Journal of Research in Personality 88 (2020) 104009

Contents lists available at ScienceDirect

Journal of Research in Personality

journal homepage: www.elsevier.com/locate/jrp

Moral grandstanding and political polarization: A multi-study

consideration
Joshua B. Grubbs a,⇑, Brandon Warmke b, Justin Tosi c, A. Shanti James d
a
Department of Psychology Bowling, Green State University, United States
b
Department of Philosophy Bowling, Green State University, United States
c
Department of Philosophy, Texas Tech University, United States
d
Department of Psychology Bowling, Green State University, United States

a r t i c l e i n f o a b s t r a c t

Article history: The present work posits that social motives, particularly status seeking in the form of moral grandstand-
Received 22 May 2020 ing, are likely at least partially to blame for elevated levels of affective polarization and ideological
Revised 10 August 2020 extremism in the U.S. In Study 1, results from both undergraduates (N = 981; Mean age = 19.4;
Accepted 17 August 2020
SD = 2.1; 69.7% women) and a cross-section of U.S. adults matched to 2010 census norms (N = 1,063;
Available online 24 August 2020
Mean age = 48.20, SD = 16.38; 49.8% women) indicated that prestige-motived grandstanding was consis-
tently and robustly related to more extreme ideological views on a variety of issues. In Study 2, results
Keywords:
from a weighted, nationally-representative cross-section of U.S. adults (N = 2,519; Mean age = 47.5,
Narcissism
Status-seeking
SD = 17.8; 51.4% women) found that prestige motivated grandstanding was reliably related to both ide-
Polarization ological extremism and affective polarization.
Virtue-signaling Ó 2020 Elsevier Inc. All rights reserved.
Grandstanding

1. Introduction 1.1. Ideological extremism and affective polarization

The American populace is sharply divided over political and According to recent studies, politicians (Banda & Cluverius,
moral issues (Dimock, Doherty, Kiley, & Oates, 2014). Online out- 2018), news media (Martin & Yurukoglu, 2017), and average citi-
rage about political phenomena is common (Brady & Crockett, zens (Johnston, 2018) are divided on a range of sociopolitical
2019; Crockett, 2017), and there is considerable evidence that pub- issues. Study of these divides has led to several recent empirical
lic discourse has become more ideologically polarized and less civil articles (for reviews, see: Iyengar, Lelkes, Levendusky, Malhotra,
over the past several years (Doherty, Kiley, Tyson, & Johnson, & Westwood, 2019; Lelkes, 2016), books (e.g., McCarty, 2019),
2019). Several possible explanations of both increasingly polarized and even documentaries (e.g., ‘‘Divided States of America,” 2017)
beliefs and a decline in civility have been posited, including eco- on the rifts between political partisans in the U.S. Although there
nomic trends, the behavior of political or party elites, media con- is broad consensus that some of these divides in the American pop-
sumption, technological influences, and partisan sorting (Brady, ulace are worsening (for a review, see: Iyengar et al., 2019), the
Crockett, & Van Bavel, 2019; McCarty, 2019). However, there is also nature of political and partisan polarization is a controversial topic
growing evidence that individual differences may contribute to in political science (Abramowitz & Saunders, 2008; Druckman &
these trends. The present work examines one such individual dif- Levendusky, 2019; Fiorina & Abrams, 2008) and political psychol-
ference—moral grandstanding motivation—that may be partially ogy (Blatz & Mercier, 2018; Westfall, Van Boven, Chambers, &
responsible for some of the divisions in contemporary American Judd, 2015). Some research has focused on divisions in the popu-
public discourse. lace in terms of partisan identity (Abramowitz & Saunders, 2008)
and extreme endorsement of ideological views (Brandt, Reyna,
Chambers, Crawford, & Wetherell, 2014), whereas other work has
focused on issue-based ideological divides (Fiorina & Abrams,
2008) or emotional reactions to outgroup members (Iyengar,
Sood, & Lelkes, 2012). For the present work, we were interested
⇑ Corresponding author at: Department of Psychology, Bowling Green State in studying these divides in terms of ideological extremism and
University, Bowling Green, OH, 43403, United States. affective polarization. Given that these terms are often used differ-
E-mail address: [email protected] (J.B. Grubbs). ently across different disciplines, we define our use of each below.

https://doi.org/10.1016/j.jrp.2020.104009
0092-6566/Ó 2020 Elsevier Inc. All rights reserved.
2 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009

1.1.1. Ideological extremism 2010 (Iyengar et al., 2012). Second, individual antipathy toward
We use the term ideological extremism as it has been understood ideological others has also apparently risen in recent years
in prior psychological literature, namely, the extent to which peo- (Iyengar & Westwood, 2015), with partisans themselves demon-
ple report extreme political or ideological views at either end of the strating stronger negative evaluations of opposing party members.
political spectrum (Lammers, Koch, Conway, & Brandt, 2017; van The present work seeks to describe one factor that may be con-
Prooijen & Krouwel, 2019). That is, we conceptualized ideological tributing to this rise in affective polarization: Moral
extremism as the self-reported adoption of extreme political views Grandstanding.
in either direction. Relatedly, we will use the term ideological polar-
ization to refer to the process by which an individual becomes 1.2. Status seeking and moral grandstanding
more ideologically extreme. We will therefore think of polarization
as the process that tends to result in ideological extremism. In this Psychologists have long known that, when likeminded individ-
project, we were specifically interested in examining factors that uals discuss issues together as a group, they often leave such dis-
predict ideological extremism (i.e., self-report of extreme political cussions with more extreme ideological positions than they
or ideological views), as such factors may also be related to the entered with individually (Moscovici & Zavalloni, 1969). This pro-
process of ideological polarization (i.e., the process by which one’s cess of adopting more extreme views is known as group polariza-
reported views move to extremes). tion (Mackie & Cooper, 1984; Mackie, 1986). Similarly, groups may
For the present work, we considered ideological extremism as a demonstrate collective emotions together, wherein extreme
key outcome variable in our analyses. However, we note ideologi- antipathy toward outgroup members is higher at the group level
cal extremism itself often functions as an important predictor of than at the individual level (Smith et al., 2007). Whereas past
symmetric effects in political and moral psychology (Brandt research, informed by social identity theory, has often speculated
et al., 2014). Ideological extremists demonstrate overly simplistic that conformity within the group is part of the reason for this pro-
evaluations of political phenomena (Fernbach, Rogers, Fox, & cess of polarization, the present work seeks to extend beyond these
Sloman, 2013; Lammers et al., 2017; van Prooijen, Krouwel, & associations and examine how individual differences, such as a
Emmer, 2018). Similarly, extremists are also likely to describe their desire for status, might predict such divides.
beliefs as being objectively correct, regardless of the actual evi- As a hypersocial species, humans are characterized by a number
dence for their beliefs (Toner, Leary, Asher, & Jongman-Sereno, of fundamental social motives that seem to influence behaviors
2013). These same extremists are also more likely to reject outside and beliefs in a wide variety of domains (Neel, Kenrick, White, &
influences on their decision making process (Brandt, Evans, & Neuberg, 2015). One such motive is status seeking (Anderson,
Crawford, 2015). Finally, political extremism is likely associated Hildreth, & Howland, 2015), which refers to the innate human
with troubling outcomes, such as greater intolerance for non- drive to improve one’s rank, station, and relative standing among
likeminded others, greater interpersonal conflict, greater incivility others (Cheng, Tracy, Foulsham, Kingstone, & Henrich, 2013). This
toward ideological others, and greater affective polarization (for drive for status may be seen in a variety of contexts including
reviews, see: Brandt et al., 2014; van Prooijen & Krouwel, 2019). the workplace (Loch, Yaziji, & Langen, 2001), social media environ-
ments (Nesi & Prinstein, 2019), and even leisure pursuits (Walasek
1.1.2. Affective polarization & Brown, 2015). This tendency toward status seeking also seems to
Voluminous bodies of work in social psychology demonstrate manifest in behavior in the context of public discourse (Grubbs,
that people often show strong antipathy toward outgroup mem- Warmke, Tosi, James, & Campbell, 2019). Specifically, some people
bers and strong preference for ingroup members (for reviews, seem particularly prone to use moral talk (i.e., public discussions
see: Hewstone, Rubin, & Willis, 2002; Mackie, Maitner, & Smith, invoking moral rights, moral principles, moral ideals, and moral
2016; Riek, Mania, & Gaertner, 2006; Robbins & Krueger, 2005). practices) as a means to attain higher status or standing in others’
Similarly, groups often share emotions at the group level, particu- eyes. In both philosophical work (Tosi & Warmke, 2016, 2020) and
larly negative emotions in relation to outgroup members (Smith, empirical inquiry (Grubbs et al., 2019), this use of moral talk to
Seger, & Mackie, 2007). Researchers have found such antipathy seek status is known as Moral Grandstanding (hereafter: MG).
across political lines as well. Consistent with previous research, First defined in the philosophical literature (Tosi & Warmke,
we use the term affective polarization to refer to the phenomenon 2016), MG refers to the use of speech (broadly construed) about
of demonstrating extreme positive affect toward members of one’s morality, politics, or other important topics to improve relative sta-
preferred political party and extreme negative affect toward mem- tion or rank. People share their values, morals, or political beliefs in
bers of opposing political parties (Iyengar et al., 2019). That is, a variety of contexts for a variety of reasons. Sometimes these
affective polarization is best understood as the gap between the motivations are altruistic—they aim to help others (e.g., seeking
extremity of one’s affect toward the political outgroup and the to improve the quality of a conversation by providing an insight
extremity of one’s affect toward the one’s political ingroup. Impor- or new information; seeking to ameliorate suffering by defending
tantly, affective polarization seems to function at least somewhat an underdog). Motivations for engaging in public moral discourse
independently of ideological extremity. Even when people might may also be principled—some people simply aim to do what is
be ideologically similar to members of another political party, right (e.g., seeking to stand up for what one takes as the right moral
affective reactions to those partisan others are often quite negative principles; seeking to do one’s duty by telling the truth). However,
(Iyengar et al., 2012). People might experience extreme negative some instances of moral talk are likely motivated by a strong desire
emotions toward political outgroup members independent of to impress others, either by (1) demonstrating one’s impressive
how ideologically different they are from said others (Iyengar moral qualities to one’s in-group or by (2) attacking ideological riv-
et al., 2019). als or opposites for allegedly inferior moral or political views. Prior
There is clear evidence that affective polarization has increased psychological work has validated this notion of MG and found that
over the past several decades (Iyengar et al., 2012; Mason, 2015). the motivation to engage in MG is expressed in ways consistent
This has happened in multiple ways. First, more people have come with either a desire for prestige or a desire for dominance
to hold negative attitudes about those in the other party. For exam- (Grubbs et al., 2019).
ple, over a 50-year time span, the proportion of American adults One way humans seek status is by trying to gain prestige, a kind
who would object to their children marrying a member of the of social status associated with being seen as knowledgeable,
opposing political party rose from 4 to 5% in 1960 to 33–50% in skilled, or impressive in some way (Cheng et al., 2013). Consistent
 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009 3

with this conceptualization of status-seeking more broadly, one as well. This ramping up effect may happen not only with respect
type of MG seems to be characterized by a desire for respect or to one’s ideological views, but also with one’s affective reactions to
admiration from others (e.g., ‘‘My beliefs should be inspiring to the other side, such that individuals in each group coming to hate
others”). This form of MG is associated with desires for prestige the other side more. MG may therefore not only cause ideological
more generally and is well-predicted by narcissistic extraversion extremism but also affective polarization.
(Grubbs et al., 2019). As public discourse comes to resemble a moral arms race, both
MG can also be motivated by a desire for status that results in affective polarization and ideological polarization (i.e., increasing
dominating others. Here, grandstanders shame, hurt, or otherwise amounts of ideological extremism) are predictable consequences.
exploit potential rivals or ideological opponents (e.g., ‘‘When I As such, we might expect to find that MG is linked to more ideolog-
share my beliefs, I do so in the hope that people different than ical extremism and affective polarization, particularly among those
me will feel ashamed of their beliefs”). Very much akin to more who are seeking prestige or respect from likeminded others.
general status-striving via dominance (Cheng et al., 2013), this Consistent with these philosophical insights, preliminary evi-
expression of MG seems to be motivated by a desire to assert one’s dence suggests that MG is associated with more extreme political
rank or power over another person, or to elevate oneself by putting views. MG motivations—particularly in the domain of prestige—
others down. This form of MG is particularly associated with more are associated with ideological extremism, demonstrating a curvi-
aggressive tendencies and is very well-predicted by narcissistic linear relationship with single-item ratings of political ideology
antagonism (Grubbs et al., 2019). (Grubbs et al., 2019). In these preliminary samples of college stu-
Importantly, preliminary empirical examinations of MG have dents and weighted nationally-representative cross-sections in
shown that it is related to a number of potentially concerning the U.S., individuals who identified as being more extreme in their
behaviors and outcomes in public discourse (Grubbs et al., 2019). political views (i.e., more to the ‘‘left” or the ‘‘right” on a sliding
For example, MG is related to greater conflict with others over scale rating of political ideology) also reported a greater desire to
political and moral issues (both cross-sectionally and over time). use moral talk as a means of appearing respectable, admirable, or
MG is also related to greater engagement in antagonistic social inspirational to others. However, to date, no work has systemati-
media behaviors in discussions about politics (e.g., sharing mes- cally examined the nature of these links beyond simple correla-
sages or posts with the sole intent of shaming or embarrassing tional analyses or determined whether these links persist in
the original author of the post). Building on these findings, there measurement environments more rigorous than single-item ideol-
may also be reason to suspect that MG relates to affective polariza- ogy scales. The present work seeks to address these areas of deficit.
tion and ideological extremism, as we explore below.
1.4. The present work
1.3. Moral grandstanding and ideological extremism
Building on the literature reviewed above, the present work
Philosophical work on MG predicts that grandstanders tend to examines the relationships between MG and both ideological
stake out extreme moral and political positions to differentiate extremism and affective polarization in the general public, specif-
themselves from others in search of attention and praise (Tosi & ically testing whether MG predicts greater levels of both. In Study
Warmke, 2016, 2020). The desire for status associated with MG 1, we focused solely on ideological extremism using novel mea-
likely leads people to express increasingly extreme political and sures of self-identified political orientation in general and in rela-
moral opinions as a means of seeming more virtuous, respectable, tion to specific issue/positions. That is, we examined ideological
or worthy of admiration than others in their group. MG leads peo- extremism as the gap between participants self-report of them-
ple to ‘‘ramp up” their ideological positions (or at least their public selves being conservative/traditional or liberal/progressive in gen-
presentations of them) to impress or appear superior to others. As eral and in relation to specific policy domains. In Study 2, we
individuals seek status, they may take progressively more extreme examined both general ideological extremism using traditional
positions in their respective groups as a means of presenting them- measures of ideological extremity and affective polarization using
selves as moral paragons. In short, the motivation to engage in MG widely accepted measures of gaps in partisan affect.
is likely associated with adopting more extreme ideological Given that prestige-motivated aspects of MG are closely related
positions. to a desire to appear morally impressive to likeminded others, it
For example, consider a discussion between likeminded indi- seems likely that individuals who report greater prestige-
viduals, each of whom thinks of herself as caring deeply for the motivated MG would also report more extreme views within their
poor. If someone says that justice demands a $15 per hour mini- own partisan groups. Similar to more seminal works in social psy-
mum wage, another can simply respond that it would be even chology around optimal distinctiveness theory (Brewer, 2011;
more just to institute a $20 per hour minimum wage. In a Leonardelli, Pickett, & Brewer, 2010), the basic thought is that peo-
grandstanding-rich environment, the others in the group now ple who are strongly inclined to use moral discourse to impress
must either accept that this person seems to care more for the poor others will also be inclined to adopt more extreme ideological posi-
or adjust their own stated views to keep up. If enough others in the tions because these views help them ‘‘stand out” among the crowd.
group accept the ramped-up claim, those with doubts or concerns Though such views are seen as extreme by outsiders, they may be
risk being left behind or called out as naysayers. Consistent with seen as more respectable or admirable within their group. Essen-
previously referenced literature on group polarization and social tially, it seems that these individuals will take extreme positions
conformity (e.g., Mackie, 1986), many people find it more palatable in order to garner status and esteem among ideological peers.
to subtly calibrate their views according to their in-group’s plat- Therefore, we expected to find that prestige-motivated MG would
form than to risk apostate status. In short, MG might lead to be associated with greater levels of ideological extremism.
extremism about any given issue. In contrast with the desire for respect and esteem associated
A similar dynamic can play out among a likeminded group in with prestige seeking, dominance-motivated aspects of MG convey
response to the views of an out-group. Ramping up can occur when a desire to assert one’s rank or power over another by causing
group members engage in a competition to see who despises the others shame or feelings of inferiority. These tendencies are inher-
out-group the most, with the winner holding the strongest con- ently related to antipathy toward ideological opponents. As such,
trary view, and thus, the moral high ground. Meanwhile, the we might expect to find that dominance-motivated aspects of
inverse of this competition could happen among the other group MG are related to more extreme affective reactions toward
4 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009

outgroup members. The idea is that people who are strongly 1000 participants. Although participants were recruited based on
inclined to use moral discourse to dominate and humiliate others U.S. nationally representative norms, participants were able to
will be also more inclined to feel and express extreme emotional select the racial or ethnic category that they felt best applied to
responses, since these emotional responses can effectively intimi- themselves (including multiple categories); as such, totals reported
date and instill fear. below may exceed 100%. Participants were primarily White
Although there is very little prior work suggesting that narcis- (73.8%), followed by African-American/Black (11.4%), Asian-
sism might be related to ideological extremism, some recent find- Pacific Islander (8.4%), Latino/a/Hispanic (7.1%), American-Indian
ings have demonstrated that narcissistic antagonism and /Native-American (3.2%), and other (1.6%). Participants primarily
narcissistic extraversion are very strongly related to MG motiva- identified as Democrat (35.7%), followed by Republican (28.9%),
tions and that narcissistic antagonism often predicts conflict with Independent (21.4%), no party affiliation (11.2%), Libertarian
others over political and moral issues (Grubbs et al., 2019). Given (1.8%), and Other (1.1%).
these associations, we included narcissistic antagonism and narcis-
sistic extraversion as covariates in our analyses, allowing us to 2.2. Measures
address the possibility that links between MG and polarization
might be related to the status-oriented personality traits underly- 2.2.1. Moral grandstanding
ing MG rather than MG alone. We included the Moral Grandstanding Motivations Scale
Finally, we note that the present work is a secondary analysis of (Grubbs et al., 2019). This 10-item measure requires participants
previously collected data that were partially described in a prior to respond on a scale of 1 (Strongly Disagree) to 7 (Strongly Agree)
publication (Grubbs et al., 2019). Specifically, this earlier work to items on two subscales. The first, six-item subscale assesses
reported associations between basic personality traits and MG, the extent to which participants are motivated to use moral talk
with a focus on validating the construct and measurement of to seek prestige or respect (e.g., ‘‘My beliefs should be inspiring
MG. However, several of the datasets on which the prior work to others” and ‘‘I hope that my beliefs cause other people to want
reported also included measures of polarization and ideological to share those beliefs”). The resulting subscale score is titled Pres-
extremism that were not included in the earlier publication. tige Strivings. The second, four-item subscale assesses the extent to
Although our analytic plan was theoretically driven by the above which participants are motivated to use moral talk to dominate or
ideas, this work was not pre-registered, and we do not believe that shame others (e.g., ‘‘When I share my beliefs, I do so to show peo-
our results should be interpreted as confirmation of a prespecified ple who disagree with me that I am better than them”). The result-
hypothesis. ing subscale score is titled Dominance Strivings.
More details regarding this entire project, including analytic
code and data, are available via the Open Science Framework at 2.2.2. Ideological extremism
https://osf.io/3mpx2/. For these samples, we assessed ideological extremism via a
novel measure of absolute ideological gap. That is, we measured
the absolute magnitude of the gap between participants’ self-
2. Study 1
reported endorsement of Conservative/Traditional ideology and
Liberal/Progressive ideology, so that a greater gap value repre-
To initially test the above ideas and examine the links between
sented more polarized views of various political domains.
MG and polarization, we collected two samples, as detailed below.
Political views were assessed in five domains: overall political
views (prompt: ‘‘To what extent do the following items describe
2.1. Participants your political views?”), fiscal views (prompt: ‘‘To what extent are
your fiscal (money, budget, tax, and finance) views politically:”),
Participants for Sample 1 were undergraduates at a large, public social views (prompt: ‘‘To what extent are your social (welfare,
university in the Midwest. Participants (N = 981; Mage = 19.4; entitlements, social-security, and education) views politically:”),
SD = 2.1; 69.7% women) were recruited from undergraduate psy- moral views (prompt: ‘‘To what extent are your moral (marriage,
chology classes throughout the university over the course of four sexuality, abortion/reproductive-rights) views politically:”), and
semesters (Fall 2017, Spring, 2018, Fall 2018, and Spring 2019). foreign-policy views (prompt: ‘‘To what extent are your foreign
Sample size was determined by maximizing participation over four policy views (military and war; U.S. involvement overseas; U.S.
semesters, rather than by a priori power analyses. Participants involvement in nation building) politically:”).
were able to select the racial or ethnic category that they felt best For each of the above categories, we asked participants to rate
applied to themselves (including multiple categories); as such, the extent to which their political views were ‘‘Conservative,” ‘‘Lib-
totals reported below may exceed 100%. Participants primarily eral,” ‘‘Progressive,” or ‘‘Traditional” on a seven-point scale of 0
identified as White (84%), followed by African-American/Black (Not at all) to 6 (Extremely). That is, we asked participants to
(12.9%), Latino/a/Hispanic (3.6%), Asian/South-Asian/Pacific-Islan respond to all four descriptors (‘‘Conservative,” ‘‘Liberal,” ‘‘Progres-
der (2.1%), American Indian/Native-American (1.0%), and other sive,” and ‘‘Traditional”) for all five domains.1
(1.5%). Political affiliations reported by participants were as fol- We calculated our index of ideological extremism using the fol-
lows: Democrat (31%), Republican (26.4%), no party affiliation lowing formula:
(27.1%); Libertarian (5.0%), Green (0.8%), Independent (8.6%), and qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðMeanðConser v ativ e þ TraditionalÞ  MeanðLiberal þ Progressiv eÞÞ
2
other (1.1%).
Participants for Sample 2 were internet-using adults in the Uni-
ted States recruited by the Turkprime data-acquisition platform
(now known as Cloud Research; Litman, Robinson, & Abberbock, 1
Given the novelty of this approach, we tested the general validity of responses by
2017). Participants (N = 1063; Mage = 48.20, SD = 16.38; 49.8% examining how participants in each sample differed on raw ratings in each domain.
women) were recruited in the Summer of 2018 based on U.S. Specifically, we compared self-identified Democrats and self-identified Republicans
demographic norms as of the 2010 census for Race, Ethnicity, within each sample on each of the items described above. For all domains, self-
identified Democrats scored substantially higher than self-identified Republicans for
Income, Census Region, Age, and Gender. Sample size was deter- ratings of Liberal and Progressive and substantially lower than self-identified
mined by funding limits, rather than by a priori power analyses. Republicans for ratings of Conservative and Traditional. These analyses are available
More simply, budgetary constraints led us to recruit a target of via the OSF page for this project at https://osf.io/jgecm/ and https://osf.io/85mec/.
 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009 5

Within each domain, we averaged participant ratings of Conser- tively associated with polarization measures. Finally, in both
vative and Traditional responses options, then subtracted from this samples, the intercorrelations of the various polarization indices
value the average participant ratings for Progressive and Liberal were large and positive (i.e., r = 0.653–0.821).
response options. This process created an index with a possible With regard to our planned SEM analyses: Our specified models
range of 6 to +6, wherein a more liberal ideological gap was rep- demonstrated good fit in both samples (Robust Fit Indices; Sample
resented by lower values and a more conservative ideological gap 1:v2[1 1 1] = 297.99, CFI = 0.975, TLI = 0.969, RMSEA = 0.041,
was represented by more positive values. We then squared these SRMR = 0.044; Sample 2: v2[1 1 1] = 311.78, CFI = 0.977,
indices and took the square root of the resulting values to create TLI = 971, RMSEA = 0.046, SRMR = 0.035).2 In both samples, FFNI
a series of indicators of unidirectional ideological extremism so Antagonism was unrelated to MG: Prestige. Similarly, in both sam-
that more extreme values (i.e., values closer to |6|) were indicative ples, FFNI Antagonism demonstrated strong positive associations
of higher ideological gap and less extreme values (i.e., values closer with MG: Dominance, and FFNI Extraversion demonstrated strong
to zero) were indicative of lower ideological gap. In sum, this cre- positive associations with MG: Prestige. In turn, MG: Prestige demon-
ated an index of absolute extremism in each domain, regardless of strated robust associations with Ideological extremism, whereas the
direction. relationship between MG: Dominance and Ideological extremism did
not reach statistical significance. Moreover, in both samples, the
2.2.3. Narcissism total variance accounted for in Ideological extremism was similar
Given the robust associations between narcissistic traits and (Sample 1, R2 = 0.110; Sample 2, R2 = 0.112).
MG in prior works, we included the Five Factor Narcissism Inven- We also considered an alternative interpretation of our data in
tory Short Form in our analyses as a covariate (Sherman et al., which extremism may be understood as predicting MG. That is, we
2015). The FFNI-SF measures numerous domains of narcissistic tested alternatives to our structural model (where MG predicts
personality along several facets. We focused on two composite extremism) rather than assuming that ours is the most plausible
indices that have previously shown robust associations with MG: interpretation of our data. To accomplish this, in both samples,
Narcissistic Antagonism (exploitative, aggressive, and entitled we compared the above described model to an alternative model
aspects of narcissism that are strongly associated with MG: Dom- in which MG: Dominance and MG: Prestige were regressed on Ideo-
inance Strivings) and Narcissistic Extraversion (grandiose and logical extremism (i.e., where Ideological extremism predicted MG:
self-important aspects of narcissism that are strongly associated Dominance and MG: Prestige) rather than the converse. However,
with MG: Prestige Strivings). such a model demonstrated a noticeably inferior fit in both sam-
ples (Sample 1, alternative model fit: v2[1 1 3] = 333.87,
CFI = 0.971, TLI = 0.965, RMSEA = 0.048, SRMR = 0.059; v2dif
2.3. Analytic plan
[2] = 40.316, p < .001; Sample 2, alternative model fit:
v2[1 1 3] = 343.41, CFI = 0.973, TLI = 0.968, RMSEA = 0.048,
In both samples, we computed full descriptive statistics, analy-
SRMR = 0.060; v2dif [2] = 35.267, p < .001). That is, our theoretically
sis of internal consistency (Cronbach’s a and McDonald’s x), and
expected models (See Fig. 1), are more consistent with the data
simple product-moment correlations between all key variables.
obtained than an alternative model wherein polarization may be
These variables are all reported in Table 1.
seen as predicting grandstanding.
Subsequent to these analyses, we conducted a Structural Equa-
tion Model (SEM) in each sample using the lavaan package
(Rosseel, 2012) for R Statistical Software (R Core Team, 2018). All 3.1. Study 1 Discussion
analyses were conducted using maximum likelihood estimation
with robust (Huber-White) standard errors and a scaled test statis- Results of our first study demonstrated that, in two large sam-
tic. In these analyses, for each sample, we defined the latent vari- ples, greater motivation to engage in moral grandstanding for
ables MG: Prestige and MG: Dominance by the items of the prestige-based reasons was consistently related to greater ideolog-
corresponding subscales of the MGMS. We also defined the latent ical extremism. This finding persisted even when narcissistic traits
variable Ideological extremism by the five absolute indices of parti- were held constant statistically, suggesting that prestige motivated
san ideological gap (Overall, Fiscal, Social, Moral, and Foreign Pol- MG is uniquely related to extremism, beyond its links to narcis-
icy) described above. In the full SEM, MG: Prestige and MG: sism. Moreover, consistent with our predictions, our specified
Dominance were regressed on observed exogenous variables FFNI: model (where MG predicted extremism) demonstrated better fit
Antagonism and FFNI: Extraversion. In turn, Ideological extremism than an alternative model where extremism predicted MG. Of note,
was regressed on MG: Prestige, MG: Dominance, FFNI: Antagonism, we found consistently negative links between dominance moti-
and FFNI: Extraversion. MG: Prestige and MG: Dominance were vated MG and extremism when narcissistic antagonism was not
allowed to covary. Similarly, FFNI: Antagonism and FFNI: Extraver- included, and insignificant links between dominance and extrem-
sion were allowed to covary. This model is summarized in Fig. 1. ism when narcissism was included. These findings were contrary
to our initial hypotheses and suggest that, to the extent MG: Dom-
3. Study 1 Results inance is related to ideological extremism, it is likely a function of
the close relationship between narcissistic antagonism and
In both samples, Pearson correlations revealed small to moder- dominance.
ate associations between MG motivations and ideological extrem- Despite our findings regarding prestige motivated MG, these
ism across all indicators (See Tables 1 and 2). In Sample 1, for MG: results demonstrated only preliminary links between MG and
Prestige these associations were most often positive and moderate partisan divides, for many reasons. For example, these results did
in size (i.e., r > 0.2; Funder & Ozer, 2019). In contrast, associations
for MG Dominance and ideological gaps were consistently small 2
We also conducted these analyses without a latent ideological extremism variable
and negative. In Sample 2, a similar pattern was observed, with (i.e., with just observed gap indices) and another model without narcissism as a
small positive associations (i.e., r < 0.1; Funder & Ozer, 2019) being covariate. These results are available via the Open Science Framework at https://osf.
io/68wy9/ as Supplemental Fig. 1 and Supplemental Table 4. Of note, patterns of
observed between MG Prestige and polarization measures, and association for MG: Prestige remained highly similar across all alternate analyses,
small negative associations between MG Dominance and polariza- though MG: Dominance demonstrated significant negative associations with ideolog-
tion measures. In both samples, narcissistic antagonism was nega- ical extremism in models where narcissism was not included.
6 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009

Table 1
Study 1, Sample 1: Descriptive Statistics, Internal Consistency Coefficients, and Correlations.

MG MG Narc. Narc. Fiscal Social Moral Foreign Overall

Prestige Dominance Antagonism Extraversion Gap Gap Gap Gap Gap
Sample 1 MG Prestige M = 4.20
(N = 981) SD = 1.11
a = 0.815
x = 0.818
MG M = 2.24 0.114**
Dominance SD = 1.28
a = 0.876
x = 0.880
Narc. M = 2.33 0.054 0.510**
Antagonism SD = 0.59
a = 0.835
x = 0.845
Narc. M = 3.33 0.276** 0.071* 0.323**
Extraversion SD = 0.62
a = 0.640
x = 0.657
Fiscal Gap M = 1.65 0.215** 0.076* 0.120** 0.071*
SD = 1.70
Social Gap M = 1.92 0.216** 0.111** 0.172** 0.057 0.759**
SD = 1.92
Moral Gap M = 2.39 0.200** 0.141** 0.175** 0.036 0.653** 0.752**
SD = 2.16
Foreign Gap M = 1.75 0.196** 0.078* 0.137** 0.027 0.762** 0.762** 0.685**
SD = 1.89
Overall Gap M = 1.89 0.209** 0.084** 0.172** 0.010 0.774** 0.821** 0.775** 0.769**
SD = 1.77

a = Cronbach’s Alpha; x = McDonald’s Omega.

MG = Moral Grandstanding; Narc = Narcissism.
**
Correlation is significant at the 0.01 level (2-tailed).
*
Correlation is significant at the 0.05 level (2-tailed).

Fig. 1. Structural Equation Models for Study 1, Samples 1 and 2, Demonstrating the Effect of Moral Grandstanding on Ideological Extremism. Sample 1 estimates are before
the slash and Sample 2 estimates are after the slash. Directional arrows represent standardized path estimates; Bi-directional arrows represent standardized covariances.
Item loadings for Moral Grandstanding Motivations Scale items are omitted from the diagram for the sake of clarity but available at https://osf.io/q2egp/. FFNI = Five Factor
Narcissism Inventory. MG = Moral Grandstanding. CFI = Comparative Fit Index; TLI = Tucker-Lewis Index; RMSEA = Root Mean Square Error of Approximation;
SRMR = Standardized Root Mean Square Residual.

not address affective polarization, were based on novel and that a series of questions related to moral and political beliefs may
idiosyncratic measures of ideological extremism, and did not con- be indirectly measuring political interest rather than status seeking
trol for general political interest. Given that political polarization via contributions to moral and political discourse. More simply, it
implies some measure of political and news interest, it is possible may be that links between MG and political polarization are better
 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009 7

Table 2
Study 1, Sample 2: Descriptive Statistics, Internal Consistency Coefficients, and Correlations.

MG MG Narc. Narc. Fiscal Social Moral Foreign

Prestige Dominance Antagonism Extraversion Gap Gap Gap Gap
Sample 2 (N = 1063) MG Prestige M = 4.30
SD = 1.25
a = 0.843
x = 0.844

MG Dominance M = 2.41 0.263**

SD = 1.44
a = 0.912
x = 0.912

Narc. Antagonism M = 2.24 0.142**

SD = 0.63 0.599**
a = 0.852
x = 0.858

Narc. Extraversion M = 2.96 0.389** 0.463**

SD = 0.72 0.333**
a = 0.733
x = 0.739

Fiscal Gap M = 2.00 0.105** 0.126** 0.095**

SD = 1.84 0.070*

Social Gap M = 2.19 0.129** 0.113** 0.168** 0.112** 0.725**

SD = 1.92
Moral Gap M = 2.54 0.093** 0.179** 0.206** 0.132** 0.653** 0.772**
SD = 2.07
Foreign Gap M = 1.92 0.129** 0.073* 0.141** 0.081** 0.757** 0.760** 0.696**
SD = 1.90
Overall Gap M = 2.27 0.131** 0.073* 0.180** 0.114** 0.743** 0.803** 0.730** 0.747**
SD = 1.85

a = Cronbach’s Alpha; x = McDonald’s Omega.

MG = Moral Grandstanding; Narc = Narcissism.
**
Correlation is significant at the 0.01 level (2-tailed).
*
Correlation is significant at the 0.05 level (2-tailed).

explained by how politically engaged a person is, rather than a true However, we assessed ideological extremism differently than in
relationship between engaging in MG and adopting more extreme Study 1 and also assessed affective polarization as described below.
views. We sought to address these possible limitations in Study 2.
4.3. Ideological extremism
4. Study 2
We asked participants to rate their political beliefs from ‘‘left”
to ‘‘right” on a sliding scale with an anchor of 100 (liberal or pro-
To extend the findings from Study 1 to a weighted, nationally
gressive) on the left and +100 (conservative or traditional) on the
representative sample, to account for general political interest, to
right. We then folded this measure, keeping the absolute value of
measure ideological extremism using a more traditional measure,
one’s self-rating, with more extreme values signifying polarization,
and to extend our models to affective polarization as well, we col-
regardless of the direction of such polarization. This resulting index
lected a cross-sectional sample of the U.S. population, as described
was termed Ideological Extremism: Left-Right Spectrum.
below.
We also included panel data provided by YouGov on the
strength of political ideology. This data was collected prior to the
4.1. Participants administration of our study and is a standard part of the political
information maintained by YouGov as a part of all panel-based
Participants were adults in the U.S. (N = 2519; Mage = 47.5, research they facilitate. For this item, participants responded to
SD = 17.8; 51.4% women) recruited via YouGov polling and the item, ‘‘In general, how would you describe your own political
matched to nationally representative (2016 American Community viewpoint?” on a scale of 1 (very liberal) to 5 (very conservative).
Survey) norms for age, gender, race, education, and U.S. Census For this index of polarization, consistent with prior research
Region. Participants identified primarily as 38.7% Democrat, (Brandt et al., 2015; Garrett & Bankert, 2018; Wegener, Downing,
27.2% Republican, 25.4% Independent, 3.6% other, and 5.1% not Krosnick, & Petty, 1995), we folded this scale at the midpoint, with
sure. Race was reported as follows: 64.1% White, 12.0% Black, the highest score (2) representing very liberal or very conservative
15.7% Hispanic, 3.3% Asian, 0.9% Native American, 2.5% Mixed, and the lowest score (0) indicating moderate ideology. In keeping
1.5% other, 0.2% Middle Eastern. Responses were weighted using with prior literature on this topic, the resulting index was titled:
YouGov’s proprietary weighting formulae (Rivers, 2016). Ideological Extremism: Panel Ideology.

4.2. Measures 4.4. Affective polarization

We measured moral grandstanding motivations and narcissistic We assessed affective polarization via two methods. First, we
extraversion/antagonism using the same measures as Study 1. used a gap index based on a feeling thermometer wherein partici-
8 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009

pants were asked to rate their feelings toward individuals with 5. Results
specific political affiliations on a scale of 0 (very cold or completely
negative) to 100 (very warm or completely positive). Participants Pearson correlations revealed highly similar results to those
were asked to rate their feelings toward Democrats and Republi- found in Study 1 (See Table 3). Specifically, we consistently found
cans. Consistent with recent works examining gap in partisan positive associations between computed indices of partisan polar-
affect (Garrett & Bankert, 2018; Garrett, Long, & Jeong, 2019; ization (both ideological and affective) and MG motivations, partic-
Iyengar & Westwood, 2015), we looked at the gap in feelings ularly prestige motivations, though these associations were
toward each party by subtracting feelings toward Republicans typically small to moderate in size (r = 0.1–0.2; Funder & Ozer,
from feelings toward Democrats. This created a variable on a scale 2019). See Table 2.
of 100 to +100. We then squared these results and then took the In our subsequent SEM analysis, we found similar patterns to
square root of these squares so that all answers were on a scale of those observed in Study 1. The fit of our model was good (Robust
0–100. By doing so, we created a linear absolute-value index of Fit Indices: CFI = 0.977, TLI = 0.966, RMSEA = 0.043, SRMR = 0.038).
polarization wherein more extreme affective polarization was rep- These results are summarized in Fig. 2. Specifically, consistent with
resented by higher values and lower levels of affective polarization Study 1, we saw strong positive associations between FFNI: Antag-
were represented by lower values. The resulting variable was onism and MG: Dominance and between FFNI: Extraversion: and
called Affective Polarization: Feeling Thermometer Gap. MG: Prestige. Additionally, analyses demonstrated positive associa-
Consistent with prior work in political science (i.e., Iyengar tions between general political interest and both MG: Dominance
et al., 2012), we also asked participants how they would feel about and MG: Prestige.
a child or other close loved one married a member of certain polit- With regard to our outcome variables, results were generally
ical parties (Republicans and Democrats). Participants rated their consistent with Study 1. General interest in politics was positively
reactions to the following prompt: ‘‘How would you feel if your associated with all forms of polarization, as was MG: Prestige. MG:
child married someone who identified as: ______________ (note: Dominance was only positively associated with Ideological
if you do not have children, you can imagine how you feel if a close Strength and affective polarization in the form of feelings about
friend or family member married someone who identified as the children’s marriages. Both FFNI Antagonism and FFNI Extraversion
following descriptors)” on a scale of 0 (completely negative) to demonstrated sporadic relationships with polarization, either
100 (completely positive). For this variable, we followed a similar being negatively related to polarization or not significantly related
technique as we did above. We created an affective gap index by at all. Variance accounted for in each outcome variable by our
subtracting feelings toward Republicans from feelings toward model ranged from R2 = 0.092 for ideological identification to
Democrats, squaring this resulting gap index, then taking the R2 = 0.240 for affective polarization as indicated by feeling ther-
square root of the squared value to create a single, linear measure mometer responses.
of polarization. The resulting variable was called Affective Polariza- Consistent with Study 1, we also tested alternative SEMs and
tion: Child’s Marriage Gap. compared them to our preferred model reported above. The first
alternative model we tested followed a similar pattern to alterna-
tive models in Study 1, wherein MG variables were treated as out-
4.5. Political interest comes predicted by all other variables. In this alternative model,
MG: Prestige and MG: Dominance were regressed on observed
In addition to the above measures of polarization, we also ana- polarization and extremism indices. This model demonstrated
lyzed panel data from YouGov regarding participants’ interest in much worse fit than the above reported model (Robust Fit Indices:
politics using a standard item from the American National Election CFI = 0.768, TLI = 0.671, RMSEA = 0.105, SRMR = 0.096;
Study (ANES). YouGov maintains participant responses to the stan- v2dif[6] = 2089.3, p < .001). Additionally, we specified another alter-
dard prompt: ‘‘Some people seem to follow what’s going on in gov- native model wherein we defined a latent variable, General Polar-
ernment and public affairs most of the time, whether there’s an ization using all observed polarization indices as indicators. This
election going on or not. Others aren’t that interested. Would you model also demonstrated worse fit than the above described model
say you follow what’s going on in government and public affairs. . .” wherein each observed polarization index was treated as an inde-
Responses were recorded on a scale of 1 (Most of the time) to 4 pendent outcome (Robust Fit Indices: CFI = 0.953, TLI = 0.941,
(Hardly at all). For ease of interpretation, we reversed the scoring RMSEA = 0.057, SRMR = 0.045; v2dif[17] = 318.6, p < .001). Finally,
of this item so that greater values indicated more interest in poli- we specified an alternative model where Ideological Extremism:
tics. This resulting score was termed Political Interest. Panel Ideology was included as an exogenous covariate in the
model, predicting both MG: Prestige and MG: Dominance, as well
as the remaining three measures (Ideological Extremism: Left/
4.6. Analytic plan Right Spectrum; Affective Polarization: Feeling Thermometer;
and Affective Polarization: Child’s Marriage Gap). In testing this
All analyses were weighted using population weights provided model, we found adequate fit (CFI = 0.975, TLI = 0.963,
by YouGov. Analyses for Study 2 were very similar to Study 1. We RMSEA = 0.045, SRMR = 0.040), but again found that our
conducted Pearson product-moment correlations between all vari- original model demonstrated better fit by a significant margin
ables. We followed these correlations with an SEM analyses in (v2dif [2] = 15.383, p < .001).
lavaan similar to those reported in Study 1. Specifically, we defined
the latent variables MG Prestige and MG Dominance by the respec-
tive subscale items of the MGMS. These variables were again
regressed on observed variables FFNI Antagonism and FFNI 5.1. General discussion
Extraversion, as well as the observed variable Political Interest. In
turn, Affective Polarization: Feeling Thermometer Gap, Affective Past work in philosophy has argued that MG has a polarizing
Polarization: Child’s Marriage Gap, Ideological extremism: Left- effect in public discourse. We evaluated this possibility in three
Right Spectrum, and Ideological Extremism: Panel Ideology were samples, examining both ideological extremism and affective
all regressed on the aforementioned variables described above. polarization. Below we summarize our findings and discuss the
This model is summarized in Fig. 2. implications of the present work.
 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009 9

Fig. 2. Structural Equation Model in Study 2 Predicting Polarization and Extremism Indicators. Directional arrows represent standardized path estimates; Bi-directional
arrows represent standardized covariances. Item loadings for Moral Grandstanding Motivations Scale items are omitted from the diagram for the sake of clarity but available
at https://osf.io/q2egp/. FFNI = Five Factor Narcissism Inventory. MG = Moral Grandstanding. CFI = Comparative Fit Index; TLI = Tucker-Lewis Index; RMSEA = Root Mean
Square Error of Approximation; SRMR = Standardized Root Mean Square Residual. Robust Fit Indices: CFI = 0.977, TLI = 0.966, RMSEA = 0.043, SRMR = 0.038.

Table 3
Study 2: Descriptive Statistics, Internal Consistency Coefficients, and Correlations.

M (SD) a x 1 2 3 4 5 6 7 8
1. Moral Grandstanding: Prestige Strivings 4.86 (0.99) 0.767 0.770
2. Moral Grandstanding: Dominance Strivings 2.33 (1.45) 0.908 0.909 0.163**
3. Narc. Antagonism 2.24 (0.70) 0.877 0.881 0.162** 0.663**
4. Narc. Extraversion 2.92 (0.74) 0.741 0.744 0.407** 0.303** 0.568**
5. Ideological Extremism: Panel Ideology 0.95 (0.80) – – 0.126** 0.009 0.079** 0.049*
6. Ideological extremism: Left-Right Spectrum 56.31 – – 0.132** 0.057** 0.149** 0.084** 0.587**
(35.49)
7. Political Interest 3.20 (0.09) – – 0.154** 0.125** 0.206** 0.021 0.262** 0.328** .
8. Affective Polarization: Feeling Thermometer 44.42 – – 0.121** 0.155** 0.296** 0.150** 0.456** 0.619** 0.415**
Gap (30.22)
9 Affective Polarization: Child’s Marriage Gap 34.14 – – 0.128** 0.064** 0.199** 0.091** 0.445** 0.560** 0.337** 0.775**
(31.20)

a = Cronbach’s Alpha; x = McDonald’s Omega.

MG = Moral Grandstanding; Narc = Narcissism.
**
Correlation is significant at the 0.01 level (2-tailed).

We consistently found robust evidence that prestige-motivated for this notion and some support for its opposite. That is, in Study
aspects of MG were positively related to affective polarization and 1, we found negative correlations between MG Dominance and
ideological extremism in all ways in which we measured it. This is ideological extremism. Moreover, in Study 2, raw correlations
consistent with preliminary links between prestige-motivated MG observed between dominance-motivated MG and affective polar-
and single-item measures of ideological extremism (Grubbs et al., ization were reduced to insignificance in latent variable analyses.
2019). That is, prestige-motivated MG was a uniformly positive In our weighted, nationally-representative sample, dominance-
associate of more extreme political ideology across single-item motivated MG was not significantly related to any of our measures
ideology measures, affective gap measures, and multi-item ideo- of affective polarization.
logical gap measures. These links persisted even when general The reasons for the lack of support for our hypotheses regarding
political interest and involvement were statistically controlled, dominance-motivated MG are unclear. In all three samples, the
and testing of alternative models suggest that our conceptualized raw associations between dominance-motivated MG and narcissis-
pathways (i.e., that polarization and extremism are predicted by tic antagonism were quite large (r > 0.5), and the magnitude of
prestige-motivated MG) were more consistent with the data than these relationships appeared even larger in latent variable analy-
alternative models (i.e., that polarization or extremism predicts ses. Additionally, across samples, narcissistic antagonism was most
prestige-motivated MG). often negatively related to polarization and political interest. As
Though we expected that dominance-motivated aspects of MG such, it is possible that any relationship between dominance-
would be predictive of affective polarization, we found no support motivated MG and polarization is overshadowed by the negative
10 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009

links between narcissistic antagonism and such polarization, and the other hand, often attempt to appear more morally impressive
that any residual relationships in structural models are a function or enlightened. These judgments are typically comparative (one is
of suppression effects. Alternatively, and perhaps more simply, the trying to look morally better than others) and contextual (one is
relationship between prestige-motivated MG and polarization may using a particular issue or discursive environment to carry out this
overshadow any possible connection between dominance- task). Thus, we hypothesize that attempting to enhance appear-
motivated MG and polarization. This would suggest that simple ances by prestige-MG suggests a more adroit mechanism (i.e.,
correlations between MG Dominance and affective polarization shifting one’s moral or political views to a more extreme position)
(as seen in study 2) may be a function of the shared variance to make oneself morally impressive within a relevant network. In
between prestige-motivated MG and dominance-motivated MG short, we suspect that built into the very nature of prestige-
such that dominance-motivated MG is unrelated to polarization striving is a mechanism for ideological polarization which leads
when controlling for prestige-motivated MG. to greater extremism, and that with respect to dominance-
striving, this mechanism is either not present, rarely activated, or
5.2. Implications and future directions overridden by some other process more efficient for attaining
dominance.
Links between MG and polarization are not evenly distributed. A similar dynamic could be at play concerning affective polar-
Although prior empirical work has conceptualized MG as being ization. One need not take up highly polarized emotions to abuse
expressed in terms of prestige strivings (i.e., a desire to be others via dominance grandstanding. For those motivated by pres-
respected, admired, or emulated for one’s moral beliefs) and dom- tige, however, being seen as especially warm toward or approving
inance strivings (i.e., a desire to shame or embarrass those that dis- of one’s in-group, and especially cold toward or disapproving of
agree with one’s moral beliefs), only the former of these categories one’s out-group, could promote one’s status within one’s in-
is consistently related to affective polarization or more extreme group. Thus, affective polarization also seems more useful as an
ideological positions. This point is particularly fascinating given avenue for seeking prestige than it does for seeking dominance.
the established links between prestige strivings and other politi- On the other hand, however, highly polarized emotions would
cally relevant domains, which we explore below. seem to facilitate dominance grandstanding, too. Hatred and
Prior works have linked the prestige-motived aspects of MG to resentment of partisan others appear to be natural motivations
many generally positive outcomes. For example, both cross- outlets for status-seeking via dominance behaviors. We admit that
sectionally and over time, prestige-motivated MG is related to the lack of closeness between affective polarization and dominance
greater civic engagement in the form of voting behavior, volun- grandstanding is puzzling in light of the apparent similarities.
teerism, and philanthropy (Grubbs, Tosi & Warmke, 2020). Simi- The present work also may inform future work related to group
larly, both cross-sectionally and over time, prestige motivated polarization more generally as well as intergroup relations.
MG is related to reports of growing closer to others over political Although the current work focuses on the role that individual dif-
and moral issues (Grubbs et al., 2019). In contrast, dominance- ferences might play in promoting more extreme ideology and
oriented aspects of MG are associated with a reduced likelihood affective reactions to ideological others, the current results also
of voting and greater reports of conflict and discord over political suggest that attempts to seek prestige may also influence inter-
issues (Grubbs et al., 2019, 2020). Yet, the present results seem group relations. Desires to be seen as inspirational and admirable
to run somewhat counter to these previous works by demonstrat- for one’s moral views (i.e., prestige-motivated grandstanding) con-
ing that prestige-motivated MG also seems to be associated with sistently predicted not only more extreme ideological positions but
adverse outcomes for civic life. The present work consistently also more extreme affective responses to others. Accordingly, it
shows that more extreme political views, affective reactions to seems that status seeking via prestige within one’s own group
political views, and partisan ideology are related to prestige- might ultimately predict more volatile and negative evaluations
motivated MG. of members of the outgroup.
What follows from these findings is a more nuanced under-
standing of how MG might be related to public discourse and polit- 6. Limitations
ical polarization more broadly. Specifically, we can consider that
prestige-motivated MG might be associated with putatively posi- The primary limitations of the present work are obvious. We
tive outcomes such as greater civic engagement while also indicat- relied exclusively on self-report data using cross-sectional samples
ing more extreme political views. This is not surprising as more in a purely U.S. context. As such, we urge caution before interpret-
extreme partisans are more politically active (McCarty, 2019). Sim- ing the present findings as being representative of anything more
ilarly, although prestige-motivated MG may predict growing closer than associations between related domains in the American cul-
to others over political issues, such relationship-strengthening tural context. Although our sampling methods were rigorous and
does not preclude the possibility of polarization. Prestige- representative, future longitudinal work is needed to determine
motivated MG may simply be associated with stronger relation- if prestige-motivated aspects of MG are causing polarization or
ships to polarized others (i.e., polarized people bond with others driving some parts of polarization. Given the cross-sectional nature
over shared extreme views). In short, prestige-motivated MG of our work, we acknowledge that another plausible interpretation
may be related to seemingly good things while also facilitating of our data is that various aspects of polarization are driving
polarization and extremism, which are known to drive a range of prestige-motivated MG. However, given our tests of such alterna-
negative outcomes at societal levels (McCarty, 2019) and in family tive models, we contend that this is less likely than our proposed
relationships (Chen & Rohla, 2018). model positing that prestige-motivated MG leads to polarization.
Regarding the weaker relationship between dominance- We also note that ideological extremism is likely to occur with
motivated MG and extremism, we suspect that this is due to inher- respect to specific issues, such as the morality of abortion, mini-
ent differences between dominance and prestige approaches to mum wage laws, immigration, or tax rates. Although we evaluated
status seeking. To achieve status, dominance grandstanders are ideological extremism within the context of specific domains in
less likely to employ a kind of moral one-upmanship than they Study 1 (e.g., fiscal policy), we did not examine specific issues
are to simply call names or use harsh, condemning language. themselves (e.g., tax law). MG may lead individuals to take up
Crucially, one needs not shift one’s moral or political views to increasingly extreme views or attitudes on specific issues, particu-
assert dominance in this way. Prestige-seeking grandstanders, on larly those that are deemed especially important to one’s social
 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009 11

group. Thus, we would expect MG to be associated with the adop- experimenter-generated anchors. Psychological Science, 26(2), 189–202.
https://doi.org/10.1177/0956797614559730.
tion of more extreme views on issues that are likely to confer social
Brandt, M. J., Reyna, C., Chambers, J. R., Crawford, J. T., & Wetherell, G. (2014). The
status within one’s network. This presents a need for both longitu- ideological-conflict hypothesis: Intolerance among both liberals and
dinal research examining MG and polarization broadly and future conservatives. Current Directions in Psychological Science, 23(1), 27–34. https://
work examining how MG might lead to extremism along specific doi.org/10.1177/0963721413510932.
Brewer, M. B. (2011). Optimal distinctiveness theory: Its history and development.
moral and political issues over time. P. AM Van Lange, A. Kruglanski, & ET Higgins, Handbook of Theories of Social
Finally, we note that a number of potentially important covari- Psychology, 2, 81–98.
ates were not included in our study. For example, we did not assess Chen, M. K., & Rohla, R. (2018). The effect of partisanship and political advertising
on close family ties. Science, 360(6392), 1020–1024. https://doi.org/
general political knowledge, which could impact our findings as 10.1126/science.aaq1433.
prior works have shown that political knowledge is often associ- Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., & Henrich, J. (2013). Two ways to
ated with greater ideological extremism (Bakker, Lelkes, & Malka, the top: Evidence that dominance and prestige are distinct yet viable avenues to
social rank and influence. Journal of Personality and Social Psychology, 104(1),
2020; Drummond & Fischhoff, 2017; Guay & Johnston, 2020; 103–125. https://doi.org/10.1037/a0030398.
Zaller, 1992). Similarly, we did not assess general belief superiority Crockett, M. J. (2017). Moral outrage in the digital age. Nature Human Behaviour, 1
(i.e., the notion that one has better beliefs or ideas than others). (11), 769. https://doi.org/10.1038/s41562-017-0213-3.
Dimock, M., Doherty, C., Kiley, J., & Oates, R. (2014). Political Polarization in the
This may be particularly relevant given literature suggesting ideo- American Public (p. 124).
logical extremism is indeed related to belief superiority (Toner Divided States of America (2017). In Frontline. Public Broadcasting Corporation.
et al., 2013) and prestige motivated MG is characterized by a desire Doherty, C., Kiley, J., Tyson, A., & Johnson, B. (2019). Public Highly Critical of State of
Political Discourse in the U.S. (p. 104).
to be seen as special or admirable for one’s beliefs.
Druckman, J. N., & Levendusky, M. S. (2019). What do we measure when we
measure affective polarization?. Public Opinion Quarterly, 83(1), 114–122.
https://doi.org/10.1093/poq/nfz003.
7. Conclusions
Drummond, C., & Fischhoff, B. (2017). Individuals with greater science literacy and
education have more polarized beliefs on controversial science topics. In
Across three samples, involving undergraduate students and Proceedings of the National Academy of Sciences. https://doi.org/10.1073/
pnas.1704882114.
nationally-representative U.S. cross-sections, we found consistent
Fernbach, P. M., Rogers, T., Fox, C. R., & Sloman, S. A. (2013). Political extremism is
support for the notion that that moral grandstanding is linked to supported by an illusion of understanding. Psychological Science, 24(6),
ideological extremism and affective polarization. These findings 939–946. https://doi.org/10.1177/0956797612464058.
are particularly clear for prestige-motivated grandstanding, and Fiorina, M. P., & Abrams, S. J. (2008). Political polarization in the American public.
Annual Review of Political Science, 11(1), 563–588. https://doi.org/10.1146/
less so for dominance motivated grandstanding. Collectively, our annurev.polisci.11.053106.153836.
findings point to the importance of considering status seeking as Funder, D. C., & Ozer, D. J. (2019). Evaluating effect size in psychological research:
an associate of polarization—both ideological and affective--and Sense and nonsense. Advances in Methods and Practices in Psychological
Science, 251524591984720. https://doi.org/10.1177/2515245919847202.
indicate a need for future research examining whether status seek- Garrett, K. N., & Bankert, A. (2018). The moral roots of partisan division: How moral
ing predicts increases in polarization over time or in specific issue conviction heightens affective polarization. British Journal of Political Science,
domains. 1–20. https://doi.org/10.1017/S000712341700059X.
Garrett, R. K., Long, J. A., & Jeong, M. S. (2019). New evidence on group polarization
from partisan media to misperception: Affective polarization as mediator.
Declaration of Competing Interest Journal of Communication, 69(5), 490–517. https://doi.org/10.1093/joc/jqz028.
Grubbs, J. B., Warmke, B., Tosi, J., James, A. S., & Campbell, W. K. (2019). Moral
grandstanding in public discourse: Status-seeking motives as a potential
The authors declare that they have no known competing finan- explanatory mechanism in predicting conflict. PLOS ONE, 14(10), e0223749.
cial interests or personal relationships that could have appeared https://doi.org/10.1371/journal.pone.0223749.
to influence the work reported in this paper. Grubbs, Tosi, and Warmke (2020) Moral Grandstanding and Civic Engagement: An
Exploratory Consideration. Manuscript under publication review.
Guay, B., & Johnston, C. D. (2020). Ideological asymmetries and the determinants of
Acknowledgements politically motivated reasoning. /publication/ja_2020_pmr_ajps/.
Hewstone, M., Rubin, M., & Willis, H. (2002). Intergroup bias. Annual Review of Psychology,
53(1), 575–604. https://doi.org/10.1146/annurev.psych.53.100901.135109.
We gratefully acknowledge the support of the Charles Koch Iyengar, S., Lelkes, Y., Levendusky, M., Malhotra, N., & Westwood, S. J. (2019). The
Foundation, the National Institute of Civil Discourse, and Bowling origins and consequences of affective polarization in the United States. Annual
Green State University in funding this research. Review of Political Science, 22(1), annurev-polisci-051117-073034. https://doi.
org/10.1146/annurev-polisci-051117-073034.
Iyengar, S., Sood, G., & Lelkes, Y. (2012). Affect not ideology. Public Opinion Quarterly,
References 76(3), 405–431. https://doi.org/10.1093/poq/nfs038.
Iyengar, S., & Westwood, S. J. (2015). Fear and loathing across party lines: New
evidence on group polarization. American Journal of Political Science, 59(3),
Abramowitz, A. I., & Saunders, K. L. (2008). Is polarization a myth?. The Journal of
690–707. https://doi.org/10.1111/ajps.12152.
Politics, 70(2), 542–555. https://doi.org/10.1017/S0022381608080493.
Johnston, C. D. (2018). Authoritarianism, affective polarization, and economic ideology.
Anderson, C., Hildreth, J. A. D., & Howland, L. (2015). Is the desire for status a
Political Psychology, 39(S1), 219–238. https://doi.org/10.1111/pops.12483.
fundamental human motive? A review of the empirical literature. Psychological
Lammers, J., Koch, A., Conway, P., & Brandt, M. J. (2017). The political domain
Bulletin, 141(3), 574–601. https://doi.org/10.1037/a0038781.
appears simpler to the politically extreme than to political moderates. Social
Bakker, B. N., Lelkes, Y., & Malka, A. (2020). Understanding partisan cue receptivity:
Psychological and Personality Science, 8(6), 612–622. https://doi.org/10.1177/
Tests of predictions from the bounded rationality and expressive utility
1948550616678456.
perspectives. The Journal of Politics, 82(3), 1061–1077. https://doi.org/10.1086/
Lelkes, Y. (2016). Mass polarization: Manifestations and measurements. Public
707616.
Opinion Quarterly, 80(S1), 392–410. https://doi.org/10.1093/poq/nfw005.
Banda, K. K., & Cluverius, J. (2018). Elite polarization, party extremity, and affective
Leonardelli, G. J., Pickett, C. L., & Brewer, M. B. (2010). Optimal distinctiveness
polarization. Electoral Studies, 56, 90–101. https://doi.org/10.1016/
theory: A framework for social identity, social cognition, and intergroup
j.electstud.2018.09.009.
relations. Advances in experimental social psychology (Vol. 43, pp. 63–113).
Blatz, C. W., & Mercier, B. (2018). False polarization and false moderation: Political
Elsevier.
opponents overestimate the extremity of each other’s ideologies but
Litman, L., Robinson, J., & Abberbock, T. (2017). TurkPrime.com: A versatile
underestimate each other’s certainty. Social Psychological and Personality
crowdsourcing data acquisition platform for the behavioral sciences. Behavior
Science, 9(5), 521–529. https://doi.org/10.1177/1948550617712034.
Research Methods, 49(2), 433–442. https://doi.org/10.3758/s13428-016-0727-z.
Brady, W. J., & Crockett, M. J. (2019). How effective is online outrage?. Trends in
Loch, C., Yaziji, M., & Langen, C. (2001). The fight for the alpha position: Channeling
Cognitive Sciences, 23(2), 79–80. https://doi.org/10.1016/j.tics.2018.11.004.
status competition in organizations. European Management Journal, 19(1),
Brady, W. J., Crockett, M., & Van Bavel, J. J. (2019). The MAD model of moral
16–25. https://doi.org/10.1016/S0263-2373(00)00067-0.
contagion: The role of motivation, attention and design in the spread of
Mackie, D., & Cooper, J. (1984). Attitude polarization: Effects of group membership.
moralized content online [Preprint]. PsyArXiv. https://doi.org/10.31234/osf.io/
Journal of Personality and Social Psychology, 46(3), 575.
pz9g6.
Mackie, D. M. (1986). Social identification effects in group polarization. Journal of
Brandt, M. J., Evans, A. M., & Crawford, J. T. (2015). The unthinking or confident
Personality and Social Psychology, 50(4), 720.
extremist? Political extremists are more likely than moderates to reject
12 J.B. Grubbs et al. / Journal of Research in Personality 88 (2020) 104009

Mackie, D. M., Maitner, A. T., & Smith, E. R. (2016). Intergroup emotions theory. In Sherman, E. D., Miller, J. D., Few, L. R., Campbell, W. K., Widiger, T. A., Crego, C., &
Handbook of prejudice, stereotyping, and discrimination, pp. 149–174). Lynam, D. R. (2015). Development of a short form of the five-factor narcissism
Psychology Press. inventory: The FFNI-SF. Psychological Assessment, 27(3), 1110–1116. https://doi.
Martin, G. J., & Yurukoglu, A. (2017). Bias in cable news: Persuasion and org/10.1037/pas0000100.
polarization. American Economic Review, 107(9), 2565–2599. https://doi.org/ Smith, E. R., Seger, C. R., & Mackie, D. M. (2007). Can emotions be truly group level?
10.1257/aer.20160812. Evidence regarding four conceptual criteria. Journal of Personality and Social
Mason, L. (2015). ‘‘I Disrespectfully Agree”: The differential effects of partisan Psychology, 93(3), 431–446. https://doi.org/10.1037/0022-3514.93.3.431.
sorting on social and issue polarization. American Journal of Political Science, 59 Toner, K., Leary, M. R., Asher, M. W., & Jongman-Sereno, K. P. (2013). Feeling
(1), 128–145. https://doi.org/10.1111/ajps.12089. superior is a bipartisan issue: Extremity (not direction) of political views
McCarty, N. (2019). Polarization: What everyone needs to knowÒ. Oxford University predicts perceived belief superiority. Psychological Science, 24(12), 2454–2462.
Press. https://doi.org/10.1177/0956797613494848.
Moscovici, S., & Zavalloni, M. (1969). The group as a polarizer of attitudes. Journal of Tosi, J., & Warmke, B. (2016). Moral grandstanding. Philosophy & Public Affairs, 44(3),
Personality and Social Psychology, 12(2), 125–135. https://doi.org/10.1037/ 197–217. https://doi.org/10.1111/papa.12075.
h0027568. Tosi, J., & Warmke, B. (2020). Grandstanding: The use and abuse of moral talk. Oxford
Neel, R., Kenrick, D. T., White, A. E., & Neuberg, S. L. (2015). Individual differences in University Press.
fundamental social motives. Journal of Personality and Social Psychology, 110(6), van Prooijen, J.-W., & Krouwel, A. P. M. (2019). Psychological features of extreme
887. https://doi.org/10.1037/pspp0000068. political ideologies. Current Directions in Psychological Science, 28(2), 159–163.
Nesi, J., & Prinstein, M. J. (2019). In search of likes: Longitudinal associations https://doi.org/10.1177/0963721418817755.
between adolescents’ digital status seeking and health-risk behaviors. Journal of van Prooijen, J.-W., Krouwel, A. P. M., & Emmer, J. (2018). Ideological responses to the
Clinical Child & Adolescent Psychology, 48(5), 740–748. https://doi.org/10.1080/ EU refugee crisis: The left, the right, and the extremes. Social Psychological and
15374416.2018.1437733. Personality Science, 9(2), 143–150. https://doi.org/10.1177/1948550617731501.
R Core Team (2018). R: A language and environment for statistical computing. R Walasek, L., & Brown, G. D. A. (2015). Income inequality and status seeking:
Foundation for Statistical Computing. https://www.R-project.org/. Searching for positional goods in unequal U.S. States. Psychological Science, 26
Riek, B. M., Mania, E. W., & Gaertner, S. L. (2006). Intergroup threat and outgroup (4), 527–533. https://doi.org/10.1177/0956797614567511.
attitudes: A meta-analytic review. Personality and Social Psychology Review, 10 Wegener, D. T., Downing, J., Krosnick, J. A., & Petty, R. E. (1995). Measures and
(4), 336–353. https://doi.org/10.1207/s15327957pspr1004_4. manipulations of strength-related properties of attitudes: Current practice and
Rivers, D. (2016). Pew research: YouGov consistently outperforms competitors on future directions. In Attitude strength: Antecedents and consequences
accuracy | YouGov. https://today.yougov.com/topics/finance/articles-reports/ (pp. 455–487). Lawrence Erlbaum Associates Inc..
2016/05/13/pew-research-yougov. Westfall, J., Van Boven, L., Chambers, J. R., & Judd, C. M. (2015). Perceiving political
Robbins, J. M., & Krueger, J. I. (2005). Social projection to ingroups and outgroups: A polarization in the United States: Party identity strength and attitude extremity
review and meta-analysis. Personality and Social Psychology Review, 9(1), 32–47. exacerbate the perceived partisan divide. Perspectives on Psychological Science,
https://doi.org/10.1207/s15327957pspr0901_3. 10(2), 145–158. https://doi.org/10.1177/1745691615569849.
Rosseel, Y. (2012). lavaan: An R package for structural equation modeling and more Zaller, J. R. (1992). The nature and origins of mass opinion. Cambridge University
Version 0.5-12 (BETA). Journal of Statistical Software, 42, 1–36. Press.