that Obama practices Islam (Carnevale 2008). Not only are voters on the right
inclined to believe the Obama Muslim rumor, they are also less likely to reject messages
that propagate this rumor, for people tend to accept information (e.g., a name cue) that

confirms their beliefs (for a review of research on "confirmation bias," see Nickerson
1998). Therefore, name presentation should have the intended effect on conservatives and

Republicans: the more "Muslim" they believe Obama's name sounds, the less supportive
of the Illinois senator they might be (H2a).
Presenting Obama's middle name can affect voters on the left and in the center in

several potential ways. For example, attempts to portray Obama as a Muslim may have

little impact on liberals, Democrats, moderates, and independents. This is mainly
because anti-Obama attacks come from the offensive players of the name game, and
centrist and leftward-leaning voters (particularly the latter) are, by definition, more
inclined to share the viewpoints of the defense. Recall that studies of confirmation bias

conclude that citizens tend to accept without question information that comports with
their current beliefs. The inverse is also true: people tend to resist contradicting ideas.

Despite the many inconclusive findings in the political advertisement literature (Lau

et al. 1999; Sigelman and Kugler 2003), one firmly established conclusion is that
negative campaigning fails to convert voters from their preexisting preferences (see Ansola

behere and Iyengar 1995, 81; Ansolabehere et al. 1994). As noted, Obama's popularity
is appreciably higher among non—"right wing" voters. Campaign tricks, particularly
those using Obama's middle name, should not change these voters' impression of the
then-senator; therefore, name plays may have little influence on how much voters on the
left and center support the candidate (H2b).
Conversely, the name game could have the opposite impact: it could backfire in the sense

that seeing Obama's middle name might strengthen a person's support (H2C). In fact, Benen

(2008) worried that anti-Obama attacks, while useful for mobilizing "the base"—strong

partisans and those with extreme ideological views—could push centrist and leftward
leaning voters away from McCain, the perceived "sponsor" of the attack, and toward Obama.

If Benen's concern is legitimate (and polling results suggest that it is; see Cooper and Thee

2008), then negative campaigning may reaffirm a voter's cynicism with the combative

nature of campaign politics (Fineman and Hosenball 1996; Lau and Pomper 2001, 2004),
and such cynicism can motivate voters who grew tired of mudslinging to throw their support

behind Obama. The potential for name plays "backfiring" is especially strong for voters in

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms




the center, for Ansolabehere and Iyengar (1995, 81, 192) find that negative campaigning
can dissuade noncommitted voters. Either out of ambivalence over which electoral camp to
choose, or out of disdain for the adversarial system that promotes taking sides, centrist voters

(compared to their more strongly affiliated counterparts) tend not to identify themselves

with traditional partisan or ideological labels (Dennis 1988a; 1988b; Geer 2006; Weisberg
1980). Because they are not entrenched in party or ideology politics, citizens in the center
are not only more likely to view the name game as a scheme to exploit voters' anxieties, but

also less likely to endorse (or reward) anti-Obama attacks.

To summarize, we anticipate that name presentation and political orientation
influence voters' support for Obama. Generally, a voter's support for Obama will depend
on his or her political orientation: left-leaning voters should think most highly of Obama,

voters on the right should give the least support to the candidate, and centrists should

give evaluations that fall between these two extremes (Hi). Furthermore, the effect of
name presentation on Obama support should vary by party and ideology (H2): Voters on

the right (who are the least agreeable to the former Illinois senator) are most likely to
penalize Obama when his middle name appears (H2a). Name plays may have no influence
whatsoever on left-leaning and centerline constituents (H2b), whose candidate fidelity can

immunize them to anti-Obama attacks. Presenting Obama's middle name to liberals,
moderates, Democrats, and independents could even backfire, provoking a backlash
against negative campaigning that ultimately benefits Obama (H2c). We evaluate these
claims next.

Experiment Design, Procedures, and Measures
In all, 789 voting-age nonstudent subjects participated in the Name Presentation
Study, an Internet-based experiment. We added this experiment to the 2008 Cooperative

Campaign Analysis Project (CCAP), a six-wave panel survey of 20,000 registered voters
that started in December 2007 and ended after the election in November 2008. CCAP

is a collaborative project run by Simon Jackman (Stanford University) and Lynn Vavreck

(University of California, Los Angeles) that allows schools to purchase time on a multi
wave, online probability survey about social and political attitudes (for details, see
http://www.polimetrix.com/news/ccap.html). Professors in Florida State University's
Economics, Political Science, and Sociology departments combined resources to include
experiments in the summer 2008 on various waves of the CCAP. The following analyses
utilize the September (9/17-9/29) 2008 wave.8
The design of the Name Presentation Study is simple. Subjects received two survey
questions gauging their "support" for Obama. The first question is a feeling thermometer

in which "favorability" scores range from 0 to 100, with higher scores representing
8. The timing of this survey experiment makes it an ideal test of the name game hypotheses. During
the earlier months of the general election, support for Obama and McCain varied greatly, making conclusions
drawn from polling data necessarily tentative. It is unlikely that subjects' views of the candidates (which had
been crystallizing over the general election) would change drastically in the weeks leading up to election day
(Blais 2004; Gelman and King 1993); therefore, results from later studies are more likely to represent the
sincere viewpoints of potential voters.

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms

Block and Onwunli / MANAGING MONIKERS


evaluations that are more favorable toward the candidate.9 The feeling thermometer varia

has a mean of 52.94 and a standard deviation of 37.80, which suggests that evaluations o

Obama tended to be relatively favorable but widely dispersed at this point in the campaig

The election had not occurred at the time of the experiment, so the second indicator of

candidate support is a dichotomous item measuring a subject's intention to vote for Oba

(mean = .5 2, standard deviation = .49). We code this "intended vote choice" variable so t

subjects receive a score of 1 if they claim that they would vote for the Obama-Biden tick

if the election were held today; subjects who intend to vote for McCain-Palin get a scor

0.10 The candidate favorability and intended vote choice variables represent different asp

of support: the latter reflects subjects' preference for Obama, while the former gauges t
strength of a subject's preference.

Name presentation is a central determinant of candidate support, for we belie
that the inclusion of Obama's middle name makes him seem more "Muslim." Accord

ingly, we added an experimental stimulus within the September 2008 wave of the CCAP

that allowed us to randomly assign half of the subjects to receive presentations of the
candidate's full name, Barack Hussein Obama, or BHO (n = 415); the remaining subjects
(« = 374) only see his first and last name when they receive the feeling thermometer and
vote choice questions. We record the results of this "name presentation experiment" in a

dichotomous variable, which we coded so that subjects who see Obama's middle name
get a score of 1 and those who do not received a score of 0.
We also expect political orientation to moderate the effect of name presentation on

Obama support, and we use political ideology and party identification as indicators of
orientation. Accordingly, we include dummy variables representing various categories of

a subject's self-reported political ideology (the categories of interest are "liberal" for
subjects on the left [n = 213], "moderate" for those in the center [n = 247], and "con
servative" for subjects on the right [n = 256]).11 Likewise, self-identified "Democrats"

(« = 3l6), "independents" (« = 205), and "Republicans" (« = 221) represent subjects
with left-leaning, centrist, and rightward political orientations, respectively.12

9. This question has the following wording: "We would like to get your feelings about the candidate
using a 'feeling thermometer.' As indicated by the scale below, ratings between 0 and 49 degrees mean that
you do not feel favorably toward the candidate and that you feel cold towards that candidate. Ratings between

51 and 100 degrees mean that you feel favorably and warm toward the candidate. If you do not feel

particularly warm or cold toward a candidate, you would rate them at 50 degrees. If you do not know enough

about a candidate to answer, please write 'don't know' in the space provided." The candidates for which
feeling thermometer ratings are available are John McCain and Barack Obama/Barack Hussein Obama;
however, the analyses that follow utilize only the Obama thermometers because name presentation had
no statistically significant effect on how subjects evaluated McCain. For all variables, "don't know" and
nonusable responses are treated as missing values.
10. Here is the exact wording of this variable: "If the Presidential election were held today and the
candidates were Barack Obama/Barack Hussein Obama and Joe Biden, the Democrats and John McCain and
Sarah Palin, the Republicans, and you have to choose, for whom would you vote?"
11. The original question was a five-point scale that asked subjects to rate the "level" of their ideology:

"Below is a scale on which the political views that people might hold are arranged from 'strongly conser
vative' to 'strongly liberal.' Please indicate where you would place yourself on this scale." We collapse the
"strongly liberal (conservative) and liberal (conservative) categories to create a three-category measure of a
subject's "type" of ideology.
12. The wording for the party identification question is: "Generally speaking, do you consider yourself

a Democrat, an independent, a Republican, or something else (please specify)?"

This content downloaded from on Tue, 17 Jan 2017 20:31:57 UTC
All use subject to http://about.jstor.org/terms

Related documents

national extemp championcraft
nader spolier

Related keywords