Unlock document.
This document is partially blurred.
Unlock all pages and 1 million more documents.
Get Access
In a survey of AMS members, perceived scientific consensus was the strongest predictor
of views on global warming, followed by political ideology, climate science expertise, and
perceived organizational conflict.
METEOROLOGISTS’ VIEWS
ABOUT GLOBAL WARMING
A Survey of American Meteorological Society
Professional Members
by Neil SteNhouSe, edward Maibach, Sara cobb, ray baN, aNdrea bleiSteiN, Paul croft,
eugeNe bierly, Keith Seitter, gary raSMuSSeN, aNd aNthoNy leiSerowitz
AFFILIATIONS: SteNhouSe, Maibach, cobb, aNd bierly—
George Mason University, Fairfax, Virginia; baN—The Weather
Channel, Atlanta, Georgia; bleiSteiN—NOAA/Office of the
Assistant Secretary for Environmental Observation and
Prediction, Washington, D.C.; croft—Department of Geology
and Meteorology, Kean University, Union, New Jersey; Seitter
aNd raSMuSSeN —American Meteorological Society, Boston,
Massachusetts; leiSerowitz—Yale University, New Haven,
Connecticut
CORRESPONDING AUTHOR: Neil Stenhouse, George Mason
University, 4400 University Drive MSN 6A8, Fairfax, VA 22030
E-mail: nstenhou@gmu.edu
The abstract for this article can be found in this issue, following the
table of contents.
DOI:10.1175/BAMS-D-13-00091.1
A supplement to this article is available online (10.1175/BAMS-D-13-00091.2)
In final form 19 September 2013
©2014 American Meteorological Society
M ounting an effective societal response to climate
change will require the involvement of the
government, business, and nongovernmental
organization sectors, as well as a myriad of profes-
sional groups and members of the public at large.
Meteorologists and other experts in atmospheric and
related sciences are one group of professionals whose
involvement is particularly important. As experts
on weather and weather prediction, they will play a
variety of important roles in helping other stakeholder
groups make informed decisions based on changing
expectations about climate and weather. Their techni-
cal expertise is complemented by the fact that mem-
bers of the public see climate scientists and broadcast
meteorologists as trustworthy sources of information
on climate change (Leiserowitz et al. 2012).
Research conducted to date with meteorologists
and other atmospheric scientists has shown that
they are not unanimous in their views of climate
change. In a survey of Earth scientists, Doran and
Zimmerman (2009) found that, while a majority
of meteorologists surveyed are convinced humans
have contributed to global warming (GW; 64%), this
was a substantially smaller majority than that found
among all Earth scientists (82%). Another survey,
by Farnsworth and Lichter (2009), found that 83%
of meteorologists surveyed were convinced human-
induced climate change is occurring, again a smaller
majority than among experts in related areas, such as
ocean sciences (91%) and geophysics (88%).
1029
JULY 2014AMERICAN METEOROLOGICAL SOCIETY |
There has been tension in recent years among
American Meteorological Society (AMS) mem-
bers who hold different views on climate change
(Schweizer et al. 2011). Some members have expressed
that their views, which question the view that human-
caused global warming was occurring, are treated
with hostility within the AMS (Schweizer et al. 2011).
In response to this conflict, the AMS created the
Committee to Improve Climate Change Communica-
tion (CICCC) (AMS Committee to Improve Climate
Change Communication 2011). The CICCC’s mission
is to bring all constituencies of opinion to the discus-
sion table and to provide venues and modes of inter-
action that help facilitate respectful and constructive
dialogue on climate change. The CICCC’s mission
does not explicitly include addressing specific areas
of climate science or try to influence the outcomes
of discussions.
To better understand members’ views about cli-
mate change and their perception of any remaining
conflicts about climate change within AMS
membership, the CICCC commissioned George
Mason University researchers to survey AMS mem-
bers; the top-line findings of that survey have been
reported elsewhere (Maibach et al. 2012). In this
paper, we report the results of two additional sets
of analysis. First, to update previous research on
the extent to which meteorologists are convinced
of human-caused global warming, we conducted a
modified replication of Doran and Zimmerman’s
(2009) study. Next, we tested four specific hypotheses
about factors believed to influence meteorologists’
views about climate change, specifically their level
of certainty that climate change is occurring, their
views on whether it is mostly human caused, and
their views on how harmful or beneficial its results
might be. The four hypothesized influencing factors
are climate science expertise, political orientation,
perceived scientific consensus, and perceived conflict
about climate change within AMS; the specific hy-
potheses are presented and explained in detail below.
Last, we analyzed open-ended responses from survey
participants about the nature of the conflict about
climate change within AMS; these findings will be
reported in a subsequent paper.
LITERATURE REVIEW AND HYPOTHESES.
Climate science expertise. Previous research using
survey data (Doran and Zimmerman 2009) and
citation analysis (Anderegg et al. 2010) has suggested
that greater expertise in climate science, measured in
terms of academic background and publishing record,
is associated with higher conviction that human-
caused global warming is occurring. For example,
in Doran and Zimmerman’s survey study, while only
82% of the total sample indicated they are convinced
that humans have contributed to global warming,
89% of active publishers in the peer-reviewed scien-
tific literature and 97% of climate experts who publish
primarily on climate change in the peer-reviewed
scientific literature indicated they were convinced
(Doran and Zimmerman 2009; Kendall Zimmerman
2008). As a result, our first hypothesis is as follows:
H1: As compared with professionals with less
expertise in climate change, professionals with more
expertise will have higher levels of personal certainty
that global warming is happening, will be more likely
to view it is as mostly human caused, and will be more
likely to view it as harmful rather than beneficial.
Political ideology. Decision making about how to mount
an effective societal response to climate change in
the United States has been complicated by increasing
polarization over the issue, which has occurred
largely along political lines. In the late 1990s, similar
proportions of liberals and conservatives saw global
warming as real; by 2008 (Dunlap and McCright
2008)—and continuing to the present (Leiserowitz
et al. 2012)—large differences had emerged such
that liberals were more likely to see it as real, and
conservatives had become increasingly skeptical.
This growing polarization appears not to be caused by
differences in scientific understanding—indeed, most
Americans know very little about the science of global
warming (Leiserowitz et al. 2010)—but rather by dif-
ferences in political ideology and deeper underlying
values (Kahan et al. 2011). Many conservatives see
the solutions proposed to mitigate global warming as
being more harmful than global warming itself due
to their effect on the economy (McCright and Dunlap
2011). Liberals, on the other hand, are more likely to
accept the dominant scientific view, as they see the
proposed responses to global warming as strengthen-
ing activities they value—namely, protection of the
environment and regulation of industrial harm.
One might expect scientists’ norms of objectivity
to prevent their political ideology from influencing
their evaluation of scientific findings. Indeed, in one
study scientists’ opinions on global warming policy
responses varied by political ideology, but their views
on the basic science did not (Rosenberg et al. 2010).
However, other studies suggest scientists’ views on
science can be influenced by ideology. A survey
of members of the American Association for the
Advancement of Science showed that conservatives
1030 JULY 2014
|
differed dramatically from liberals with regard to
their views about the science of global warming
(Nisbet 2011): 44% of conservatives saw global
warming as mostly due to human activities, com-
pared to about 94% of liberals. We therefore hypoth-
esized the following:
H2: As compared with professionals with a more
conservative political orientation, professionals with
a more liberal political orientation will have higher
levels of personal certainty that global warming is
happening, will be more likely to be view it is as
mostly human caused, and will be more likely to view
it as harmful rather than beneficial.
Perceived scientific consensus. Public opinion research
has shown that only a minority of the public (44%)
agree that “most scientists think global warming
is happening” (Leiserowitz et al. 2012, p. 18). Also,
preliminary analyses of the data from the present
survey found that only 59% of AMS members agree
that 81%–100% of climate scientists think that
global warming is happening (Maibach et al. 2012).
Members of the public who perceive agreement about
global warming among scientific experts are more
likely to view global warming as real, human caused,
and harmful than people who do not perceive agree-
ment in the scientific community (Ding et al. 2011;
Dunlap and McCright 2008; Krosnick et al. 2006).
We therefore hypothesized the following:
H3: As compared with professionals who perceive less
scientific consensus about global warming, profes-
sionals who perceive more scientific consensus will
have higher levels of personal certainty that global
warming is happening, will be more likely to be view
it as mostly human caused, and will be more likely to
view it as harmful rather than beneficial.
Perceived conflict. Schweizer et al. (2011) found that
broadcast meteorologists who perceived conflict
about global warming among their peers had
disengaged from the issue, and they reported hav-
ing done so because of the pressure they felt from
committed partisans in the conflict to “choose
sides.” This finding parallels an argument that has
been advanced in the political science literature that
increasing levels of polarization in American politics
has caused the moderate majority to disengage from
political participation (Fiorina and Abrams 2008).
One proposed reason for this political disengagement
is that moderate voters feel no incentive to participate
when they see the debate as primarily being driven by
ideological positioning rather than a wish for problem
solving (Fiorina and Abrams 2008). Schweizer et al.
(2011) found similar sentiments expressed by broad-
cast meteorologists in relation to conflict about
climate change within AMS. We believe that, when
members of a professional group (including but not
limited to meteorologists) perceive conflict within
their peer group, they will withdraw from the conflict
by moderating their views on the issue that is causing
the conflict.
This moderation of views would entail AMS mem-
bers on both ends of the spectrum of views shifting
closer toward the center. However, because we
expected a larger number of members to have begun
with views favoring human-caused global warming
(prior to perceiving conflict at the AMS), we also ex-
pected any moderating influence to result in members
revising their views of global warming downward
more often than upward. For this reason, when
averaged across the whole population of members,
we would expect greater perception of conflict to
be associated with a reduced level of conviction that
human-caused global warming was occurring. We
therefore hypothesized the following:
H4: As compared with professionals who perceive less
conflict about global warming within the member-
ship base of their professional society, professionals
who perceive more conflict will report lower levels of
personal certainty that global warming is happening,
will be less likely to be view it is as mostly human
caused, and will be less likely to view it as harmful
rather than beneficial.
METHOD. On 29 December 2011, we e-mailed
all 7,197 AMS members for whom AMS had an
e-mail address, excluding associate members and
student members. The e-mail—signed by the CICCC
chairs—requested participation in our survey
and provided a link to the web-based survey form
(including the consent form). On 6 January 2012,
and again on 11 January 2012, participants who had
not yet responded received reminders by e-mail. On
27 March 2012, 375 participants who had given one
specific answer (to a question on global warming
causation) were e-mailed a request to answer one
additional question (designed to clarify their view on
causation). This question is described further below.
Of the 7,197 people invited to participate, 135 peo-
ple were ineligible due to invalid addresses. Therefore,
the valid denominator of our sample was 7,062. Of
these, 1,854 people completed at least some portion
of the survey beyond the consent form, yielding a
1031
JULY 2014AMERICAN METEOROLOGICAL SOCIETY |
minimum response rate of 26.3% (which assumes
that all nonrespondents were eligible to participate);
this is slightly lower than the average rate for web
surveys (Shih and Fan 2008). Of the 375 people who
were sent the follow-up question, 271 responded. One
of these respondents refused to answer the question,
for a response rate to the follow-up question of 270
out of 375, or 72%.
Independent variables. exPertiSe. To assess climate
science expertise, we examined three measures:
whether respondents indicated climate science as
their area of expertise, their highest degree obtained,
and their peer-reviewed publishing record on climate
change over the last five years. Using principal axis
factor analysis, we found these measures formed a
single factor; loadings for all variables were within
the acceptable range (Gorsuch 1983). We therefore
decided to sum all three variables into an index
of climate science expertise. For area of expertise,
participants who indicated climate science as their
area of expertise were scored 1; all others were scored
0. For highest degree obtained, participants who
answered “PhD (or other doctoral degree)” and “MS
or MA” were scored 2 and 1, respectively; all others
were scored 0. For publishing record, respondents
who said they published more than 50% of their
21%–40%, 41%–60%, 61%–80%, 81%–100% (coded
1–5, respectively), and “I don’t know enough to say.”
Respondents who indicated “I don’t know” (n = 156)
were excluded from the analyses, so that perceived
consensus could be analyzed as a single continuous
variable. To assess if the exclusion of “Don’t know”
respondents influenced our findings, we conducted
additional analyses with all respondents using
dummy coding, a statistical technique that allows
noncontinuous variables to be entered into linear
regressions. This produced no significant change to
the results.
Perceived coNflict. Respondents were asked if they
agreed or disagreed with the statement, “There
is conflict among AMS members on the issue of
global warming.” Response options were “Strongly
disagree,” “Somewhat disagree,” “Neither agree nor
disagree,” “Somewhat agree,” and “Strongly agree.”
The responses were coded 0–4, respectively.
Dependent variables. certaiNty. To establish the cer-
tainty of respondents’ views on global warming, we
asked two questions. The first asked whether global
warming is happening, and the second assessed the
respondents’ level of certainty that it is happening.
The first question was worded as follows: “In this
