Content uploaded by Nathaniel Geiger
Author content
All content in this area was uploaded by Nathaniel Geiger on Oct 11, 2017
Content may be subject to copyright.
UNCORRECTED PROOF
ARTICLE INFO
Article history:
Received 21 August 2015
Received in revised form 29 April 2016
Accepted 6 May 2016
Available online xxx
Keywords:
Climate change
Pluralistic ignorance
Discussion
Impression management
Stereotype content model
ABSTRACT
Despite the importance of interpersonal public communication about climate change, most citizens rarely discuss the
topic. In two studies, we find that inaccurate perceptions of others' opinions (i.e. pluralistic ignorance) contribute to self-
silencing among those concerned about climate change. Study 1 illustrates that those who are aware of others' concern
about climate change report greater willingness to discuss the issue than those with inaccurate perceptions of others' opin-
ions. Study 2 demonstrates that correcting pluralistic ignorance increases concerned participants' willingness to discuss
climate change. In both studies, pluralistic ignorance leads to self-silencing because perceptions that others do not share
one's opinion are associated with expecting to be perceived as less competent in a conversation about climate change. In
contrast to previous research on confronting prejudice, in the present research expectations about being disliked did not
explain self-silencing. We discuss the implications for self-silencing and promoting interpersonal communication about
climate change.
© 2016 Published by Elsevier Ltd.
Journal of Environmental Psychology xxx (2016) xxx-xxx
Contents lists available at ScienceDirect
Journal of Environmental Psychology
journal homepage: www.elsevier.com
Climate of silence: Pluralistic ignorance as a barrier to climate change discussion
Nathaniel Geiger, Janet K. Swim∗
Department of Psychology, The Pennsylvania State University, USA
1. Introduction
The challenge of climate change requires major economic and
social changes, both to transition to a low-carbon economy and to
adapt to the changes that are already “locked in”by previous pat-
terns of carbon emissions (IPCC, 2014). A strong limiting factor to
the success of these changes is the public's willingness to accept, sup-
port, and actively engage in shaping economic, sociocultural, political,
and structural changes that help to address climate change (Clayton
et al., 2015; Jacobson & Delucchi, 2011). This public response is
most likely to occur when social changes coincide with shared mean-
ing and value structures held by a majority of citizens (Dietz, 2013;
Habermas, 1971, p 27). Thus, interpersonal communication about top-
ics is crucial to build public acceptance and support for social change:
scientifically grounded public discussion can increase public under-
standing of the problem, community engagement, and development
of consensus for locally appropriate mitigation and adaptation solu-
tions (Clayton et al., 2015; Swim, Fraser, & Geiger, 2014). Currently,
however, engagement in these conversations are uncommon: only a
quarter of the American public report regularly discussing climate
change (Leiserowitz, Maibach, Roser-Renouf, Feinberg, & Rosenthal,
2015), and similar levels of silence are found among the British public
(Capstick et al., 2015; Rowson, 2013).
We suggest that the social dynamics surrounding climate change
are barriers to discussion –a socially constructed silence (Marshall,
2014; Norgaard, 2011, p 82). First, we propose that pluralistic igno-
rance –the tendency for a majority to misperceive others' opinions on
a topic, falsely believing that fewer people share their opinion than
actually do (Prentice & Miller, 1993) –contributes to the lack of dis-
cussion about climate change. Despite a solid majority of the public
∗∗ Corresponding author. 511 Moore Building, University Park, PA 16802, USA.
Email address: janet.swim@gmail.com (J. K. Swim)
being concerned about climate change, most underestimate the de-
gree to which others are concerned (Leviston, Walker, & Morwinski,
2013). Second, we propose that pluralistic ignorance leads people to
avoid discussing climate change because people anticipate being eval-
uated more negatively by those who disagree with them than those
who agree with them in anticipated conversations about the topic. Re-
search on core dimensions of social evaluation suggests that antici-
pated negative evaluations would be in the form of anticipating being
perceived to lack warmth, competence, or both (Fiske, Xu, Cuddy, &
Glick, 1999).
2. Pluralistic ignorance and self-silencing
It is perhaps unsurprising that the public has demonstrated plu-
ralistic ignorance about climate change (Leviston et al., 2013). Plu-
ralistic ignorance has been demonstrated across many topics: support
for racial segregation in the 1970s (most white Americans supported
desegregation but believed that most others supported segregation;
O'Gorman & Garry, 1976), norms of alcohol consumption (univer-
sity students believed that norms of alcohol consumption were exces-
sive but perceived that most others supported them; Prentice & Miller,
1993), opinions on foreign policy (most Americans support multilat-
eral foreign policy but perceive that most other Americans support
unilateral foreign policy; Todorov & Mandisodza, 2004) and comfort
with “hooking up”(students estimated that others felt more comfort-
able engaging in uncommitted sexual activity than they did; Lambert,
Kahn, & Apple, 2003). Pluralistic ignorance could in part be due to the
lack of regular conversations about climate change (Leiserowitz et al.,
2015), which could lead to individuals having little insight into others'
internal beliefs. Interestingly, pluralistic ignorance on climate change
has even been found among climate scientists who underestimate con-
cern among other scientific experts (Lewandowsky, Oreskes, Risbey,
Newell, & Smithson, 2015).
http://dx.doi.org/10.1016/j.jenvp.2016.05.002
0272-4944/© 2016 Published by Elsevier Ltd.
UNCORRECTED PROOF
2 Journal of Environmental Psychology xxx (2016) xxx-xxx
Pluralistic ignorance can have significant consequences for effec-
tively addressing social issues. Pluralistic ignorance is associated with
attitude change shifting toward the perceived norm (Leviston et al.,
2013; Prentice & Miller, 1993); behavioral conformity to the per-
ceived norm (Prentice & Miller, 1993; Schroeder & Prentice, 1998),
and relevant to the present study, reduced willingness to share one's
opinion on a topic (Miller & McFarland, 1987; Rios & Chen, 2014;
Taylor, 1982). Conversely, correcting pluralistic ignorance by provid-
ing information about the true beliefs of others can reverse these ef-
fects (Schroeder & Prentice, 1998).
The spiral of silence theory specifically addresses the impact of
pluralistic ignorance on public discourse (Noelle-Neumann, 1993;
Taylor, 1982). This theory proposes that individuals scan their social
environment for information about others' opinions and that people are
less willing to share their opinion when informational cues lead them
to believe that they hold a minority view (vs. majority view), espe-
cially when the topic is perceived as controversial or morally charged
(Noelle-Neumann, 1993). Silencing is proposed to be self-reinforcing:
if many who hold a particular view believe that they are in the minor-
ity and remain silent, the silence leads others who share this view to
believe that their opinion is uncommon and encourages them to also
remain silent. Motivation to self-silence is also proposed to increase
when individuals believe that their opinion is declining in public pop-
ularity (Taylor, 1982).
The premises derived from pluralistic ignorance and spiral of si-
lence theory may explain why people are hesitant to discuss climate
change. The principles outlined in spiral of silence theory are pur-
ported to apply primarily to morally controversial or value-laden top-
ics, such as abortion, support for addressing racial inequality, and po-
litical party preference in national elections (Moy, Domke, & Stamm,
2001; Noelle-Neumann, 1993). Climate change might appear to differ
from these more commonly studied topics because climate change is
a scientific topic supported by a solid body of evidence and an over-
whelming consensus of scientific experts whom agree that human-
caused climate change is occurring and presents a significant threat
to global civilization (Cook et al., 2013, 2016; Oreskes, 2004). Yet,
expression of opinions about climate change has taken on a cultural
significance distinct from scientific understanding of the topic due to
its politicization. About half of U.S. senators recently voted to pub-
licly deny that “human activity significantly contributes to climate
change”(Goldenberg, 2015), despite scientific consensus and only ap-
proximately 10% of Americans similarly dismissing the scientific evi-
dence behind anthropogenic climate change (Leiserowitz et al., 2015).
Further, many perceive the topic principally as a moral topic be-
cause of the potential negative impacts of unchecked climate change
(Markowitz, 2012). Another point of view, expressed by a vocal mi-
nority who question the scientific consensus, is that climate change
is a conspiracy that is immorally being promoted as scientific fact by
those who wish to promote a specific political agenda (Lewandowsky,
Oberauer, & Gignac, 2013). In sum, it appears that climate change has
culturally acquired a controversial, moral connotation in modern soci-
ety, and thus we propose that the processes described in pluralistic ig-
norance and spiral of silence will also apply to climate change. Thus,
we make the following prediction:
Hypothesis 1
Participants will be less willing to talk about climate change
when they perceive that their opinions are in the minority (vs. the
majority).
3. Impression management and self-silencing
Self-silencing may be a form of impression management. Individ-
uals desire to be viewed in a positive light and sharing an unpopular
opinion could result in others perceiving them negatively. Researchers
have proposed that people self-silence because of fear of isolation
(Noelle-Neumann, 1993), rejection (Bergsieker, Shelton, & Richeson,
2010), social retributions for violating cultural norms prescribing si-
lence (Norgaard, 2011), embarrassment (Miller & McFarland, 1987),
being dismissed as a “complainer”(Kaiser & Miller, 2001; Swim &
Hyers, 1999), and being seen as ignorant (Salmon & Neuwirth, 1990).
The varied explanations for self-silencing listed in the above para-
graph can be organized along two core dimensions of impressions:
fears about being a) disliked or b) losing respect. Interpersonal evalu-
ation research suggests that up to 90% of initial impressions of others
can be organized along these two core dimensions, which directly re-
flect the core dimensions of social cognition: warmth (those perceived
as low in warmth are disliked) and competence (those perceived as
low in competence are not respected) (Abele & Wojciszke, 2007;
Fiske et al., 1999). These two dimensions have been consistently de-
scribed across various literature (Fiske, Cuddy, Glick, & Xu, 2002;
Heider, 1958; Rosenberg, Nelson, & Vivekanathan, 1968; Singh, Ho,
Tan, & Bell, 2007) and align with two basic impression management
goals: the desire to have an audience think favorably about oneself and
the desire to present one's ideal self to others (Bergsieker et al., 2010;
Schlenker, 1975). Thus, warmth reflects being perceived as friendly
and cooperative, while competence corresponds with being respected
and achieving high social status (Fiske et al., 1999, 2002). Being per-
ceived as either cold (i.e., confrontational and unlikeable) or incompe-
tent (i.e., not respected and low status) are distinct grounds for antic-
ipated social rejection, and thus people may alter their behavior in at-
tempts to manage others' impressions of them on one or both of these
dimensions (Holoien & Fiske, 2013).
3.1. Avoiding being disliked
The desire to avoid being disliked has been well established as a
motive for self-silencing when one is a target of discrimination and
prejudice. (Sechrist et al., 2004; Shelton & Stewart, 2004; Stangor
et al., 2003; Swim & Hyers, 1999). This desire leads individuals to
refrain from confronting discrimination despite their wishes to do so
(e.g., Swim, Eyssell, Murdoch, & Ferguson, 2010) or despite what
they expect they would do (Shelton & Stewart, 2004; Swim & Hyers,
1999; Woodzicka & LaFrance, 2001), particularly in the presence of
others expected to not share one's own point of view (Swim & Hyers,
1999). Individuals faced with discrimination often perceive the possi-
bility of confronting as impolite (Swim & Hyers, 1999) and those who
do confront are devalued as difficult to interact with and “complain-
ers”(Kaiser & Miller, 2001). Consistent with the argument that indi-
viduals self-silence to avoid being disliked, women were less likely to
assertively respond to sexist comments during a job interview when
the desire to be liked was emphasized than when the desire to be re-
spected was valued more highly (Mallett & Melchiori, 2014).
Fear of being disliked may also motivate suppression of opinions
about climate change. Individuals may be concerned about being per-
ceived as an “alarmist”or environmental activist if they were to ex-
press their concern about the topic, labels which carry potentially
negative connotations of being disliked by others. Similar to those
who confront discrimination, those who are “alarmed”about climate
UNCORRECTED PROOF
Journal of Environmental Psychology xxx (2016) xxx-xxx 3
change are perceived as “whiny”,“nagging”, and “complainers”
(Swim & Geiger, 2016b) and prototypic environmental activists are
commonly stereotyped as “eccentric”,“self-righteous,”and “reactive”
(Bashir et al., 2013), all traits associated with being seen as cold and
disliked by others. These negative impressions are associated with re-
duced willingness to engage in climate change activism and to affil-
iate with environmental activists. Expectations about being disliked
for speaking one's opinions may be accentuated in particular contexts;
namely, when individuals anticipate that others do not share their
views and thus believe that expressing their opinion would be con-
frontational within a given context (Noelle-Neumann, 1993). Based
upon the above, we make the following prediction.
Hypothesis 2
Individuals
’
hesitation to discuss climate change in situations
when they perceive their opinions are in the minority (vs. the ma-
jority) will be partly explained by expectations of appearing less
warm in the conversation.
3.2. Avoiding losing respect
Another motive for self-silencing is the concern that one would
lose others' respect following a conversation about a topic. Express-
ing an unpopular opinion could result in appearing ignorant to oth-
ers (Salmon & Neuwirth, 1990), and people may remain silent out of
fear of embarrassing themselves when they believe that they are less
knowledgeable about a topic than others (Miller & McFarland, 1987).
Yet, research suggests that confronting discrimination does not lead
to the confronter being perceived as incompetent (i.e. losing respect;
Stangor et al., 2003; Swim, Gervais, Pearson, & Stangor, 2009), and
women more interested in being respected than liked were more likely
to confront sexism during a job interview than other women (Mallett
& Melchiori, 2014). This could suggest that concerns about losing re-
spect are less central than concerns about being disliked when individ-
uals consider whether to self-silence unpopular opinions.
However, in contrast to confronting discrimination, the degree to
which an individual expects to be perceived as competent may affect
willingness to engage in discussions about climate change. Since cli-
mate change is a scientific topic, expectations of appearing competent
may be more salient than expectations of appearing warm since un-
derstanding of scientific topics maps onto the competence dimension,
but not the warmth dimension, of social cognition (Fiske, Cuddy, &
Glick, 2007). This proposition is supported by work examining infor-
mal scientific educators' concerns about incorporating climate change
into their education curriculum (Swim & Fraser, 2013, 2014). The
more concerned educators were about being able to competently com-
municate about climate change the more likely they were to avoid ex-
tensively communicating with visitors about this topic.
Given that even trained scientific educators express concern about
being capable of communicating climate change, nonscientists may be
even more likely to hold these concerns. Research shows that most
nonscientists have limited understanding of the scientific mechanisms
of climate change (Leiserowitz, Smith, & Marlon, 2010; Swim et al.,
2014), and thus may be concerned about appearing ignorant or incom-
petent when discussing this topic. Further, the expectation of appear-
ing incompetent may be amplified when considering a discussion with
an audience not expected to share one's views, partly because a dis-
senting audience may challenge the speaker or question their assump-
tions. Based upon the above analyses we make the following hypoth-
esis:
Hypothesis 3
Individuals
’
hesitation to discuss climate change in situations
when they perceive their opinions are in the minority (vs.the ma-
jority)will be partly explained by expectations of appearing less
competent in the conversation.
4. Present research
In two studies, we examine the effects of pluralistic ignorance on
willingness to discuss climate change. We first conducted two pilot
studies to verify that the pattern of pluralistic ignorance about climate
change observed in Leviston et al. (2013) work would replicate in our
target population. Next, in Study 1 we examine whether participants
who do not themselves doubt the scientific view on climate change
and hold inaccurate perceptions of others' opinions are less willing to
discuss the topic than those who endorse similar views about climate
change but hold accurate perceptions of others' opinions. In Study 2,
we experimentally manipulate perceptions of others' opinions and ex-
amine the effects of correcting pluralistic ignorance on facilitating dis-
cussion relative to emphasizing the false perceptions of others' opin-
ions. Study 2 participants include a full range of personal opinions
about climate change ranging from those who are very concerned
about climate change to those who consider themselves nonbelievers.
In both studies, we examine whether expectations of being perceived
as a) warm and/or b) competent explain the psychological process un-
derlying these effects.
5. Pilot studies
Pilot testing with undergraduate students in introductory psychol-
ogy courses (reflecting a range of students across the campus) were
consistent with previous findings about pluralistic ignorance on cli-
mate change (Leviston et al., 2013). One pilot sample completed a
screening instrument that has been used to categorize the public into
different levels of concern about climate change (i.e., the Six Amer-
icas' questionnaire, Maibach, Leiserowitz, Roser-Renouf, & Mertz,
2011) and self-categorized into different levels of concern about cli-
mate change as assessed by the same screening instrument. Both meth-
ods indicated that a majority of students were on concerned side of
the opinion spectrum: survey instrument (N = 365): 7% Alarmed, 40%
Concerned, 40% Cautious, 3% Disengaged, 8% Doubtful, 3% Dis-
missive; self-categorization (N = 368): 8% Alarmed, 28% Concerned,
39% Cautious, 19% Disengaged, 3% Doubtful, 3% Dismissive. This
pattern is similar to this age group in the American public (Leiserowitz
et al., 2015) and in subsequent tests in the same participant pool
(Swim & Geiger, 2016a). Yet, despite this majority concern, in a sec-
ond pilot test (N = 89), only 30% of respondents accurately perceived
that a majority of other students were concerned about climate change.
The most common misperceptions were that: 1) most undergraduates
were disengaged with the topic, 2) most undergraduates were doubtful
about climate change, or 3) undergraduates' opinions were polarized.
6. Study 1
In Study 1, we examine whether pluralistic ignorance predicts si-
lence on climate change among undergraduate college students who
do not doubt the existence of climate change. In this study, we fo-
cus on the three most common misperceptions of other students' opin-
ions identified in the pilot study above. Specifically, Study 1 tests
whether those who had accurately perceived that most undergraduates
were concerned about climate change in a pre-screening would later
be more willing to engage in a discussion about climate change than
UNCORRECTED PROOF
4 Journal of Environmental Psychology xxx (2016) xxx-xxx
those who had endorsed one of three other types of inaccurate percep-
tions of others' opinions in the pre-screening.
Previous research on pluralistic ignorance (Goode, Balzarini, &
Smith, 2014; Larimer, 2010) and willingness to engage in discussion
(Oshagan, 1996; Salmon & Kline, 1985) suggests that perceptions of
others' behaviors and opinions can influence conformity only to the
extent that the particular “others”referenced are perceived to be rel-
evant to the individual in a given situation. Specifically, people are
likely to modify their behavior or discussion based on the perceived
opinions of those with whom they identify (Larimer, 2010; Neighbors
et al., 2010) or the perceived opinions of those with whom they are
speaking (Azjen & Fishbein, 1977), while those with whom they do
not identify or a more distant reference group may be unlikely to in-
duce conformity. To verify this proposition, we also assess percep-
tions of the American public's (a more abstract and distant reference
group) opinion in the pre-screening. We anticipated that these percep-
tions would not be associated with later willingness to discuss climate
change.
6.1. Methods
6.1.1. Participants
Respondents were recruited based on their answers in pre-screen-
ing as part of a battery of measures submitted by several psychol-
ogy labs in the university department distributed during the first two
weeks of classes. A subset of the initial 1148 participants from the
pre-screening were recruited based upon their opinions about climate
change and their perception of other students' opinions about climate
change (measures are described in sections 5.1.3.1 and 5.1.3.2) We did
not recruit those who indicated that they were “Doubtful”or a “Non-
believer”in order to study self-silencing among those who hold the
majority opinion. Perceptions of other students' opinions were used to
select an adequate representation of the four categories of opinion per-
ceptions which made up the primary predictor variable in the present
study.
Respondents were 305 undergraduate students (124 males, 180 fe-
males, and one student who did not indicate gender) enrolled in intro-
ductory psychology classes at Pennsylvania State University. The av-
erage age was 20 (range 18–48), and most students (76%) identified as
White, with the largest three ethnic minority groups being Asian (8%),
Hispanic (6%), and Black (5%). Politically, more students were lib-
eral (34%) than conservative (19%), with 36% describing themselves
as moderate and 7% as libertarian.
6.1.2. Procedure
Approximately six weeks after completing pre-screening mea-
sures, participants were directed to an online survey. Participants were
instructed to imagine the following scenario:
You are assigned a group class project in your Intro to English
class at Penn State. You are in a group with 4 other people, and
you decide to meet together in the library to work on this project.
After a few productive hours of work, your group begins to get
distracted. The topic of the weather comes up. You are thinking
about bringing up how climate change may be affecting the current
weather.
Participants then indicated their expectations about how they
would be perceived if they initiated a discussion about climate change
and their willingness to discuss climate change in this setting by an-
swering the survey measures described below.
6.1.3. Pre-screening measures
6.1.3.1. Personal opinion about climate change
Respondents self-categorized into one of six groups arranged in a
continuum from most to least concerned about climate change (Very
Concerned, Concerned, Cautious, Disengaged, Doubtful, and Nonbe-
liever). We used the terms “Very Concerned”and “Nonbeliever”in-
stead of the original “Six Americas”category labels “Alarmed”and
“Dismissive”(Leiserowitz et al., 2015) because of the potential pejo-
rative nature of these labels. We provided descriptions of each group
to participants based on prototypical descriptions given in the “Six
Americas”report (Maibach, Leiserowitz, & Roser-Renouf, 2009, also
see Swim & Geiger, 2016a for validity of self-categorization).
Similar to pilot data, most students who had completed the pre-
screening were more concerned than unconcerned (Very Concerned
11%, Concerned 27%, Cautious 27%, Disengaged 26%, Doubtful 6%,
Nonbeliever 3%.).
6.1.3.2. Perceptions of others' opinions (proposed predictor)
Respondents chose one of five graphs that they believed best rep-
resented the distribution of: a) fellow university undergraduates' and
b) the American public's opinions about climate change (Fig. 1). The
categories of concern depicted in the graphs were derived from the
descriptions which respondents had previously read in order to de-
scribe their own opinions (section 5.1.3.1). The five graphs were de-
signed to represent the following five options: 1) most are concerned
(concerned model that most accurately represents the university stu-
dent population; Option C); 2) most doubt climate change is occurring
(doubtful model; Option D), 3) most are in the middle of the distribu-
tion (disengaged model; Option B), 4) a bimodal distribution (polar-
ized model; Option E) and 5) people are equally distributed across all
six possibilities (rectangular model, Option A). Pilot testing indicated
that respondents understood the meanings of the different graphs and
suggested that Option A indicates uncertainty or lack of an ability to
decide.
As noted above, we recruited participants to adequately represent
participants who endorsed each of the four distributions assessed in
the study. We did not recruit the respondents who chose the rectan-
gular distribution because few endorsed this distribution in pilot test-
ing and our predictions were about comparison among respondents
who chose the other four distributions. We did not use responses on
perceptions of the American public's opinions to guide participant re-
cruitment. This resulted in the inclusion of a few participants who had
selected the rectangular distribution for perceptions of the American
public's opinions (n = 7). These seven participants were removed from
all analyses which involved this variable.
6.1.4. Survey measures
The survey measures were assessed six weeks after the pre-screen-
ing measures detailed above.
6.1.4.1. Anticipated evaluations by others (proposed mediators)
Participants rated their expectations of being perceived as warm
(friendly, fun, good-natured, likeable, nice; α= 0.89) and competent
(competent, responsible, intelligent, level-headed, successful;
α= 0.76) (each on a −2 “Very Unlikely”to 2 “Very Likely”scale)
if they were to talk about climate change. We also assessed expec-
tations of appearing as a complainer, an alarmist, and an environ-
mentalist to analyze since expectations about being disliked (i.e., per-
ceived as cold) might be grounded in expectations about being cate-
gorized in one or more of these groups. Similar to expectations of be-
ing perceived as warm (as presented below), results show null results
UNCORRECTED PROOF
Journal of Environmental Psychology xxx (2016) xxx-xxx 5
Fig. 1. Opinion distribution answer choices in Study 1. Those who chose Option A were not recruited to participate in the study.
for relationships between these variables and perceptions of others'
opinions. Thus, to simplify presentation of results, we do not present
analyses related to these variables.
6.1.4.2. Willingness to discuss climate change (proposed outcome)
Following other research measuring willingness to discuss contro-
versial topics (e.g. Noelle-Neumann, 1993; Scheufele, Shanahan, &
Lee, 2001), respondents answered the question, “How likely would
you be to discuss climate change in the above situation?”(−2 “Very
Unlikely”to 2 “Very Likely”). Responses were normally distributed
(skewness = −0.23, kurtosis = −1.00), suggesting that ordinary least
squares (OLS) regression techniques would be appropriate for data
analysis.
6.2. Results
6.2.1. Overview
We used one-way ANOVAs to test whether perceptions of a) other
undergraduates' and b) the American public's opinions are related to
i) willingness to discuss climate change and ii) anticipated evaluations
by others. The independent variable corresponded to the four opinion
distributions the respondent chose for the relevant target group. We
next conducted mediation analyses to test whether anticipated evalua-
tions mediated the relation between opinion perceptions and willing-
ness to discuss climate change.
6.2.2. Pluralistic ignorance and discussions
Perceptions about undergraduates' opinions were associated with
willingness to discuss climate change, F(3, 301) = 2.96, p= 0.03,
η
2= .03.1Contrast tests compared those who accurately perceived
that most undergraduates were concerned about climate change ver-
sus each of the other three possible inaccurate perceptions. As hy-
pothesized, those who inaccurately perceived that most undergradu-
ates were doubtful about climate change were significantly less will
1For this and all other analyses presented below, we examined whether results
were moderated by participants' personal opinion about climate change. We found
no significant interaction effects in any analyses, ps > 0.17, perhaps partly due to
the restricted range of opinions among participants recruited to be in this sample.
ing to discuss climate change (M= −0.85, S.d. = 1.00) than those who
accurately perceived that most undergraduates were concerned
(M= −0.23, S.d. = 1.15), p= 0.004,
η
p
2= 0.03. In contrast, those who
inaccurately perceived that most undergraduates were disengaged
(M=−0.40,S.d =1.11), or polarized in their opinions (M= −0.44,
S.d =1.24), were not statistically less willing to discuss the topic than
those with accurate opinions (ps >0.25), although the difference was
in the expected direction.
We conducted a separate ANOVA to examine whether perceptions
of the American public's opinions were related to willingness to dis-
cuss climate change. In contrast to the above findings with percep-
tions of undergraduates' opinions, perceptions of the American pub-
lic's opinions were not related to willingness to talk about climate
change, F(3, 294) = 0.37, p= 0.77,
η
2= 0.004.
6.2.3. Impression management
Perceptions of other undergraduates' opinions were associated with
expectations about appearing competent, F(3, 300) = 4.31, p= 0.005,
η
2= 0.04, but not with expectations about appearing warm, F(3,
300) = 1.65, p= 0.18,
η
2= 0.02. Contrast tests again compared those
who accurately perceived that most undergraduates were concerned
about climate change versus each of the other three possible inaccu-
rate perceptions. As hypothesized, participants who believed that most
undergraduates were doubtful about climate change expected to ap-
pear less competent (M= 0.46, S.d. = 0.65) than undergraduates who
held more accurate perceptions (M= 0.81, S.d. = 0.52), p= 0.001,
η
p
2= 0.04. Undergraduates who held the more accurate perception
also perceived that others would perceive them as more competent
than those who believed others were disengaged (M= 0.60,
S.d = 0.54), p= 0.01,
η
p
2= 0.02, but not more than undergraduates
who believed others were mostly polarized, (M = 0.69, S.d. = 0.53),
p= 0.16.
We conducted separate ANOVAs to examine whether perceptions
of the American public's opinions were related to expectations of ap-
pearing warm or competent. In contrast to perceptions of other un-
dergraduates' opinions, perceptions of the American public's opin-
ions were not related to expectations of appearing competent, F(3,
294) = 2.02, p= 0.11,
η
2= 0.02, or expectations of appearing warm,
F(3, 294) = 0.09, p= 0.97,
η
2= .001.
UNCORRECTED PROOF
6 Journal of Environmental Psychology xxx (2016) xxx-xxx
6.2.4. Mediation analyses
We used PROCESS with 5000 bootstraps to conduct a regres-
sion-based parallel mediation analysis (Hayes, 2013, Model 4) test-
ing whether the observed difference between those who believed other
students were concerned and those who believed other students were
doubtful was explained by perceptions of appearing warm or com-
petent. We created the following dummy code based on participants'
perceptions of other students to test this relationship: concerned = 0,
disinterested = 0, polarized = 0, doubtful = −1 (entered as the indepen-
dent variable) and included the following two other dummy coded
variables as covariates in the mediation model to control for other
possibilities examined in the ANOVA: a) concerned = 0, disinter-
ested = −1, polarized = 0, doubtful = 0 and b) concerned = 0, disinter-
ested = 0, polarized = −1, doubtful = 0 (see Cohen, Cohen, West, &
Aiken, 2003). We entered a) perceptions of appearing warm and b)
perceptions of appearing competent as parallel mediators and willing-
ness to discuss climate change as the dependent variable. We omit-
ted the variable corresponding to perceptions of the American pub-
lic's opinions from these analyses because the prior analyses showed
that this variable was unrelated to the other variables in the mediation
analysis.
As shown in Fig. 2, those who believed most other students
doubted climate change were less willing to discuss the topic than
those who believed most other students were concerned because the
former expected to be perceived as less competent in a discussion
than did the latter, indirect effect = 0.09, SE = 0.06, 95% CI [0.01,
0.25], while expectations of appearing warm did not play a role in this
process, indirect effect = 0.01, SE = 0.03, 95% CI [−0.02, 0.11].
6.3. Discussion
Among students who do not doubt climate change, those who en-
dorse the misperception that most others doubt climate change were
less likely to engage in discussions about climate change than those
who held accurate perceptions of others' opinions. In contrast, those
who held other types of inaccurate perceptions—that others are dis-
engaged or polarized–were not significantly less willing to discuss
climate change than those who held the accurate perception. The lack
of statistical significance in differences between these three groups
prevented us from making any conclusions about difference among
these three groups. However, examining the mean values for each of
the four groups leads to the speculation that the effect on perceptions
of others' opinions on self-silencing is a matter of degree: the more
an individual's perception diverges from the reality that others share
their concern about climate change, the more hesitant they may be to
speak up. Future research could consider assessing perceptions of oth-
ers' opinions as a continuous variable to examine whether the relation-
ship between perceptions of others' opinions and self-silencing best
fits a linear pattern or if there is a certain tipping point upon which in-
dividuals become more likely to self-silence.
Results also show that the reason that those who believe that most
others doubt climate change are less willing to discuss the topic than
those who accurately perceive others' opinions is because the former
expect to lose respect (appear less competent) in a discussion, and not
because they expect to be disliked (perceived as less warm). This sug-
gests that many individuals are uncomfortable engaging in discussions
about climate change with a dissenting audience because they believe
that they do not have the ability to do so, rather than being concerned
about appearing socially deviant or hostile.
7. Study 2
Study 2 tested whether experimentally manipulating perceptions of
others' opinions would affect willingness to discuss climate change.
Most published research on pluralistic ignorance is correlational; to
our knowledge, only one study has experimentally manipulated per-
ceptions of others' opinions (opinions about drinking behavior;
Schroeder & Prentice, 1998). Similarly, although research on the spi-
ral of silence links perceptions of others' opinions to willingness to
speak out on topics (Glynn & Huge, 2014; Noelle-Neumann, 1993),
to our knowledge only one publication has experimentally manipu-
lated perceptions of others' opinions to verify that these perceptions
exert a direct causal influence on willingness to discuss topics (Rios
& Chen, 2014). Other research, however, points to the effects of ma-
nipulating perceptions of others opinions on behaviors; for example,
relative to information indicating that others are not prejudiced, infor
Fig. 2. Expectations of appearing competent and warm as mediators of the effects between the perceived opinions of others and willingness to discuss climate change in Study 1.
Higher values on Perceived Opinions of Others indicate more accurate perceptions of others' opinions. See text for a full description of how these variables were entered into the
model. ***p< 0.001; **p< 0.01; *p< 0.05; +p= 0.056.
UNCORRECTED PROOF
Journal of Environmental Psychology xxx (2016) xxx-xxx 7
mation that indicates others are prejudiced increases participants' dis-
criminatory behavior (Sechrist & Stangor, 2001). Thus, in Study 2, we
extend this research by examining whether manipulating perceptions
of other's opinions about climate change–thereby reinforcing or cor-
recting pluralistic ignorance–influences willingness to discuss climate
change. In addition, we reassess expectations of appearing warm and
competent to test whether manipulating pluralistic ignorance affects
willingness to discuss climate change via increasing expectations of
being liked or respected by others.
We add two additional dimensions to the design of Study 2. First,
we expand our sample to include people with a full range of beliefs
about climate change, thereby including those who disagree with the
scientific view on climate change as well as those who are concerned
about climate change. We include the full range to test whether cor-
recting inaccurate perceptions that few are concerned about climate
change reduces willingness to discuss the topic among those who do
not share the scientific view on climate change relative to exacerbating
the incorrect perception that overestimates the extent to which others'
share their doubts (Leviston et al., 2013). That is, we test whether the
same information provides inverse effects for those who doubt versus
those who are concerned about climate change. Those who doubt cli-
mate change may also harbor concerns about expressing their opinions
about the topic because people also report negative impressions of this
group consistent with low warmth (Swim & Geiger, 2016b) –perceiv-
ing them as arrogant, aggressive, and dictatorial.
Second, we examined the effects of whether participants were led
to believe that concern about climate change was increasing or de-
creasing. Spiral of Silence theory (Noelle-Neumann, 1993) and pre-
vious research (Petrič& Pinter, 2002; Taylor, 1982) suggest that per-
ceptions of changing opinions affect willingness to discuss topics such
that individuals may be more willing to discuss topics when they be-
lieve their opinion will become more popular in the future than when
they think their opinion will become less popular in the future. Yet,
to our knowledge, the causal relationship between these two variables
has not been demonstrated in an experimental study. In Study 2, we
test whether participants led to believe that their opinion about climate
change is increasing report greater willingness to discuss the topic
than those led to believe that their opinion about climate change is de-
creasing.
In addition to these added dimensions, we also altered a key as-
pect of the design from Study 1 to increase the external validity of our
study: in contrast to the hypothetical situation used in Study 1, par-
ticipants learned that they would actually speak about climate change
in a classroom setting. This has been shown to enhance the validity
of findings related to willingness to discuss a topic (Scheufele et al.,
2001).
7.1. Methods
7.1.1. Participants
Participants were 194 students recruited from Pennsylvania State
University introductory psychology classes, made up of 105 males, 88
females, and one who did not indicate gender. Most students (73%)
identified as White, with the largest three ethnic minority groups being
Asian (10%), Black (6%), and Hispanic (6%). Politically, more stu-
dents reported being liberal (33%) than conservative (22%), with 33%
identifying as moderate and 9% as libertarian. Most students (74%)
reported that at least one of their parents had a 4-year college degree
or higher.
7.1.2. Procedure
Participants reported to a classroom along with 10–23 others. They
completed a brief survey assessing their personal concern about cli-
mate change and then answered the same question using an electronic
clicker allegedly so that they could “get a feel for what the classroom
thinks as a whole.”The first independent variable in the 2(others' cur-
rent opinions) x 2(changes in others' opinions over time) was intro-
duced, as follows. The experimenter displayed a graph that allegedly
showed the class clicker responses. However, participants were actu-
ally randomly assigned to view one of two previously created graphs
that indicated that either: a) most others in the classroom were con-
cerned about climate change, or b) most others in the classroom were
unconcerned about climate change. To ensure that participants under-
stood the graph's meaning, the experimenter explained the graph to the
participants. The second independent variable was introduced through
this description: the experimenter described the results as either indi-
cating that concern was a) increasing or b) decreasing from previous
years. All participants within a single timeslot were exposed to the
same experimental condition.
After viewing the graph, participants were informed that they
would separate into small groups and have a discussion about climate
change. Before doing so, they completed a second survey, which as-
sessed anticipated evaluations by others and willingness to participate
in the discussion, as well as anticipatory affect measures which we an-
alyzed for a different research project. After the survey, participants
conducted a group discussion about climate change and completed ad-
ditional surveys which we analyzed for the additional research project
mentioned.
7.1.3. Measures
7.1.3.1. Personal opinion about climate change
Participants self-categorized into one of the groups presented in
Study 1, with the modification that the Cautious option was not pre-
sented because the electronic clicker we used as a prop only had five
buttons.
7.1.3.2. Anticipated evaluations by others
Using the same items as Study 1, participants rated their expecta-
tions of being perceived as warm (α= 0.83) and competent (α= 0.75)
in the upcoming discussion about climate change. As in Study 1, we
also assessed expectations of appearing as an environmentalist, an
alarmist, and as a complainer. Again, results show null results for rela-
tionships between these variables and perceptions of others' opinions,
so we do not present analyses related to these variables in the results
to simplify the presentation.
7.1.3.3. Willingness to discuss climate change
Similar to Study 1, Study 2 participants answered the question:
“Please indicate your willingness to participate in the group discus-
sion on climate change,”(−2 “Very Hesitant”to 2 “Very Willing”).
Again, this variable was normally distributed (skewness = −0.34, kur-
tosis = −0.23), suggesting that OLS regression techniques were appro-
priate for data analysis.
7.2. Results
7.2.1. Overview
We first tested the hypotheses that the effects of both experimen-
tal manipulations would be moderated by personal opinions about cli-
mate change. We conducted a three-step hierarchical regression pro
UNCORRECTED PROOF
8 Journal of Environmental Psychology xxx (2016) xxx-xxx
cedure (Cohen et al., 2003) to examine main effects and interactions,
entering a) students' willingness to discuss climate change and b) an-
ticipated evaluations by others were regressed on i) others' current al-
leged opinions, ii) whether participants were informed that concern
was increasing or decreasing, and iii) personal opinions about climate
change in Step 1, all two-way interactions in Step 2, and the three-
way interaction (for exploratory purposes) in Step 3 (see Table 1). We
next conducted a conditional mediation analysis to examine whether
the mediation findings in Study 1 would replicate for those upon both
ends of the opinion spectrum.
7.2.2. Pluralistic ignorance and discussions
The effect of others' alleged opinions about climate change on will-
ingness to discuss climate change was moderated by participants' own
concern about the topic, b= 0.41, SE = 0.16, p= 0.01 (Table 1). As
predicted, undergraduates were more willing to discuss the topic when
they were led to believe others would share their opinion than when
they were led to believe they would not (see Fig. 3). The effect of the
manipulation was of similar magnitude (but of opposite direction) for
participants of opposing opinions.
In contrast, students who were led to believe that concern was
increasing (vs. decreasing) were no more willing to discuss climate
change and this effect was not moderated by participants' own concern
about the topic, ps > 0.10.
7.2.3. Impression management
The effect of others' alleged opinions about climate change on
expectations of appearing competent was moderated by participants'
own concern about the topic, b= .23, SE = 0.09, p= 0.01. As hypoth-
esized, participants expected that others would perceive them as more
competent in the discussion when they believed others shared their
opinions (vs. did not share their opinions) (Fig. 3). In contrast, being
led to believe that concern was increasing (vs. decreasing) did not in-
fluence expectations of appearing competent and this effect was not
moderated by participants' concern about the topic, ps > 0.10.
The effect of others' alleged opinions about climate change on ex-
pectations of appearing warm was marginally moderated by partici-
pants' own concern about the topic, b= .18, SE = 0.10, p= 0.08.2As
before, students being led to believe that concern was increasing (vs.
decreasing) did not predict expectations of appearing warm and this
null relationship was not moderated by participants' concern about the
topic, ps > 0.10.
7.2.4. Mediation analyses
We conducted a conditional mediation analysis (Hayes, 2013,
model 8 with 5000 bootstraps) to explore whether the interaction be-
tween others' alleged opinions and personal opinions on willingness
to discuss climate change was mediated by expectations of appear-
ing competent or warm. As shown in Fig. 4, expectations about ap-
pearing competent, but not expectations of appearing warm, mediated
the relation between the interaction between undergraduates' opin-
ions and of others' alleged opinions and willingness to discuss climate
change. Spotlight analyses revealed that those who were concerned
about climate change were more willing to discuss climate change
when they were led to believe that others were concerned (vs. not
concerned) because they believed they would be perceived as more
competent when others shared their views, indirect effect b= 0.06,
SE = 0.05, 95% CI [0.004, 0.156]. In contrast, those who were doubt
2As anticipated, this trend was driven by participants expecting that others would
perceive them as warmer when they believed that others shared their opinions (vs.
did not share their opinions).
Table 1
Effects of others' alleged opinions, changes in others alleged opinions over time, and
participants' own opinions on dependent measures (Study 2).
Independent Variables
Willingness to
discuss climate
change
Expectations of
appearing competent
Expectations of
appearing warm
B SE η2
pB SE η2
pB SE η2
p
Step 1
Others concerned (vs.
unconcerned)
0.20 0.14 0.01 0.03 0.07 0.00 0.11 0.08 0.01
Concern increasing
(vs. decreasing)
0.00 0.14 0.00 −0.04 0.07 0.00 −0.06 0.08 0.00
Personal opinion
(more concerned)
−0.06 0.08 0.00 0.12 0.04 0.04** 0.08 0.05 0.01
Step 2
Others
concerned ×personal
opinion
0.41 0.17 0.04* 0.23 0.09 0.04* 0.18 0.10 0.02
Concern
increasing ×personal
opinion
−0.11 0.16 0.01 −0.03 0.09 0.00 −0.13 0.10 0.01
Others
concerned ×concern
increasing
0.30 0.27 0.01 −0.10 0.15 0.00 −0.21 0.17 0.01
Step 3
Others
concerned ×concern
increasing ×personal
opinion
−0.64 0.33 0.02 −0.25 0.18 0.02 −0.05 0.20 0.00
**p< 0.01; *p< 0.05.
ful about climate change were less willing to discuss climate change
when they believed others were concerned (vs. not concerned) be-
cause they believed they would be perceived as less competent when
others disagreed with their views, indirect effect b= −0.15, SE = 0.10,
95% CI [−0.36, −.02].
7.3. Discussion
Results replicate and extend Study 1 findings that concerned par-
ticipants are less willing to discuss climate change when they believe
others do not share their concern (vs. share their concern). Study 2 re-
sults extend this finding by revealing that this effect applies to those
upon both ends of the opinion spectrum. Similarly, Study 2 results
replicate and extend Study 1 findings that concerned participants self-
silence because they expect to be perceived as less competent in a con-
versation about climate change with those who disagree (vs. agree)
with them. Study 2 reveals that this effect occurs for those upon both
ends of the opinion spectrum. Study 2 further reveals that self-silenc-
ing among those concerned about climate change can be countered by
providing accurate information: those who hold the majority opinion
are more willing to share their opinion when pluralistic ignorance is
corrected (vs. exacerbated) by providing accurate information. In con-
trast, this accurate information is likely to silence the minority if they
previously believed that their view was more common than it actually
was.
In contrast to our predictions, there were no effects related to
whether participants were led to believe concern was increasing (vs.
decreasing). These results suggest that perceptions that others' opin-
ions are changing to become more (vs. less) like one's own opinion do
not appear to influence willingness to speak up. Instead, perceptions
that others' current opinions differ from one's own reduce willingness
to engage in discussion.
UNCORRECTED PROOF
Journal of Environmental Psychology xxx (2016) xxx-xxx 9
Fig. 3. Effect of perceived opinion climate on a) willingness to discuss climate change and b) expectations of being perceived as competent among participants with different personal
opinions about climate change. (Study 2). Values, simple difference tests and standard error bars depicted in the figure were assessed using simple slopes analyses.
Fig. 4. Expectations of appearing competent and warm as mediators of the effects between opinion congruence and willingness to discuss climate change in Study 2. Perceived Opin-
ions of Others ×Personal Opinion reflects the degree to which one's opinion is congruent with perceptions of others' opinions, with higher values reflecting higher perceived opinion
congruence than lower values. **p< 0.01; *p< 0.05.
UNCORRECTED PROOF
10 Journal of Environmental Psychology xxx (2016) xxx-xxx
8. General discussion
The present research demonstrates the effects of pluralistic igno-
rance in promoting public silence on the socially relevant topic of cli-
mate change. Study 1 results reveal the costs of pluralistic ignorance
on discussion about climate change among those who do not doubt
the science. Survey respondents who did not themselves doubt climate
change were less willing to discuss the topic when they inaccurately
believed fellow students would not share their opinion than when they
accurately perceived they were in the majority. Study 2 results show
that when accurate portrayals of others' beliefs were presented, those
who were concerned about climate change were more willing to dis-
cuss the topic relative to concerned individuals led to inaccurately
believe that others would not share their views. Both studies show
that the reason individuals are more willing to discuss climate change
when they perceive that others agree than when they perceive that oth-
ers disagree is because they expected to be respected more (i.e., appear
more competent) by the former audience. In contrast, expectations of
being liked (i.e., appearing warm) do not play a role facilitating dis-
cussion in either study.
Our findings have practical implications for those who seek to fa-
cilitate greater public engagement with climate change among the ma-
jority of the population that express concern about climate change.
One way to promote discussion is to correct pluralistic ignorance, in-
forming them that a majority of others share their concern. In contrast,
correcting pluralistic ignorance is likely to reduce discussion among
those who are not concerned about climate change. Thus, while the
correction of pluralistic ignorance could increase participation by the
silenced majority group it could decrease participation by those hold-
ing minority opinions if the minority opinion holders previously be-
lieved that they were in the majority, which may not be desirable to
those interested in hearing what individuals who reject the scientific
consensus believe.
Consistent with previous research (Larimer, 2010; Oshagan, 1996),
our results suggest that correcting pluralistic ignorance may only be
effective at facilitating discussion when the opinions of the audience
with whom one will discuss the topic are made salient, rather than
the opinions of the general American public. This points to a practi-
cal weakness of accurately informing individuals of others' opinions:
given geographic and political polarization of climate change concern
(Howe, Mildenbarger, Marlon & Leiserowitz, 2015), in some sub-
groups a majority of people are unconcerned about climate change.
Thus, informing members of these subgroups that a majority of others
are concerned about climate change may be unbelievable or may be
disconfirmed if they talk about climate change within their subgroup.
In these situations, it may be more effective to instead focus on de-
creasing expectations about losing respect from these audiences in a
discussion, as we describe below.
8.1. Impression management and self-silencing
The present research indicates that impression management con-
cerns about losing respect (appearing less competent), but not con-
cerns about being disliked (appearing less warm), explain self-silenc-
ing on the topic of climate change. These results differ from previous
research on confronting prejudice which revealed the silencing effect
of concerns about being disliked (e.g., Shelton & Stewart, 2004; Swim
& Hyers, 1999; Woodzicka & LaFrance, 2001). As we describe in the
introduction, impression management concerns related to not being re-
spected may play a role in self-silencing on climate change due to the
scientific nature of the topic.
These impression management concerns suggest that improving
individuals' confidence in their ability to talk about climate change
(i.e., self-efficacy) could also counteract the negative effects of per-
ceiving themselves to be in the minority by allowing individuals to
overcome expectations of appearing incompetent. This idea is sup-
ported by correlational research suggesting that those with greater
self-efficacy about their ability to discuss climate change discuss the
topic more frequently (Swim et al., 2014) and experimental research
showing that watching short videos about climate change which in-
crease individuals' self-efficacy about their ability to contribute to
conversations bolster willingness to engage in discussions about the
topic (Geiger, Swim, & Fraser, 2016). Individuals could potentially in-
crease their confidence about discussing climate change through learn-
ing about the mechanisms by which climate change is occurring (e.g.,
Ramney & Clark, 2016) or receiving information that most scientists
agree that human caused climate change is occurring (e.g., Cook &
Lewandowsky, 2016), possibilities which could be tested in future re-
search. Confidence could also be obtained via observing others com-
petently discuss climate change. These role models could decrease
concerns about appearing incompetent because they could illustrate
how to talk about climate change (Geiger et al., 2016). Observing high
status individuals discuss climate change could be another potential
buffer to expectations of appearing incompetent. Incorporating more
high-status characters who discuss climate change into the media (e.g.,
Cameron, Weintraub, & Schwarzenegger, 2014; Cornwell, Bajger, &
Higgins, 2015) or promoting greater public discussion among those
holding positions of power (Corner, 2014) could encourage others to
follow their lead. Last, altered social norms could encourage conver-
sations about climate change if the norms promoted discussion about
aspects of the topic that were more accessible to nonscientists, thus in-
creasing individuals' self-efficacy about contributing to these conver-
sations.
8.2. Future directions
Future research might address alternate contexts in which expecta-
tions of being disliked following an interaction would play a role in
silencing. In the present research, Study 1 involved a hypothetical sit-
uation and Study 2 involved a setting where participants did not an-
ticipate repeated interactions with the audience in the future. Expec-
tations of being disliked may play a greater role in self-silencing in
situations where extended contact with others is expected. Contexts
more similar to those found in the prejudice literature could also lead
to increased salience of expectations of being disliked. For instance, in
some contexts speaking up about climate change would require con-
fronting another's environmentally harmful behavior. In this context,
individuals may self-silence out of concern of being appearing hostile
and unfriendly (i.e., cold). Individuals may also expect to be disliked
or rejected from a group if speaking up required directly arguing with
an in-group member who denies the existence of anthropogenic cli-
mate change.
Future research might examine whether individual differences
moderate effects. Those who are high self-monitors (Snyder, 1974),
for instance, may be more likely than others to self-silence due to a
greater tendency to attend to audience opinions. Participant age might
also moderate the findings: older participants may have more practice
negotiating difficult interactions or be less influenced by peer pressure
than the university age students in the present sample (Sears, 1986).
The ubiquity of climate change silence among a variety of demograph-
ics (Norgaard, 2011) and Study 2 results showing that pluralistic igno-
rance can lead to silencing for those on both ends of the opinion spec-
trum suggests that our results may generalize to other demographics
not captured in the present samples.
UNCORRECTED PROOF
Journal of Environmental Psychology xxx (2016) xxx-xxx 11
More detailed assessments of self-silencing should also be consid-
ered in future research to gain a broader insight into this process. Lack
of discussion about climate change could be an intentional choice to
not speak up, or simply indicate lack of personal interest in discussing
the topic, a contrast which should be teased apart in future research.
Future research could also assess self-silencing using behavioral mea-
sures, such as response latency –the tendency to hesitate before shar-
ing one's opinion (Bassili, 2003; Rios & Chen, 2014). Response la-
tency could be practically important to examine because subtle tem-
poral delays can alter the tone of interactions (Pearson et al., 2008).
Using behavioral measures also hold benefits related to not relying on
self-report measures.
Alternative methods for assessing perceptions of others' opinions
could also be assessed in future research. As we note in section 5.3,
these perceptions could be assessed as a continuous measure to deter-
mine whether the relationship between opinion perceptions and will-
ingness to discuss climate change is linear or whether self-silencing
primarily occurs only when individuals believe that a majority of oth-
ers hold opinions directly opposed to their own. These perceptions
could also be assessed using the decision by sampling model (Stewart,
Chater, & Brown, 2006) in which individuals consider what percent-
age of relevant others are perceived to be less concerned about climate
change than they are by comparing themselves to each relevant other
(cf. Wood, Brown, & Maltby, 2012). Wood et al. (2012) found that
using this approach predicts perceived riskiness of individuals drink-
ing behavior more strongly than participants' perceptions of how their
drinking behavior compares to the “average.”
8.3. Conclusion
The present research demonstrates that pluralistic ignorance can be
a barrier to discussions about climate change among those most con-
cerned about climate change and this barrier can be removed by cor-
recting this pluralistic ignorance. Our findings advance the theoretical
understanding of self-silencing, showing that with regard to this topic,
pluralistic ignorance hinders discussions because individuals expect to
be respected less by a dissenting audience, rather than expectations of
being disliked by those who disagree. This suggests that building re-
silience to communicating with audiences of differing opinions may
be facilitated by improving communication skills or boosting effi-
cacy about discussing climate change (Swim et al., 2014; Geiger et al.,
2016). Indeed, a large-scale effort is currently underway to develop
communication skills and efficacy among interpreters at and visitors
to aquariums and zoos around the US (Bunten & Arvizu, 2013; Geiger
et al., 2016; Swim & Fraser, 2013, 2014).
Uncited references
Chia and Lee, 2008; Griskevicius et al., 2010; Hines et al., 2002.
Acknowledgements
This investigation was supported by National Science Foundation,
Grant Number DUE-1239775 to NEAq. Portions of this paper were
presented at the German Environmental Association Conference in
Magdeburg, Germany.
We would like to thank Gamliel Sassoon and Kayla Warner for
their assistance in collecting data and Stephan Lewandowsky and two
anonymous reviewers for their feedback on the paper.
References
Abele, A.E., Wojciszke, B., 2007. Agency and communion from the perspective of self
versus others. Journal of Personality and Social Psychology 93 (5), 751Ajzen, I., &
Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review
of empirical research. Psychological bulletin, 84(5), 888. [NEW PARAGRAPH].
Bashir, N. Y., Lockwood, P., Chasteen, A. L., Nadolny, D., & Noyes, I. (2013). The
ironic impact of activists: Negative stereotypes reduce social change influence. Eu-
ropean Journal of Social Psychology, 43(7), 614-626. [NEW PARA-
GRAPH]Bassili, J.N., 2003. The minority slowness effect: subtle inhibitions in the
expression of views not shared by others. Journal of Personality and Social Psy-
chology 84, 261–276.
Bergsieker, H.B., Shelton, J.N., Richeson, J.A., 2010. To be liked versus respected: di-
vergent goals in interracial interactions. Journal of Personality and Social Psychol-
ogy 99 (2), 248.
Bunten, A., Arvizu, S., 2013. Turning visitors into citizens: using social science for
civic engagement in informal science education centers. Journal of Museum Edu-
cation 38 (3), 260–272.
Cameron, J., Weintraub, T., Schwarzenegger, A., 2014. Years of living dangerously.
Electus international.
Capstick, S.B., Demski, C.C., Sposato, R.G., Pidgeon, N.F., Spence, A., Corner, A.,
2015. Public perceptions of climate change in Britain following the winter 2013/
2014 flooding. . Working paper retrieved from http://c3wales.org/wp-content/
uploads/2015/01/URG-15-01-Flood-Climate-report-final2.pdf.
, Lee, W., 2008. Pluralistic ignorance about sex:the direct and the indirect effects of
media consumption on college students' misperception of sex-related peer norms.
International Journal of Public Opinion Research 20 (1), 52–73.
Clayton, S., Devine-Wright, P., Stern, P.C., Whitmarsh, L., Carrico, A., Steg, L.,
et al., 2015. Psychological research and global climate change. Nature Climate
Change 5 (7), 640–646.
Cohen, J., Cohen, P., West, S.G., Aiken, L.S., 2003. Applied multiple correlation/re-
gression analysis for the behavioral sciences. Routledge, New York.
Cook, J., Lewandowsky, S., 2016. Rational irrationality: modeling climate change be-
lief polarization using Bayesian networks. Topics in Cognitive Science 8 (1),
160–179.
Cook, J., Nuccitelli, D., Green, S.A., Richardson, M., Winkler, B., Painting, R.,
… Skuce, A., 2013. Quantifying the consensus on anthropogenic global warming
in the scientific literature. Environmental Research Letters 8 (2), 024024.
Cook, J., Oreskes, N., Doran, P.T., Anderegg, W.R., Verheggen, B., Maibach, E.W.,
… Nuccitelli, D., 2016. Consensus on consensus:a synthesis of consensus esti-
mates on human-caused global warming. Environmental Research Letters 11 (4),
048002.
Corner, A., 2014. Climate Silence and how to break it. . http://www.climateoutreach.
org.uk/portfolio-item/climate-silence-and-how-to-break-it/.
Cornwell, J.F., Bajger, A.T., Higgins, E.T., 2015. Judging political hearts and minds
how political dynamics drive social judgments. Personality and Social Psychology
Bulletin . .
Dietz, T., 2013. Bringing values and deliberation to science communication. Proceed-
ings of the National Academy of Sciences 110 (Suppl. 3), 14081–14087.
Fiske, S.T., Cuddy, A.J., Glick, P., 2007. Universal dimensions of social cognition:
warmth and competence. Trends in cognitive sciences 11 (2), 77–83.
Fiske, S.T., Cuddy, A.J., Glick, P., Xu, J., 2002. A model of (often mixed) stereotype
content: competence and warmth respectively follow from perceived status and
competition. Journal of Personality and Social Psychology 82 (6), 878.
Fiske, S.T., Xu, J., Cuddy, A.C., Glick, P., 1999. (Dis) respecting versus (dis) liking:
status and interdependence predict ambivalent stereotypes of competence and
warmth. Journal of Social Issues 55 (3), 473–489.
Geiger, N., Swim, J. K., & Fraser, J. (2016). Promoting public discussions about cli-
mate change. Paper presented at the Society for the Psychological Study of Social
Issues, Minneapolis, MN., , Fraser J., 2. . n.
Glynn, C.J., Huge, M.E., 2014. Speaking in spirals:an updated meta-analysis of the
spiral of silence. In: Donsbach, W., Salmon, C.T., Tsfati, Y. (Eds.), The spiral of
Silence: New perspectives on communication and public opinion. Routledge, pp.
65–72.
Goldenberg, S., 2015. US Senate refuses to accept humanity's role in global climate
change, again. The Guardian . Retrieved from http://www.theguardian.com/
environment/2015/jan/22/us-senate-man-climate-change-global-warming-hoax.
Goode, C., Balzarini, R.H., Smith, H.J., 2014. Positive peer pressure: priming member
prototypicality can decrease undergraduate drinking. Journal of Applied Social
Psychology 44 (8), 567–578. http://dx.doi.org/10.1111/jasp.12248.
Griskevicius, V., Tybur, J.M., Van den Bergh, B., 2010. Going green to be seen: status,
reputation, and conspicuous conservation. Journal of Personality and Social Psy-
chology 98 (3), 392.
Habermas, J., 1971. Toward a rational society: Student protest, science, and politics,
Vol. 404. Beacon Press.
UNCORRECTED PROOF
12 Journal of Environmental Psychology xxx (2016) xxx-xxx
Hayes, A.F., 2013. An introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. Guilford Press, New York.
Heider, F., 1958. The psychology of interpersonal relations. Wiley, New York.
Hines, D., Saris, R.N., Throckmorton-Belzer, L., 2002. Pluralistic ignorance and health
risk behaviors: do college students misperceive social approval for risky behaviors
on campus and in media. Journal of Applied Social Psychology 32 (12),
2621–2640.
Holoien, D. S., & Fiske, S. T. (2013). Downplaying positive impressions: Compensa-
tion between warmth and competence in impression management. Journal of ex-
perimental social psychology, 49(1), 33-41. [NEW PARAGRAPH] Howe, P.D.,
Mildenberger, M., Marlon, J.R., Leiserowitz, A., 2015. Geographic variation in
opinions on climate change at state and local scales in the USA. Nature Climate
Change 5 (6), 596–603.
IPCC, 2014. In: Core Writing Team, Pachauri, R.K., Meyer, L.A. (Eds.), Climate
change 2014: Synthesis report. Contribution of working groups I, II and III to the
fifth assessment report of the intergovernmental panel on climate change. IPCC,
Geneva, Switzerland, p. 151.
Jacobson, M.Z., Delucchi, M.A., 2011. Providing all global energy with wind, water,
and solar power, Part I: technologies, energy resources, quantities and areas of in-
frastructure, and materials. Energy Policy 39 (3), 1154–1169.
Kaiser, C.R., Miller, C.T., 2001. Stop complaining! the social costs of making attribu-
tions to discrimination. Personality and Social Psychology Bulletin 27 (2),
254–263.
Lambert, T.A., Kahn, A.S., Apple, K.J., 2003. Pluralistic ignorance and hooking up.
Journal of Sex Research 40 (2), 129–133.
Larimer, M.E., 2010. Group identification as a moderator of the relationship between
perceived social norms and alcohol consumption. Psychology of Addictive Behav-
iors 24 (3), 522–528. http://dx.doi.org/10.1037/a0019944.
Leiserowitz, A., Maibach, E., Roser-Renouf, C., Feinberg, G., Rosenthal, S., 2015. Cli-
mate change in the American mind: March, 2015. Yale University and George Ma-
son University, New Haven, CT (Yale Project on Climate Change Communica-
tion).
Leiserowitz, A., Smith, N., Marlon, J.R., 2010. Americans' knowledge of climate
change. Yale University, New Haven, CT. Yale Project on Climate Change Com-
munication http://environment.yale.edu/climate/files/
ClimateChangeKnowledge2010.pdf.
Leviston, Z., Walker, I., Morwinski, S., 2013. Your opinion on climate change might
not be as common as you think. Nature Climate Change 3, 334–337.
Lewandowsky, S., Oberauer, K., Gignac, G.E., 2013. NASA faked the moon land-
ing—therefore,(climate) science is a hoax an anatomy of the motivated rejection of
science. Psychological Science 24 (5), 622–633.
Lewandowsky, S., Oreskes, N., Risbey, J.S., Newell, B.R., Smithson, M., 2015. Seep-
age: climate change denial and its effect on the scientific community. Global Envi-
ronmental Change 2015 (33), 1–13. http://dx.doi.org/10.1016/j.gloenvcha.2015.02.
013.
Maibach, E., Leiserowitz, A., Roser-Renouf, C., Mertz, C.K., . Identifying like-minded
audiences for global warming public engagement campaigns: An audience seg-
mentation analysis and tool development. PloS one, 6(3), 2011, e17571.. .
Maibach, E., Roser-Renouf, C., Leiserowitz, 2009. Global Warming's six Americas
2009: An audience segmentation analysis. . http://www.
climatechangecommunication.org/images/files/
GlobalWarmingsSixAmericas2009c.pdf.
Mallett, R.K., Melchiori, K.J., 2014. Goal preference shapes confrontations of sexism.
Personality and Social Psychology Bulletin 40 (5), 646–656. http://dx.doi.org/10.
1177/0146167214521468.
Markowitz, E.M., 2012. Is climate change an ethical issue? Examining young adults'
beliefs about climate and morality. Climatic Change 114 (3–4), 479–495.
Marshall, G., 2014. Don't even think about it: Why our brains are wired to ignore cli-
mate change. Bloomsbury Publishing, USA.
Miller, D.T., McFarland, C., 1987. Pluralistic ignorance: when similarity is interpreted
as dissimilarity. Journal of Personality and social Psychology 53 (2), 298.
Moy, P., Domke, D., Stamm, K., 2001. The spiral of silence and public opinion on af-
firmative action. Journalism & Mass Communication Quarterly 78 (1), 7–25.
Neighbors, C., Labrie, J.W., Hummer, J.F., Lewis, M.A., Lee, C.M., Sruti, D.,
… Larimer, M.E., 2010. Group identification as a moderator of the relationship be-
tween perceived social norms and alcohol consumption. Psychology of Addictive
Behaviors 24 (3), 522–528. http://dx.doi.org/10.1037/a0019944.
Noelle-Neumann, E., 1993. The spiral of silence: Public opinion –our social skin, 2nd
ed. University of Chicago Press, Chicago/London.
Norgaard, K.M., 2011. Living in denial: Climate change, emotions, and everyday life.
The MIT Press.
O'Gorman, H.J., Garry, S.L., 1976. Pluralistic ignorance—a replication and extension.
Public Opinion Quarterly 40 (4), 449–458.
Oreskes, N., 2004. The scientific consensus on climate change. Science 306 (5702),
1686.
Oshagan, H., 1996. Reference group influence on opinion expression. International
Journal of Public Opinion Research 8 (4), 335–354.
Pearson, A.R., West, T.V., Dovidio, J.F., Powers, S.R., Buck, R., Henning, R., 2008.
The fragility of intergroup relations: divergent effects of delayed audiovisual feed-
back in intergroup and intragroup interaction. Psychological Science 19 (12),
1272–1279.
Petrič, G., Pinter, A., 2002. From social perception to public expression of opinion: a
structural equation modeling approach to the spiral of silence. International Journal
of Public Opinion Research 14 (1), 37–53.
Prentice, D.A., Miller, D.T., 1993. Pluralistic ignorance and alcohol use on campus:
some consequences of misperceiving the social norm. Journal of Personality and
Social Psychology 64 (2), 243.
Ranney, M.A., Clark, D., 2016. Climate change conceptual change: scientific informa-
tion can transform attitudes. Topics in Cognitive Science 8 (1), 49–75.
Rios, K., Chen, Z., 2014. Experimental evidence for minorities' hesitancy in reporting
their opinions the roles of optimal distinctiveness needs and normative influence.
Personality and Social Psychology Bulletin 40 (7), 872–883.
Rosenberg, S., Nelson, C., Vivekananthan, P., 1968. A multidimensional approach to
the structure of personality impressions. Journal of Personality and Social Psychol-
ogy 9, 283–294.
Rowson, J., 2013. Facing up to stealth denial and winding down on fossil fuels. . Re-
trieved from https://www.thersa.org/discover/publications-and-articles/reports/a-
new-agenda-on-climate-change/.
Salmon, C.T., Kline, F.G., 1985. The spiral of silence ten years later: an examination
and evaluation. In: Sanders, K.R., Kaid, L.L., Nimmo, D. (Eds.), Political commu-
nication yearbook, 1984. Southern Illinois University Press, Carbondale, pp. 3–30.
Salmon, C.T., Neuwirth, K., 1990. Perceptions of opinion “climates”and willingness
to discuss the issue of abortion. Journalism & Mass Communication Quar-
terly 67 (3), 567–577.
Scheufele, D.A., Shanahan, J., Lee, E., 2001. Real talk manipulating the dependent
variable in spiral of silence research. Communication Research 28 (3), 304–324.
Schlenker, B.R., 1975. Self-presentation:managing the impression of consistency when
reality interferes with self-enhancement. Journal of Personality and Social Psy-
chology 32 (6), 1030.
Schroeder, C.M., Prentice, D.A., 1998. Exposing pluralistic ignorance to reduce alco-
hol use among college students. Journal of Applied Social Psychology 28 (23),
2150–2180.
Sears, D.O., 1986. College sophomores in the laboratory: influences of a narrow data
base on social psychology's view of human nature. Journal of Personality and So-
cial Psychology 51 (3), 515.
Sechrist, G.B., Stangor, C., 2001. Perceived consensus influences intergroup behavior
and stereotype accessibility. Journal of Personality and Social Psychology 80 (4),
645–654.
Sechrist, G. B., Swim, J. K., & Stangor, C. (2004). When do the stigmatized make at-
tributions to discrimination occurring to the self and others? The roles of self-pre-
sentation and need for control. Journal of personality and social psychology, 87(1),
111. [NEW PARAGRAPH] Shelton, J.N., Stewart, R.E., 2004. Confronting perpe-
trators of prejudice: the inhibitory effects of social costs. Psychology of Women
Quarterly 28 (3), 215–223.
Singh, R., Ho, L., Tan, H.L., Bell, P.A., 2007. Attitudes, personal evaluations, cogni-
tive evaluation and interpersonal attraction:on the direct, indirect and reverse-
causal effects. British Journal of Social Psychology 46 (1), 19–42.
Snyder, M., 1974. Self-monitoring of expressive behavior. Journal of Personality and
Social Psychology 30 (4), 526.
Stangor, C., Swim, J.K., Sechrist, G.B., DeCoster, J., Van Allen, K.L., et al., 2003.
Ask, answer and announce: three stages in perceiving and responding to discrimi-
nation. European Review of Social Psychology 14, 277–311.
Stewart, N., Chater, N., Brown, G.D.A., 2006. Decision by sampling. Cognitive Psy-
chology 53, 1–26.
Swim, J.K., Eyssell, K.M., Murdoch, E.Q., 2010. Self-silencing to sexism. Journal of
Social Issues 66 (3), 493–507.
Swim, J.K., Fraser, J., 2013. Fostering hope in climate change educators. Journal of
Museum Education 38 (3), 286–297.
Swim, J., Fraser, J., 2014. Zoo and aquarium professionals' concerns and confidence
about climate change education. Journal of Geoscience Education 62 (3), 495–501.
Swim, J.K., Fraser, J., Geiger, N., 2014. Teaching the choir to sing: use of social sci-
ence information to promote public discourse on climate change. Journal of Land
Use & Environmental Law 30 (1), 91–117.
Swim, J.K., Geiger, N., 2016a. Climate change audience segmentation: An alternative
assessment of the “Six Americas”. (Manuscript under review).
Swim, J.K., Geiger, N., 2016b. Perceptions of the prototypical six Americas: Gendered
and valenced attributes. (Unpublished data).
Swim, J.K., Gervais, S.J., Pearson, N., Stangor, C., 2009. Managing the message: us-
ing social influence and attitude change. Strategies to confront interpersonal dis-
crimination. In: Butera, F., Levine, J.M. (Eds.), Coping with minority Status: Re-
sponses to exclusion and inclusion. Cambridge University Press, New York.
Swim, J.K., Hyers, L.L., 1999. Excuse me—what did you just say?!: women's public
and private responses to sexist remarks. Journal of Experimental Social Psychol-
ogy 35 (1), 68–88.
UNCORRECTED PROOF
Journal of Environmental Psychology xxx (2016) xxx-xxx 13
Taylor, D.G., 1982. Pluralistic ignorance and the spiral of silence: a formal analysis.
Public Opinion Quarterly 46 (3), 311–335.
Todorov, A., Mandisodza, A.N., 2004. Public opinion on foreign policy: the multilat-
eral public that perceives itself as unilateral. Public Opinion Quarterly 68 (3),
323–348.
Wood, A.M., Brown, G.D., Maltby, J., 2012. Social norm influences on evaluations of
the risks associated with alcohol consumption: applying the rank-based decision by
sampling model to health judgments. Alcohol and alcoholism 47 (1), 57–62.
Woodzicka, J., LaFrance, A., 2001. Marianne. Real versus imagined gender Harass-
ment. Journal of Social Issues 57 (1), 15–30. Blackwell Publishers.