Explaining Perceptions of Climate Change in the US

AuthorChiara Binelli,Matthew Loveless,Brian F. Schaffner
Published date01 March 2023
Date01 March 2023
Subject MatterArticles
Political Research Quarterly
2023, Vol. 76(1) 365380
© The Author(s) 2022
Article reuse guidelines:
DOI: 10.1177/10659129211070856
Explaining Perceptions of Climate Change
in the US
Chiara Binelli
, Matthew Loveless
, and Brian F. Schaffner
A signif‌icant proportion of the US population does not believe that climate change is a serious problem and immediate
action is necessary. We ask whether individualsexperiences with long-run changes in their local climate can override the
power of partisanship that appears to dominate this opinion process. We merge individual-level data on climate change
perceptions and the main determinants previously identif‌ied by the literature with county-level data on an exogenous
measure of local climate change. While we f‌ind that local climate change signif‌icantly affects percep tions and in the
expected direction, partisanship and political ideology maintain the strongest effect. We then f‌ield a randomi zed online
experiment to test whether partisanship also drives support for pro-climate policies and the willingness to make
environmentally friendly individual choices.
perceptions of climate change, partisanship, public opinion, united states
Despite overwhelming scientif‌ic evidence, a signif‌icant
proportion of the US population does not acknowledge
that climate change is happening (Howe et al. 2015).
Partisanship and, to some extent, political ideology appear
to be the primary explanations for these divergent per-
ceptions (Egan and Mullin 2017). The mechanism
through which partisanship operates in the complex sci-
entif‌ic domain of the climate change issue is two-fold: (1)
citizens have little motivation to look for accurate evi-
dence and (2) they perceive low personal stakes in an issue
that often seems geographically and temporally distant.
Thus, many Americans delegate to partisan elites for
information (Egan and Mullin 2012;2017).
Intuitively, an effective means to form a correct as-
sessment of how the climate is changing would be direct
exposure to the reality of climate change. Mounting ev-
idence shows that personal experience with the daily
weather is more effective at persuading individuals than
statistical information provided by experts because it is
more vivid and accessible. Perceived changes in local
temperature have been linked causally to changes in
global warming beliefs, an effect termed local warming
(Zaval et al. 2014).
Yet, climate change differs from weather change as it
refers to changing weather patterns over a long period of
time. It is not clear whether individuals actually respond to
changes in their local climate. Faced with the diff‌iculty of
measuring how individuals experience climate change,
the literature has taken one of two approaches: (1)
measuring changes in the climate with changes in the
weather (e.g., Egan and Mullin 2012) or (2) exploiting
exposure to extreme weather events (such as excessive
heat, droughts, f‌looding, and hurricanes) that could have a
direct impact on climate change perceptions. Both ap-
proaches have found evidence of a positive relationship
between weather changes and expressions of concern
about climate change.
However, the effects are both modest and short-lived.
Overall, the evidence that weather shapes climate change
opinions is mixed because of the use of weather changes
as a proxy for climate changes, the heterogeneity of the
study populations, and weak causal identif‌ication strate-
gies (Howe et al. 2019).
Does the direct experience of climate change have a
causal impact on climate change perceptions, and can
Department of Political and Social Sciences, University of Bologna, Italy
Department of Political Science and Tisch College, Tufts University,
Medford, MA, USA
Corresponding Author:
Chiara Binelli, Department of Political and Social Sciences, University
of Bologna, Bologna 40126, Italy.
Email: chiara.binelli@gmail.com
this direct experience mitigate the effect of partisanship
on these perceptions? Also, how do partisanship and
experience with climate change drive support for envi-
ronmental policy actions and individual environmental-
friendly choices? Moving beyond perceptions and toward
actual choices is crucial to def‌ine an effective strategy to
enact policies to address climate change. Finding that
partisanship has a diminished effect on subsequent
behavioral choices would open the possibility of pro-
moting important behavioral changes to combat climate
As reliable climate data has become available,
several papers have discussed improved measures of
climate change that use long-term temperature trends
(Howe et al. 2019 for a review). One example is the
methodology proposed by Kaufmann et al. (2017) to
measure local climate change in the US using a county-
level index of the number of days per year for which the
year of the record high temperature is more recent than
the year of the record low temperature over the past
several decades. Exploiting this opportunity, we de-
velop a two-stage empirical analysis. First, we merge
the county-level index of climate change with detailed
individual-level data from the 2014 wave of the Co-
operative Congressional Election Study (CCES) panel
survey to estimate a model of climate change percep-
tions where we control for the climate change index in
the county of residence and all the relevant determi-
nants of climate change perceptions identif‌ied by the
previous literature. Second, we f‌ield a randomized
online experiment to test whether partisanship drives
the willingness to take action to combat climate change
and to engage in environmental-friendly choices (such
as the installation of solar panels or the purchase of
hybrid and electric cars). Whatever the role of parti-
sanship in shaping climate change perceptions, f‌inding
that policy support and individual actions were less (or
not at all) driven by partisanship would provide an
important f‌inding to promote active changes to reduce
emissions. Taken together, this evidence has the po-
tential to def‌ine both the dimensions and the nature of
the political challenges needed to combat climate
Literature Review
Egan and Mullin (2017) review the literature on the at-
titudinal determinants of climate change and identify the
following f‌ive groups of determinants, in order of im-
portance: (1) political preferences (partisanship and po-
litical ideology), (2) demographics, particularly gender
and religiosity, (3) personal experience with climate
change, (4) world views on social relationships (e.g.,
hierarchical versus egalitarian orientation), and (5) media
exposure. Overall, several studies show that partisanship
and political ideology drive Americansopinions on
climate change more than any other factor, including
personal experience with and vulnerability to changes in
the weather (Egan and Mullin 2017).
Yet, one signif‌icant limitation of the literature is the
actual measurement of climate change, most commonly
proxied with various measures of local changes in the
weather or extreme weather events. Specif‌ically, climate
heuristics are calculated by comparing the temperature
during a given day (Zaval et al. 2014;Brooks et al. 2014),
week (Egan and Mullin 2012), season (Akerlof et al.
2013;Howe and Leiseowitz 2013), or year (Goebbert
et al. 2012;Howe et al. 2013) with a long-run average for
the corresponding period and by classifying this anomaly
as either warmer or cooler than average. However,
weather changes and climate change are different: the
daily, weekly, seasonal, or annual deviations from the
mean do not represent a change in the climate. Climate
change is a shift in the long-run weather means. Thus, the
use of weather measures to proxy individualsexperiences
with climate change could potentially introduce a sub-
stantial source of measurement error.
Recent developments in big data collection and
analysis have produced substantially improved measures
of individualsexperiences with climate change. Due to
increased data availability, several papers have proposed
measures of local climate change based on long-term
temperature trends (Howe et al. 2019 for a review).
One example is the work by Kaufmann et al. (2017) that
develops a county-level index of the number of days per
year for which the year of the record high temperature is
more recent than the year of the record low temperature
during a period of the last 30, 40, and 50 years. The index
also allows for a differential impact of more recent and
extreme changes in temperature to capture two key as-
pects that inf‌luence perceptions of climate change: re-
cency weighting and an emphasis on extreme weather
Kaufmann et al. (2017) f‌ind a positive correlation
between the cross-county variation in the index and ag-
gregate changes in the proportion of the US population
that agree that global warming is happening. They also
show that recency weighting is key in that recent record
temperatures have a particularly strong effect on beliefs
and climate skepticism is greater in counties exhibiting
recent cooling versus counties that have warmed. How-
ever, empirical analyses at an aggregate level
correlations between geographical variation in climate
change and the degree of public awareness across geo-
graphical unitslimit our ability to make individual-level
inferences. Further, since weather varies geographically,
geographic patterns of a particular weather variable
may coincide with the geographic distribution of other
366 Political Research Quarterly 76(1)

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT