American Association for Public Opinion Research

The Nature of Political Ideology in the Contemporary Electorate

Author(s): Shawn Treier and D. Sunshine Hillygus
Source: The Public Opinion Quarterly, Vol. 73, No. 4 (Winter, 2009), pp. 679-703
Published by: Oxford University Press on behalf of the American Association for Public
Opinion Research
Stable URL: https://www.jstor.org/stable/40467637
Accessed: 26-07-2018 17:03 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
https://about.jstor.org/terms

American Association for Public Opinion Research, Oxford University Press are
collaborating with JSTOR to digitize, preserve and extend access to The Public Opinion
Quarterly

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
Public Opinion Quarterly, Vol. 73, No. 4, Winter 2009, pp. 679-703

THE NATURE OF POLITICAL IDEOLOGY IN THE

CONTEMPORARY ELECTORATE

SHAWN TREIER
D. SUNSHINE HILLYGUS

Abstract Given the increasingly polarized nature of American poli-

tics, renewed attention has been focused on the ideological nature of the
mass public. Using Bayesian Item Response Theory (IRT), we examine
the contemporary contours of policy attitudes as they relate to ideo-
logical identity and we consider the implications for the way scholars
conceptualize, measure, and use political ideology in empirical research.
Although political rhetoric today is clearly organized by a single ideolog-
ical dimension, we find that the belief systems of the mass public remain
multidimensional, with many in the electorate holding liberal preferences
on one dimension and conservative preferences on another. These cross-
pressured individuals tend to self-identify as moderate (or say "Don't
Know") in response to the standard liberal-conservative scale, thereby
jeopardizing the validity of this commonly used measure. Our analysis
further shows that failing to account for the multidimensional nature
of ideological preferences can produce inaccurate predictions about the
voting behavior of the American public.

There appears to be a consensus among scholars and political observers that

U.S. political elites have grown more polarized in recent decades. Democrats
and Republicans in Congress more consistently oppose each other on legis-
lation (McCarty, Poole, and Rosenthal 2006), the party platforms are more
ideologically extreme (Layman 1999), and issue activists are more committed
to one political party or the other (Stone 1991). In contemporary American

Shawn treier is with the University of Minnesota, 1414 Social Sciences Building, 267 19th Ave S,
Minneapolis, MN 55455, USA. d. sunshine hillygus is with Duke University, 409 Perkins Library,
Box 90204, Durham, NC 27708-0204, USA. The authors would like to thank Chris Federico, Paul
Goren, Dean Lacy, Andrew Martin, Caroline Tolbert, Claudine Gay, Dan Carpenter, Kevin Quinn,
Eric Schickler, the participants of the University of Minnesota American Politics Pro-seminar,
three anonymous reviewers, and editor Jamie Druckman for helpful feedback and suggestions.
Address correspondence to D. Sunshine Hillygus; e-mail: hillygus@duke.edu.

doi: 10. 1093/poq/nfp067 Advance Access publication December 4, 2009

© The Author 2009. Published by Oxford University Press on behalf of the American Association for Public O
All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 680 Treier and Hillygus

politics, Republican politicians consi

an issue while Democratic politician
across different policy domains. With
we can quite accurately predict a p
parate as taxes, health care, or aborti
political elites in the United States t
of ideology.
Can the policy preferences of the American public be similarly characterized?
Although this question has been the subject of considerable research over the
years (e.g., Marcus, Tabb, and Sullivan 1974; Conover and Feldman 1984;
Jacoby 1 99 1 ), it takes on a new prominence given the recent polarization debate.
Some scholars have argued that the sharpening of policy differences between
political elites in recent decades has increased ideological identification and
polarization in the public as well (Abramowitz and Saunders 1998). In contrast,
others argue that the majority of the public remain moderate on most policy
issues even as elected representatives have grown further apart (Dimaggio,
Evans, and Bryson 1996; Fiorina 2004). Morris Fiorina concludes that the
great mass of American people "are for the most part moderate in their views
and tolerant in their manner. . . it is not voters who have polarized, but the
candidates they are asked to choose between" (Fiorina 2004, pp. 8, 49).
Given the increasing salience of political ideology in American politics,
it seems important to examine how ideology is conceptualized by the public
relative to how it is operationalized and measured by researchers. Both sides of
the polarization debate seem to assume that the ideological labels people use are
a meaningful representation of their public policy preferences - an assumption
once challenged by early public opinion research that concluded the public
was incapable of ideological thinking (Converse 1964). The generalizations
that scholars make about the behavior, attitudes, or thinking of the American
electorate could be wholly inaccurate if the liberal-conservative continuum so
often used in empirical analysis is an inadequate measure of policy preferences.
In this article, we examine the contemporary contours of policy attitudes as
they relate to ideological identity. And we consider the implications for the
way scholars conceptualize, measure, and use political ideology in theories
and models of political behavior. Although political rhetoric today is clearly
organized by a single ideological dimension, we find that the belief systems
of the mass public are multidimensional. Using Bayesian Item Response The-
ory (IRT), a methodological approach with unique advantages over previous
estimates of ideological preferences, we show that many in the electorate hold
liberal preferences on one dimension and conservative preferences on another.
These cross-pressured individuals tend to self-identify as moderate (or say
"Don't Know") in response to the standard liberal-conservative scale, raising
questions about the validity of this commonly used measure and undermining
characterizations of the American public as either policy centrist or ideologi-
cally innocent.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
The Nature of Political Ideology 681

40% -

35% - ^-""V

30% - y^ ^

25% - ' J]l^w- -- ~""~

20% - >~^^

1 5% - ^ _ _

10% -

5% -

o% -
I I I I I ι ι I

1972 1976 1980 1984 1988 1992 1996 2000

Figure 1. Political Ideology, American Nationa

File.

Measuring Ideology
In characterizing the ideological preferences o
typically rely on a survey question asking res
a liberal-conservative continuum. The questio
sort: "When it comes to politics, do you usually
liberal, liberal, slightly liberal, moderate or m
servative, conservative, extremely conservativ
about this?"1
As Fiorina (2004) and others have pointed
of Americans consider themselves to be extrem
conservative. As shown in figure 1 the plura
themselves as ideologically moderate or say "
cans like to consider themselves "middle class
themselves as ideologically moderate. The tren
icans today are better able to place themselve

1. This is the standard question asked since 1972 in th

replicated in other surveys).
2. The "extreme" categories are collapsed with the respect
In 2000, for instance, 2 percent of respondents called them
"extremely conservative."

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
682 Treier and Hilly gus

in DK responses), but we also see a

identifying as moderate.
Does this imply that the public is n
ences? Or is this a reflection of th
In other words, what exactly do t
opinion research treated the high l
idence that the public lacked the p
(Converse 1964). This perspective
connects between self-identified i
and Fleishman 1988), response inst
(Kerlinger 1984), and sensitivity t
1981). Levitin and Miller (1979, p.
themselves as having an ideologi
something about their positions on
policy stands." Others found that id
bolic considerations, group affiliat
political issues (Conover and Feldm
In contrast, recent research concl
political environment has helped t
for the general public (Levine, Car
and Saunders 1998). When political
politics becomes packaged on an id
Thus, as candidates have polarized
ological rhetoric, ideological labels
voters, thereby allowing individual
priate ideological category (Levendu
campaigned on a message of "comp
though AI Gore did not label himse
he clearly associated himself with
unions and civil rights organization
2000 electorate finds that "citizen
quite accurately along the liberal-c
not confined to the most knowledg
prior election years" (Jacoby 20
explains: "while a generation ago d
(e.g., Free and Cantril 1967), direct
the dominant means of assessing in
Yet, even as the public might b
conservative or liberal, it does not
equately captures their policy prefe
tion assumes that individuals can
dimension but, as the political wo
differentiation between social and
Inglehart 1997; Layman and Carsey

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 The Nature of Political Ideology 683

the distinctiveness of these two dimensions of ide

quences for our ability to interpret responses to the
(Kerlinger 1984).
If ideological preferences are multidimensional, it m
unidimensional ideology question, especially the mo
likely capture not only those who are centrist but
pressured between policy domains. For someone w
one policy dimension and a conservative one on an
"conservative" labels are simply inadequate descript
such, when asked their political ideology on a one
individuals should be more likely to say DK or to s
Research on attitudinal ambivalence has shown that individuals who are torn
between competing considerations are more likely to skip the survey question
or to select the middling category (Alvarez and Brehm 1995). We expect a
similar pattern for the ideologically cross-pressured.3 On the other hand, even
if we conceptually recognize the existence of multiple dimensions, a single
dimension will remain adequate if ideology is able to predict preferences across
a variety of different issue domains, as appears to be the case for political elites
(McCarty, Poole, and Rosenthal 2006).
In the analysis that follows, we examine the extent to which the one-
dimensional ideology question captures the policy preferences of the American
public in today's polarized environment by examining the contours of policy
attitudes as they relate to ideological identification. We then scrutinize any dis-
connects between policy preferences and ideological identification to determine
if they are the result of inadequacies of the survey respondents or inadequacies
of the survey question (Achen 1975).

Data and Methods

To evaluate the relationship between policy attitudes and ideological identity,

we estimate latent measures of economic and social policy preferences using
Bayesian item response theory (IRT). The Bayesian IRT model offers a number
of methodological advantages to alternative methods, such as an additive scale
of issues (Heath, Evans, and Martin 1994; Abramowitz and Saunders 1998) or
factor analysis (Layman and Carsey 2002; Ansolabehere, Rodden, and Snyder
2008). An additive scale, although easy to compute, assumes that every issue
contributes equally to the underlying preference dimension. The IRT measure,
like factor analysis, does not require such an assumption. For instance, if social
preferences are more strongly related to abortion attitudes than to environmental
policy attitudes, this difference will be captured in the IRT discrimination
parameters. But in contrast to conventional factor analysis, the IRT measure

3. Using the 1984 NES, William Jacoby similarly finds that attitudinal consistency is related to
ideological identification (Jacoby 1991).

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 684 Treier and Hilly gus

does not assume a multivariate normal distribution for all observed variables.
When this is not an appropriate approximation (e.g., dichotomous or ordinal
variables), conventional factor analysis can produce biased preference estimates
(Kaplan 2004).4 The IRT model directly models the appropriate distribution
of the observed indicators, whether nominal, binary, ordinal, or continuous (or
any mixture of types).
Finally, with the Bayesian IRT model, the latent measures (or factor scores)
are estimated directly and simultaneously with the discrimination parameters -
rather than as postestimation by-products of the covariance structure, as is the
case with conventional factor analysis. Consequently, these traits are subject
to inference just like any other model parameter, so we can calculate the
uncertainty estimates for the latent measures. It is a simple fact that all latent
concepts are necessarily measured with error, but alternative methods require
the assumption that the resulting estimate is the "true" value. In contrast, we can
quantify if we do a better or worse job of estimating someone's placement on
the ideological dimensions. And we can then take into account this uncertainty
when we use these latent measures as independent variables in subsequent
empirical models.
In estimating our latent policy dimensions, we rely on 23 questions from the
2000 American National Election Study (survey details and question wording
reported in the appendix). Although these questions do not exhaust the universe
of policies that might be related to an individual's general belief system, they
offer a wide range of politically relevant issues. Five of the issue questions
had a split sample design, in which respondents received either a "scale" or
"branching" question format, so each respondent was asked just 18 different
policy questions. Because the ideology questions are split between a branching
and scale format in the preelection survey, this survey also offers the opportunity
to evaluate different approaches to measuring ideological identification.
We model individual issue responses as a function of the unobserved prefer-
ence dimension via an ordinal item-response model (Treier and Jackman 2008).
Given the large number of parameters in the model and the difficulty in esti-
mating the parameters jointly using classical methods of maximum likelihood,
we work in a Bayesian setting, using Markov Chain Monte Carlo (MCMC)
methods to explore the joint posterior density of the model parameters (for
a survey of these methods and their applicability to researchers see Jackman
(2000, 2004); Gill (2008)). We implement this MCMC scheme using Win-
BUGS (Lunn et al. 2000).5 We use diffuse normal priors for the discrimination
parameters with mean zero and variance 1,000 and standard normal priors for

4. Alternative approaches are available (e.g., factor analysis on polychoric correlation matrix) but
these too have limitations and it is more common for researchers to simply ignore this assumption.
5. We let the algorithm run for 100,000 iterations as burn-in, moving away from the start values
such that subsequent iterations represent samples from the joint posterior density. Estimates and
inferences are based on 500,000 iterations, thinned by 100, in order to produce 5,000 approximately
independent draws from the posterior density.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
The Nature of Political Ideology 685

the preferences. The cutpoints are parameterized as th

fined as in Treier and Jackman (2008, p. 215) and ar
ensure the ordinality of the cutpoints.
One attractive feature of the Bayesian approach is that
of missing data, which are quite high in the measur
of the question wording experiments and because of
instance, for individual issue questions we find that a
respondents refused to answer or gave a DK response.
of respondents answered all issue questions that were
Bayesian approach, an individual's latent ideology sco
the data available for that individual, and those estimate
for those with less data. Critically, that uncertainty c
in subsequent statistical models. In contrast, classical
require a correction to the "swiss cheese" data structu
the different question formats, using listwise deletion,
data to fill in the holes.
Estimation of the IRT model requires a number of re
cation. For a one-dimensional model, the location and
normalizing the mean to zero and the variance to one. F
sion, similar to a confirmatory factor analysis, the para
with appropriate restrictions on the discrimination pa
this requires identifying a representative item for eac
the discrimination parameter is set equal to 1, and re
these items to load only on that dimension (for detail
2000). To allow for a more complex and realistic late
additional restrictions, as discussed below, and condu
ensure the restrictions do not unduly impact the results

Dimensions of Ideology
Is one dimension of ideology sufficient to capture th
the American public? We compare alternative estimati
table 1. Reported are the estimated discrimination par
extent to which each issue explains variation in the la
item does not help us distinguish among respondents wi
on each dimension, the discrimination parameter will be
zero. The wide variation we see in the size of the dis
highlights the advantage of the IRT measure over a sim
would have assumed that all issues loaded equally on
continuum.

Starting with the unidimensional measure reported in the first column, we

see that the individual social issues do not load as highly on the latent scale
compared to the economic issues. We are hesitant to conclude that this means

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 686 Treier and Hilly gus

Table 1. Discrimination Parameters for One and Two-Dimensional Models

Independent Correlated
ID Economic Social Economic Social

Aid to poor spending 1.49 1.89 -0.24 1.80 0.00

Government services (branching) 1.59 1.67 0.00 1.70 0.00
Guaranteed jobs (branching) 1.33 1.60 -0.18 1.53 0.00
Health insurance (branching) 1.23 1.37 -0.10 1.36 0.00
Public school spending 1.28 1.24 0.26 1.27 0.00
Welfare spending 1.06 1.15 0.01 1.17 0.00
Guaranteed jobs (scale) 0.99 1.13 -0.09 1.12 0.00
Social security spending 0.88 1.10 -0.31 1.03 0.00
Government services (scale) 1.49 1.00 0.00 1.00 0.00
Health insurance (scale) 0.93 0.91 0.08 0.92 0.00
Tax cut from surplus 0.40 0.25 0.50 0.29 0.00
Affirmative action 0.92 0.94 0.12 0.91 0.11
Environment (scale) 1.13 0.99 0.44 0.87 0.41
Gun control 0.97 0.85 0.44 0.73 0.41
Environment (branching) 0.92 0.81 0.37 0.72 0.35
Death penalty 0.46 0.42 0.19 0.38 0.15
Abortion, partial-birth 0.44 0.21 0.76 0.00 0.72
Abortion, parental consent 0.39 0.09 1.12 0.00 0.96
Abortion 0.45 0.00 1.00 0.00 1.00
Women's role (scale) 0.65 0.41 1.06 0.00 1.09
Women's role (branching) 0.43 0.11 1.29 0.00 1.14
Gays in military 0.69 0.45 1.58 0.00 1.52
Gay adoption 0.85 0.77 2.43 0.00 2.51
DIC 65,842.6 63,839.2 64,010.7
Correlation 0.30

that social issues ar

we included more e
not terribly surpris
based. More import
on a single dimensio
a distinct dimension
fall along at least tw

6. Others have identified

1989). As a robustness ch
action attitudes (report
characteristics - small an
the small number of que
spending) had even large
to another dimension. Mo

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 The Nature of Political Ideology 687

Confirming this expectation, we find that two dim

dle columns in table) better summarize the public
evidenced by the improvement in the Bayesian dev
(DIC), the goodness of fit measure. The smaller the
model. We also see that the items which loaded wea
model are the strongest items on the second (social
This two-dimensional model was estimated with the loosest restrictions
possible - an unconstrained model where all but two indicators were allowed to
load on either dimension.8 The economic dimension is defined by the question
concerning government spending and services (scale format), while the item
on abortion is associated exclusively with the social dimension. Although this
specification makes clear that a two-dimensional model is preferable to a one
dimensional one, it seems unrealistic to assume that the economic and social
dimensions are orthogonal. To relax this assumption, allowing preferences on
these two dimensions to be correlated, we have to implement additional restric-
tions. To estimate these correlated dimensions, we restrict many of the items to
load on only one dimension or the other. Those items that did not clearly load
better on one dimension or the other in the independent models were allowed
to load on both dimensions.9 The resulting discrimination parameters are re-
ported in the final columns of the table.10 In addition to providing a clearer
structure which identifies these dimensions as economic and social preferences,
these preferences are now allowed to correlate, which they do at a moderate
level of 0.30. Reassuringly, we find the individual discrimination parameters
look nearly identical with either specification. In the remainder of the analy-
sis, we rely on this correlated two-dimensional latent measure to evaluate the
self-reported measures of ideology. ] l

dimensions are unchanged by the inclusion of a third dimension. The correlation between economic
dimensions with and without the third dimension is 0.979 (0.954 for social dimensions).
7. Indeed, the respondents preferences in the ID model match well with the economic preferences
in the 2D model (0.962 correlation), but less so with social preferences (0.580).
8. An alternative identifying restriction is to fix individual respondents on the latent dimension. We
obtain similar results when normalizing by fixing the positions of three respondents: one extremely
liberal on both dimensions [-3, -3], one extremely conservative on both [-3, -3], and one
respondent extreme on both, but cross-pressured [-3, -3]. The resulting latent scores correlate
with those reported at .995 on the economic dimension and .995 on the social dimension.
9. As a robustness check, we have also estimated a model where every item loaded only on one
dimension, with nearly identical results. The resulting latent scores correlate with those reported
at .995 on the economic dimension and .993 on the social dimension.
10. We find no evidence that the estimates do not converge. For each set of estimates, 5.5 percent
or less of the parameters fail Geweke's diagnostic, an amount, under standard levels of statistical
significance, one would expect to randomly occur if all of the parameters converged to the stationary
distribution.

1 1 . A comparison of an additive scale created using the issue items available for each respondent
finds a correlation of .82 for the economic dimension and .67 on the social dimension. The
lower correlation for the social dimension no doubt reflects the greater variability found in the
discrimination parameters, again affirming the advantages of the IRT model. Unfortunately, we

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 688 Treier and Hillygus

Figure 2. Latent Policy Preference

Comparing Ideology Meas

We start
separately comparin by
self-placement scale in the postele
responding box plot of estimated
The figure shows a clear relations

cannot create a directly comparable estimate

and the discrete nature of some of the variables.
12. In a box plot graph, the box indicates the interquartile range (marking the lower quartile,
median, and upperquartile), the whiskers show the range of the data (1.5 times the interquartile
range), and circles indicate outliers.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
The Nature of Political Ideology 689

self-placement - with self-identified liberals havin

erences on social and economic issues than self-identified conservatives and
vice-versa. At the same time, the graph highlights the considerable variabil-
ity in the policy preferences of each group, especially among self-identified
moderates and those answering "don't know." Indeed, the interquartile range
(the box) of the moderate respondents overlaps the median value of all but
the extreme ideologues, offering the first indication that the moderate label, in
particular, covers a wide range of policy attitudes.
The box plot graphs offer other interesting comparisons as well. First, relative
to self-identified liberals (of any intensity), self-identified conservatives in the
sample have more diverse policy preferences on both dimensions, as evidenced
by the longer whiskers. This no doubt reflects the negative connotations associ-
ated with the liberal label that has been prevalent in American politics in recent
decades. Differences in the social desirability of the ideological labels mean
that more people are willing to identify as "conservative" even though they
don't necessarily hold the policy preferences associated with the label (Miller
1992). Comparing across dimensions also finds that self-identified liberals and
conservatives are more polarized (i.e. further from 0) on the social dimension
than on the economic dimension.
Second, we see that the intervals between categories are not equal, as is
assumed by the 7-point self-placement scale. In particular, there is a much
larger gap in the median between those who call themselves "extremely" liberal
(conservative), and those who call themselves liberal (conservative) relative to
other gaps on the scale. In contrast to the ideological self-placement measure -
in which ideological labels can mean different things to different people -
our latent measure of policy preferences places all respondents on the same
underlying policy scale and provides finer distinctions between individuals.
Finally, we see that those who say "don't know" have a preference distri-
bution quite similar to that of self-identified moderates. Self-reported ideology
questions often have rather high levels of DK responses, leaving scholars un-
sure how to handle the missing data problem. The similarities we see here offer
some empirical justification to simply recoding DK to be moderate (rather than
omitting them from the analysis), as is common in some research. To be clear,
though, the reason the two groups look so similar is that self-identified mod-
erates, like the don't knows, have an equally diverse set of policy preferences.
Undoubtedly, both the DK and moderate responses are selected for a variety of
reasons, only some of which reflect centrism (for self-identified moderates) or
uncertainty (for DK).
The latent measures offer a benchmark for evaluating the various mea-
sures of ideology available in the 2000 American National Election Study.
Comparing the correlations between each of the ideology questions and the
latent measures, shown in table 2, finds that the postelection scale measure
outperforms the other two. It is perhaps not surprising that a postelection mea-
sure outperforms a preelection measure since people might be better able to

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 690 Treier and Hilly gm

Table 2. Comparison of Latent Mea

Latent economic Latent social

Correlation Correlation

Preelection branching ideology .425 .490

Preelection scale ideology .478 .523
Postelection scale ideology .486 .547

Note. - Polyserial correlations are reported.

select an appropriate ideological label because of cam

we also find that the preelection scale outperforms t
question. This result offers some challenge to previou
that branching measures were preferable to scale measu
but is consistent with other analysis of the 2000 ANE
and McKay 2002). For the remainder of the analysis
identification, we rely on this standard postelection 7
Turning to a comparison of the individual-level rela
social and economic dimensions, we plot in figure 3
cial and economic scores for self-reported conservati
and don't knows.13 For those who call themselves lib
see that there is a clear relationship between the la
dimensions, with respondents clustered in the corres
liberal quadrants of the graphs. Thus, the more con
preferences on social issues, the more conservative
economic issues, offering evidence of ideological cons
of the electorate.
In contrast, there is a much weaker relationship bet
issue preferences for moderates and DK respondents.
see a diffuse cloud of data points and a smaller Beta
dicates that individuals who self-identify as moderate
or moderate across political issues, but are often cro
economic and social dimension (located in quadrants 2
of moderates and 47 percent of DK respondents are i
quadrants. If we define someone as holding "centrist
fall within the middle tercile of both the economic and social latent dimen-
sions, we find that just 17 percent of self-identified moderates have centrist

13. The end categories of the post-election 7-point scale are collapsed so liberals and conservatives
include the "extreme" ideologues. Not shown are the "slightly" categories; not surprisingly, the
observed relationships for this group fall in between that of ideologues and moderates.
14. The reported Betas are the regression coefficients for the economic dimension regressed onto
the social dimension.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 The Nature of Political Ideology 691

Liberal Moderate
β = 0.37 β = 0.21
CO -ι CO -

10.9% 8.4% 20.7% 24.9%

CM - CM - · ..

I -:? -.··· · :>>fc;·'.··

b ι ι- ι .*..-·'· ; · ι t ι ι ι ■ ■'?:·'':■'·' ί^Ι; Γ1 ι
« _3 _2- -J /.%-.' ·1 2 3 -3 -2 :'^Í^V 2 3

65.3% 15.3% 31.6% 22.8%

7 J 7 J

Conservative Don't Know

β = 0.32 β = 0.21
CO -ι CO η
15.3% m * 62.1%· 34.7% . 26.1%
m * CM -.-'%.*· CM -

§ .··'··.:::<··:····· ·:.Γ·'ν·..·.:·....·
i ' · .·*". f /;ífe % · .:. ' · : ·-· ëi ^ ; · ·
Q Ι ι . ·'·..·?',· ν1 λ·1· Π Ι ι. ί*.;·_}" '· ' Λ·Λ· ' '
S -3 -2 '-A'*?:·' "il ·' · 2 3 -3 -2 '/j-t^J '.".'■ 12 3
ο ι "".'.""' .'·*:► ".·"*·"
CM _ CM !
8.9% 13.7% 27.1% 12.1%
co J co J

Economic Dimension Economic Dimension

Figure 3. Economic versus Cultural Dimension of Ide

Ideology. Regression coefficients and percent of obser
category.

preferences.15 Even expanding the definition of centrist to include respondents

with preferences in the 25th to 75th percentile of both dimensions, we still find
that only 35 percent of self-identified moderates hold centrist positions on both
policy dimensions.
We also find slight asymmetries between liberal and conservative identi-
fiers that again seem to reflect differences in the social desirability of the two

15. Using this same threshold, we find that 59 percent are cross-pressured, either centrist on one
dimension and extreme on the other or extreme on both in opposing directions, another 8 percent
are misclassified conservatives, and 15 percent are misclassified liberals.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 692 Treier and Hilly gus

labels. Self-identified liberals have

nomic and social dimensions on
servatives. And slightly more cons
quadrant (38 percent compared to 3
the various thresholds for the latent
identified liberals and conservative
their ideological labels across both
vatives are completely misclassifie
dimensions - fewer than 9 percen
fewer than 3 percent based on the
In sum, this descriptive analysis ind
ronment, the commonly asked surv
is inadequate for capturing the com
the American public. As political co
of us are mavericks, political mutts
mix of liberalisms, conservatisms
These political mutts often identify
to interpret the standard measure of

Predicting Use of Ideologic

The descriptive analysis above sugge
up of at least two very different
ideologically cross-pressured. The k
the observed variation in economi
lack of political sophistication. To a
logit model with the appropriate co
select the Moderate, Ideologue, or D
we measure cross-pressured prefe
social dimensions (multiplied by -
cross-pressured), so that the measu
dimensions. We include in the m
account for the possibility that mod
ignorance, as hypothesized by Con
potential symbolic aspects of ideolo
measure (partisan versus independe
aspects of ideological labels with de
age. Finally, we control for policy
social dimensions.

16. For instance, 10 percent of self-identified liberals have preferences in the conservative tercile
on at least one dimension, while 12 percent of conservatives have preferences in the liberal tercile
on at least one dimension.
17. Following Bartels (1996), we use the interviewer assessment of political knowledge.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 The Nature of Political Ideology 693

The model results are estimated simultaneously w

take into account the measurement error involved in the estimation of our
latent measures.18 Thus, if we have a lot of uncertainty in our estimate of the
economic or social preferences of an individual - perhaps because of skipped
questions or a random response pattern - that uncertainty is carried over into
the subsequent model estimates. Simply including the latent measure without
accounting for the uncertainty in the estimate of that measure - the approach
typically used with factor scores from a traditional factor analysis - could result
in biased coefficient estimates. In this respect, our approach offers additional
reassurance that we are controlling for political knowledge by accounting for
the uncertainty in the latent measures (since some individuals are measured
with more error than others).
The findings are reported in table 3. Reported are the coefficient estimates
and the 90 percent highest posterior density (HPD) intervals. This offers a
gauge of "significance" since we can say there is a 90 percent probability that
the coefficient lies within the interval.19 Looking first at some of our political
controls, we see that the decision to identify as a political moderate is related
to many of the characteristics hypothesized in previous research. Partisans are
less likely than independents to identify as moderates, and the more politically
knowledgeable are less likely than the politically ignorant to call themselves
moderate (and more likely to call themselves moderate than to say "Don't
Know"). Even with all of these controls, though, being cross-pressured has
a sizable and significant effect on use of the moderate label. There is a 93
percent probability that the cross-pressure coefficient is greater than zero for
the Moderate versus Ideologue comparison.20
To illustrate the substantive impact of policy cross-pressures, we graph in
figure 4 a contour plot of predicted probabilities to show the variation in the
probability of identifying as a moderate across various levels of economic and
social preferences. We see that those most cross-pressured (the darker areas
in the image) have a higher elevation or probability of calling themselves
moderate.21 Individuals who are the least cross-pressured (most ideologically
extreme on both dimensions) have just a 10 percent chance of calling themselves
ideological moderate, compared to a 25 percent chance among those with

18. The simultaneous estimation is implemented by imposing a standard multinomial logit model
for self-placement as an ideologue, moderate, or DK alongside the measurement model (with
parameters Θ), assuming vague normal priors on the logit coefficients (labeled γ). The MCMC
chain then updates the estimates of γ, given the previous estimates of Θ, and θ is updated given the
previous estimates of γ .
1 9. By contrast, with standard (frequentist) confidence intervals, we know only that 90 percent of
the time the interval contains the true value; there is no indication of how likely a set of values are.
20. There is an 89 percent probability that the coefficient is greater than zero for the DK v.
Ideologue comparison.
21 . Estimates are made holding variables at their mean category or mode for indicator variables.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 694 Treier and Hilly gus

♦-ess
«< h t β
ο ν öιo' ο
6 «ηιd ν«ci (sιo ^ηοο ι
o σ; o
ο r--o rt: ι οο ι μ. ι ο I d
SË £ © « © 7 7 £ ? 7 ο 1 I
1

î g ^ „ ^ S

I .-S o§§d|d|d
^â|îôîôo^j 1 1 i â 1

3 Cõ1 ι - ii - ir- » Co1 So1 cn'S'jr1

go ι - i^HONOomirjvq ι »n - < o
> ä oo2<2c>ooC>ooC>inG*^©Tt<? § I g I 2

^ CO

5 S s °.S ? « °:s! ' a ? s * s Ξ-s - s ' s '.ζ 5 .s

!<5!§Ϊ7ο£±277Λ
uL " "
" "
.s
g

£ω ,
I
18,
8, ^^ ||ωω§-s8ï:|P.2Na(ieS-?^dNd-SigSe2s2Sïî
ν q ο !; . g -« ^ __. ^ I >n
sß *^ S Nhò^d-!òiododinòK-tMÒ'oO^Oo'(:iooo50 ^
§ ν ο' q!; .^ges
-« _y
I >n 'qq ^^ es
-; q^ -;gjgj *>
*> «n = -; -;S ^^q qs
^ «n

I 111 Πϊ^'Π'Π'ϋ ι ι ! ^
I
I SS
υ

ώ ι

3
'%
S
3

- i

Ill« Mtia
* - ë ë §
H ^ I i S. ε| ο |

I I I 1 1 1 I I I I á *S !!!

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 The Nature of Political Ideology 695

Figure 4. Predicted Probability of Identifying as a

the greatest ideological distance between their eco

preferences (97.5th percentile of cross-pressures m
As an additional robustness check, we have estim
the politically knowledgeable respondents (rather
for them in the models), and find nearly identical re
for symbolic considerations, political knowledge, an
uals with divergent economic and social preferenc
themselves moderate than to use a liberal or cons
offers an important corrective to recent characteriza
lic as being either centrist or polarized. The reality is
the mass public are more complex - and the label
often used to describe the American public represe
attitudes.

Implications
The finding that the moderate label masks differences between policy centrists
and cross-pressured respondents has consequences for our theories and models

22. Examining changes in the predicted probabilities across varying levels of values for all three
variables still finds that the individuals are most likely to identify as a moderate when they are
centrist on both dimensions or when they are cross-pressured between them.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
696 Treier and Hilly gus

of political behavior. Ideologically m

heart of theories of electoral democ
instance, predicts that moderates are
politicians to advocate centrist pub
also considered the all-important s
support across or within elections
national elections (Converse 1964).
example, one analyst concluded: "th
our recent history - was determined
and moderates."23 There is a clear te
political moderates as a homogeneou
attitudes and behavior. Yet, our ana
the policy preferences of moderates
their voting behavior.
We evaluate this possibility empiri
choice models that either include o
controlling for ideological self-plac
widely different predictions about t
graph plots the estimated error in
as homogeneous in their policy pr
of economic and social preferences
probability of voting for Bush can
In contrast, predictions are off by n
(not statistically significant) and 13
changing predicted candidate choic
placement scale may be adequate in
conservatives and predicting their p
for self-identified moderates.
Beyond the example above, the co
we have identified may have a varie
dynamics. A more complete under
ideological identification could wel
about split ticket voting, political
over, recognizing the distinction be
cross-pressured might affect our b
centrists and cross-pressured respo
campaign strategies (Hillygus and

23. Andrew Kohut, "The Real Message of th

24. The logit model is again estimated sim
identification, age, gender, race, income, an
are calculated holding all other variables at t
available in an online appendix on the POQ w
conservative scale is replaced by dummy va

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
The Nature of Political Ideology 697

Social _ 0.4- Economic

--- 25% ..··' --- 25%
0.3- 75% / / 75%

*■ °·1- / / /' * 0.1- / / /'

% o.o- / / / I o.o- / / /'
I -0.1 - / / / I -0.1 - / / /
-0.2- / / -02- /

-0.3- .'' -03- .''

-1.5 -1.0 -0.5 0.0 0.5 1.0 -1.5 -1.0 -0.5 0.0 0.5 1.0

Economic Policy Preferences Social Policy Preferences

Figure 5. Error in the Predicted Probability of a Bush Vote.

Discussion

Given the current polarized nature of American politics, renewed attention has
been focused on the ideological preferences of the mass public. Yet, the way
we conceptualize and measure those preferences shapes our conclusions about
their distribution and influence.
Our analysis documents the multidimensional nature of policy preferences in
the American electorate, and finds a noteworthy number in the public are liberal
on one dimension and conservative on another. Because these cross-pressured
individuals tend to call themselves moderate (or say DK), it undermines inter-
pretation of the standard 7-point ideological identification scale so often used
in political research. Thus, even as scholars find that ideological labels are more
meaningful than ever before, those labels are accurate representations of policy
preferences only for those self-identifying as a liberal or conservative.
We are by no means the first to acknowledge that ideological self-placement
is a flawed measure because of mismatches between ideological identification
and policy preferences. But in contrast to early research, we cannot attribute the
disconnect between self-reported ideology and issue attitudes to the lack of po-
litical knowledge alone. Rather, many people are coherent along the economic
and social dimension separately, but are simply cross-pressured between them.
Our results show that failing to account for the multidimensional nature of
ideological preferences can produce inaccurate predictions of voting behavior

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 698 Treier and Hilly gus

for the plurality of Americans who

tive. As such, we recommend that fu
and economic preferences in empiri
dard approach to measuring ideolog
a unidimensional scale, it would be
the potential for including direct m
logical dimensions as an alternative
(see Hooghe, Marks, and Wilson 200
a set of valid and reliable survey q
include a large and diverse set of p
so that issue-based scales can be created.
To be sure, our findings do not imply that the ideological self-placement
measure should never be used. Scholars have long noted the symbolic im-
portance of liberal and conservative labels (Stimson 2004), and our results
suggest that these labels are meaningful representations of policy preferences
for self-identified liberals and conservatives. However, researchers should at
least operationalize ideological self-placement as a series of dummy variables
in their empirical models since the measure cannot be assumed to be an ordinal
scale with political moderates in the middle. And even then, this approach can-
not distinguish between so-called moderates who are centrist and those who are
cross-pressured, making it inadequate for any theory or model that is dependent
on a measure of policy centrism.
Beyond these practical implications, these findings are relevant to the on-
going polarization debate. On one side are those who say that political moder-
ates have either followed political elites to the ideological extremes or, frustrated
by the polarized environment, have dropped out of the political process alto-
gether (Abramowitz and Saunders 1998, 2005; Layman and Carsey 2002). On
the other side are those who contend that the majority of Americans have re-
mained ideologically centrist even as political elites have grown more polarized
(Dimaggio, Evans, and Bryson 1996; Fiorina 2004). It turns out that neither
portrait of the American moderate is entirely accurate. Our findings make clear
that the American public is not as ideologically extreme as often portrayed, but
nor are they truly centrist. And the heterogeneous political complexion of the
American public has consequences for the way we measure political ideology
and the way we use it in our theories and models of political behavior.

Appendix

2000 AMERICAN NATIONAL ELECTION STUDY

The 2000 ANES was conducted by the Center for Political Studies of the
Institute for Social Research. The preelection survey was conducted from
September 5 to November 6, and the postelection re-interview ran from

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
The Nature of Political Ideology 699

November 8 to December 18. The study population

voting age living in the forty-eight contiguous stat
a dual frame sample that included both a traditiona
using face-to-face (FTF) interviews (1,006 pre resp
stratified equal probability sample interviewed by
dents). The response rates, calculated as the ratio o
the total number of potential respondents, were 64
percent for phone for preelection and 57.2 percent
for phone for postelection. More details about the
at http://www.electionstudies.org/studypages/2000

QUESTION WORDING

All variables were recoded to run from liberal to conservative. Question wording
for model controls are available in an online appendix on the POQ website.

Ideological self-placement measures: Liberal-Conservative Scale

(V001368/V000440/VCF0803) [FTF]: "Where would you place yourself on
this scale, or haven't you thought much about this? Scale: (1) extremely liberal,
(2) liberal, (3) slightly liberal, (4) moderate; middle of the road, (5) slightly
conservative, (6) conservative, (7) extremely conservative." [phone]: "When
it comes to politics, do you usually think of yourself as extremely liberal,
liberal, slightly liberal, moderate or middle of the road, slightly conservative,
conservative, extremely conservative, or haven't you thought much about
this?"; Liberal-Conservative Branching Measure (V000446) (FTF and phone):
"When it comes to politics, do you usually think of yourself as a liberal, a
conservative, a moderate, or haven't you thought much about this? If you had
to choose, would you consider yourself a liberal or a conservative? Would
you call yourself a strong liberal or a not very strong liberal? Would you call
yourself a strong conservative or a not very strong conservative?"

Issue questions: V000748: "Do you think gay or lesbian couples, in other
words, homosexual couples, should be legally permitted to adopt children?";
V000545: [FTF] "Where would you place yourself on this scale, or haven't
you thought much about this? (1-7 scale) 1 govt should provide many fewer
services, 7 govt should provide many more services"; V000549: [phone]
"Which is closer to the way you feel or haven't you thought much about this?
Should the government reduce/increase services and spending a great deal or
(reduce/increase services and spending) only some."; V000609: [FTF] "Where
would you place yourself on this scale, or haven't you thought much about
this? 1-7 scale, 1 govt insurance plan, 7 private insurance plan"; V000610:
[phone] "Which is closer to the way you feel or haven't you thought much
about this? do you feel strongly or not strongly that there should be a gov-
ernment insurance plan?/do you feel strongly or not strongly that individuals

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
700 Treier and Hilly gus

should pay through private insura

you place yourself on this scale, or
scale: 1. govt should see to jobs and
person get ahead on own."; V000
you feel or haven't you thought muc
government should see to it that e
of living, or not so strongly? Do you
just let each person get ahead on t
[standard version] "Some people thi
criminating against blacks when m
required to have an affirmative act
hiring. What do you think? Should
blacks have to have an affirmative
SION] Some people think that if a
against blacks when making hiring
have an affirmative action program
do you think? Should companies t
to have an affirmative action prog
an affirmative action program? [B
not strongly (that they should not
"Should federal spending on welfar
about the same?"; V000680: "Should
be increased, decreased, or kept ab
spending on social security be incr
V000683: "Should federal spending
or kept about the same?"; V000690:
the expected federal budget surplus
or disapprove of this proposal? Do
not strongly? Do you disapprove of
V000694: [FTF] "There has been som
years. Which one of the opinions on
can just tell me the number of th
been some discussion about abortio
you a short list of opinions. Please te
with your view? You can just tell m
options: (1) by law, abortion should
mit abortion only in case of rape, in
(3) the law should permit abortion fo
to the woman's life, but only after
established. (4) by law, a woman sho
a matter of personal choice."; V00
yourstate that would require a teena
permission before she could obtain
V000705: "There has been discussio

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 The Nature of Political Ideology 701

certain types of late-term abortions, sometimes cal

Do you favor or oppose a law that would make these
Do you strongly or not strongly favor/oppose a law t
these types of abortions illegal?; V000727: "Do you
be allowed to serve in the United States Armed For
Do you feel strongly or not strongly that homose
serve? Do you feel strongly or not strongly that h
allowed to serve?"; V000731: "Do you think the fe
make it more difficult for people to buy a gun tha
for people to buy a gun, or keep these rules about
A lot easier/more difficult or somewhat easier/mor
you favor or oppose the death penalty for persons
you favor/oppose the death penalty for persons con
or not strongly?"; V000755: [FTF standard] "Where
on this scale, or haven't you thought much about
where would you place yourself on this scale? 1-7
should have equal roles, 7. a woman's place is in th
version 1] "Which is closer to the way you feel, or
about this? [phone version 2] which is closer to the
versions] Do you feel strongly or not strongly that m
equal roles? Do you feel strongly or not strongly
the home?"; V000771: [FTF] "Where would you pla
or haven't you thought much about this? 1-7 scale:
business needed to protect environment, 7. regulatio
already too much a burden on business."; V000775
to the way you feel, or haven't you thought much
toughen regulations to protect the environment a
regulations to protect the environment way too mu
or just somewhat of a burden?"

Supplementary Data
Supplementary data are available online at http://p

References

Abramowitz, Alan L, and Kyle L. Saunders. 1998. "Ideological Realignment in the U.S. Electorate."
Journal of Politics 60(3):634-52.
Abramowitz, Alan, and Kyle Saunders. 2005. "Why Can't We All Just Get Along? The Reality
of a Polarized America." The Forum: A Journal of Applied Research in Contemporary Politics
3(2): 1-24.
Achen, Christopher H. 1975. "Mass Political Attitudes and the Survey Response." American
Political Science Review 69(4): 1218-31.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 702 Treier and Hilly gus

Aguilar, Omar, and Mike West. 2000. "Bayes

Journal of Business & Economic Statistic
Aldrich, John, John Griffin, and Amy McK
Format on the National Election Studies." D
Aldrich, John, Richard G. Niemi, George R
ment of Public Opinion about Public Policy
American Journal of Political Science 26(
Alvarez, R. Michael, and John Brehm. 199
Development of a Heteroskedastic Probit
Political Science 39(4): 1055-82.
Ansolabehere, Stephen, Jonathan Rodden, an
Multiple Measures to Gauge Preference S
American Political Science Review 102(2):
Bartels, Larry. 1996. "Uninformed Voters: In
Journal of Political Science 40(1): 194-230.
Carmines, Edward, and Michael Ensley. 2004
Issue Preferences and Partisan Attitudes in
Political Science Association, Chicago.
Carmines, Edward G., and James A. Stimson
of American Politics. Princeton, NJ: Prince
Conover, Pamela Johnston, and Stanley Fe
Conservative Self-Identifications." America
Conover, Pamela Johnston, and Stanley Feld
A Schematic Model." American Journal o
Converse, Philip E. 1964. "The Nature of
Discontent, ed. David E. Apter. New York:
Dimaggio, Paul, John Evans, and Bethany
Become More Polarized?" American Journ
Downs, Anthony. 1957. An Economic Theor
Fiorina, Morris. 2004. Culture War? The My
man.

Free, Lloyd, and Hadley Cantril. 1967. The Political Beliefs of Americans: A Study of
Opinion. New Brunswick, NJ: Rutgers University Press.
Gill, Jeff. 2008. Bayesian Methods. 2nd ed. New York: Chapman and Hall/CRC Press.
Heath, Anthony, Geoffrey Evans, and Jean Martin. 1994. "The Measurement of Core Belief
Value." British Journal of Political Science 24( 1 ): 1 15-32.
Hightower, Jim. 1997. There's Nothing in the Middle of the Road but Yellow Stripes an
Armadillos. New York: HarperCollins.
Hillygus, D. Sunshine, and Todd Shields. 2008. The Persuadable Voter. Princeton, NJ: Prince
University Press.
Hinich, Melvin J., and Michael C. Munger. 1997. Analytical Politics. Cambridge: Camb
University Press.
Hooghe, L., G. Marks, and C. J. Wilson. 2002. "Does Left/Right Structure Party Positio
European Integration." Comparative Political Studies 35(8):965-89.
Inglehart, Ronald. 1997. Modernization and Postmodernization: Cultural, Economic, and Pol
Change in 43 Societies. Princeton, NJ: Princeton University Press.
Jackman, Simon. 2000. "Estimation and Inference via Bayesian Simulation: An Introduct
Markov Chain Monte Carlo." American Journal of Political Science 44(2): 375-404.
Jackman, Simon. 2004. "Bayesian Analysis for Political Research." Annual Review of Po
Science 7:483-505.
Jacoby, William. 1991. "Ideological Identification and Issue Attitudes." American Journal of
Political Science 35(1): 178-205.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms
 The Nature of Political Ideology 703

Jacoby, William. 2004. "Ideology in the 2000 Elections: A Stud

Voting in Presidential Elections, eds. Herbert Weisberg, and C
University Press.
Kaplan, David. 2004. The Sage Handbook of Quantitative Meth
Newbury Park, CA: Sage.
Kerlinger, Fred. 1984. Liberalism and Conservatism: The Nature
NJ: Hillsdale.

Knight, Kathleen. 1999. "Liberalism and Conservatism." In Measures of Political Attitudes, eds.
J. P. Robinson, L. S. Wrightsman, and P. R. Shaver. San Diego, CA: Academic.
Layman, Geoffrey. 1999. "Culture Wars in the American Party System: Religious and Cultural
Change Among Partisan Activists Since 1972." American Politics Research 27(1):89- 121.
Layman, Geoffrey, and Tom Carsey. 2002. "Party Polarization and 'Conflict Extension' in the
American Electorate." American Journal of Political Science 46(4):786-802.
Levendusky, Matt. 2009. The Partisan Sort: How Liberals Became Democrats and Conservatives
Became Revublicans. University of Chicago Press.
Levine, Jeffrey, Edward Carmines, and Robert Huckfeldt. 1997. 'The Rise of Ideology in the
Post-New Deal Party System, 1972-1992." American Politics Research 25(l):19-34.
Levitin, Teresa E., and Warren E. Miller. 1979. "Ideological Interpretations of Presidential Elec-
tions." American Political Science Review 73(3):751- 71.
Lunn, David J., Andrew Thomas, Nicky Best, and David J. Spiegelhalter. 2000. "WinBUGS - A
Bayesian Modelling Framework: Concepts, Structure, and Extensibility." Statistics and Com-
puting 10(4):325-37.
Marcus, George E., David Tabb, and John L. Sullivan. 1974. "The Application of Individual
Differences Scaling to the Measurement of Political Ideologies." American Journal of Political
Science 18(2):405-20.
McCarty, Nolan, Keith Poole, and Howard Rosenthal. 2006. Polarized America: The Dance of
Ideology and Unequal Riches. MIT Press.
Miller, Alan S. 1992. "Are Self-Proclaimed Conservatives Really Conservative? Trends in Attitudes
and Self-Identification among the Young." Social Forces 71(1): 195-210.
Robinson, John, and John Fleishman. 1988. "Ideological Identification: Trends and Interpretations
of the Liberal-Conservative Balance." Public Opinion Quarterly 52(1): 134-45.
Schuman, Howard, and Stanley Presser. 1981. Questions and Answers in Attitude Surveys. New
York: Academic.
Shafer, Byron E., and William J. M. Claggett. 1995. The Two Majorities: The Issue Context of
American Politics. Johns Hopkins University Press.
Stimson, James. 2004. Tides of Consent. New York: Cambridge.
Stone, Walter. 1991. "On Party Switching among Presidential Activists: What do We Know?"
American Journal of Political Science 35(3):598-607.
Treier, Shawn and Simon Jackman. 2008. "Democracy as a Latent Variable." American Journal of
Political Science 52(1): 201-217.

This content downloaded from 193.49.18.238 on Thu, 26 Jul 2018 17:03:18 UTC
All use subject to https://about.jstor.org/terms