This article was downloaded by: [Joanna Huxster]

On: 29 June 2015, At: 15:13

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Journal of Environmental Education

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/vjee20

Undergraduate Understanding of Climate

Change: The Influences of College Major
and Environmental Group Membership
on Survey Knowledge Scores
a

Joanna K. Huxster , Ximena Uribe-Zarain & Willett Kempton

a

Drexel University, Philadelphia, Pennsylvania, USA

University of Delaware, Newark, Delaware, USA

Published online: 29 Jun 2015.

Click for updates

To cite this article: Joanna K. Huxster, Ximena Uribe-Zarain & Willett Kempton (2015) Undergraduate
Understanding of Climate Change: The Influences of College Major and Environmental Group
Membership on Survey Knowledge Scores, The Journal of Environmental Education, 46:3, 149-165,
DOI: 10.1080/00958964.2015.1021661
To link to this article: http://dx.doi.org/10.1080/00958964.2015.1021661

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
Content) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

THE JOURNAL OF ENVIRONMENTAL EDUCATION, 46(3), 149165, 2015

C Taylor & Francis Group, LLC
Copyright 
ISSN: 0095-8964 / 1940-1892 online
DOI: 10.1080/00958964.2015.1021661

Undergraduate Understanding of Climate Change: The

Influences of College Major and Environmental Group
Membership on Survey Knowledge Scores
Joanna K. Huxster

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

Drexel University, Philadelphia, Pennsylvania, USA

Ximena Uribe-Zarain and Willett Kempton

University of Delaware, Newark, Delaware, USA

A survey covering the scientific and social aspects of climate change was administered to examine
U.S. undergraduate student mental models, and compare knowledge between groups based on major
and environmental group membership. A Knowledge Score (scale 035, mean score = 17.84) was
generated for respondents at two, central East Coast, U.S. universities (n = 465). Elements of student
mental models examined include environmental issue confusion, skepticism, and self-reported understanding. This study finds that students frequently confuse climate change with other environmental
issues, and that a substantial majority of students do not have an understanding of climate change
that closely matches the scientific model. These misconceptions extend to their understanding of
mitigation actions. Environmental group membership is shown to be a greater determinant of climate
change knowledge than enrollment in a science major.
Keywords climate change, mental models, public understanding of science, science literacy

The scientific community has reached a high level of consensus about the occurrence of anthropogenic global climate change (IPCC, 2013; Oreskes, 2004; Rosenberg, Vedlitz, & Cowman,
2010). In the United States, outside of the scientific community, there continues to be considerable debate and confusion surrounding the topic. Studies of public understanding of climate
change conducted in the 1990s by Kempton, Boster, & Hartley (1995) and Bostrom, Morgan,
Fischoff, & Read (1994) found that members of the general public frequently held inaccurate
mental models regarding the processes of climate change. Participants in these studies had often gathered misinformation and combined multiple environmental problems into their mental
models. Since the 1990s, climate change discussion and science have both advanced; however,
the level of public understanding has not noticeably increased. A study by the Yale Project on
Climate Change Communication (Leiserowitz, Smith, & Marlon, 2010) and a retest of Bostrom
et al.s 1994 study (Reynolds, Bostrom, Read, & Morgan, 2010), both revealed little change in the
Correspondence should be sent to Joanna K. Huxster, Drexel University, Philadelphia, Pennsylvania, USA. E-mail:
jkh49@drexel.edu
Color versions of one or more figures in this article are available online at www.tandfonline.com/vjee.

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

150

STUDENT UNDERSTANDING OF CLIMATE CHANGE

understanding of the United States public. The new findings again included persistent confusion
between climate change and other environmental issues such as ozone depletion.
This article examines the knowledge and mental models of an important indicator population, undergraduate students. They represent the future leaders and thinkers of the country, and,
as voters, they are capable of contributing to political and social change in the present. Their
understanding of this urgent, socioscientific issue is critical to the future of the global climate.
Significant recent work, such as that of Markowitz, Hodge, & Harp (2014), has been done to
increase climate change communication efforts. Understanding the mental models of respondents in higher education will help evaluate the efficacy of the current education and science
communication efforts and identify ways in which they can be improved.
This study samples students nearing the end of their undergraduate programs and evaluates
the understanding they have accumulated and retained throughout their academic careers. Semistructured interviews (n = 18) were conducted to develop and refine survey questions at one
public, central East Coast, U.S. university. The resulting survey was administered to respondents
from two public, central East Coast, U.S. universities (n = 853, completed to end of survey n
= 465) and Knowledge Scores for these respondents were generated based on their responses.
In this article, these Knowledge Scores, as well as the knowledge elements of participant mental
models observed through interview and survey data, are analyzed.

AMERICAN ENVIRONMENTAL VALUES AND SCIENTIFIC

UNDERSTANDING
In their book Environmental Values in American Culture, Kempton et al. (1995) conduct semistructured interviews and surveys to understand American cultural models of the environment,
and of specific issues, including climate change. Kempton et al. (1995) define a cultural model or
mental model as a set of guidelines or an imaginary map each individual builds for his or herself
about any particular subject through which he or she can run scenarios, analyze information, and
reach conclusions (1995, p. 10). A further examination of the literature on mental models and
their implications can be found in the next section of this literature review.
In Cultural Models of Weather and the Atmosphere from Kempton et al. 1995, the authors
closely examine their informants cultural models of climate change (or as it was more commonly
called in 1995, the greenhouse effect). Overall, they find that the responses of their laypersons
do not correspond to scientific models. The authors describe ways in which informants modeled
climate change that are based upon models for previously existing concepts, all of which deviate
from the scientific understanding. Some of these concepts include greenhouse gases as pollution,
or the pollution model, and ozone depletion as the mechanism of climate change, or the ozone
depletion model (Kempton, Boster, & Hartley, 1995).
A 2010 study done by the Yale Project on Climate Change Communication reveals some
similar results to the Kempton et al. studies (Leiserowitz et al., 2010). In the Yale study, 2,030
adult members of the American public were surveyed on their knowledge of climate change and
were given grades (AF) on their performance. A majority of those surveyed received Bs, Cs,
and Ds on their knowledge, with the largest proportion scoring Cs and only 2% earning an A.
The majority of the adults surveyed had heard of the greenhouse effect and could correctly
identify carbon dioxide as a greenhouse gas, but few understood how great the increase in carbon

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

HUXSTER ET AL.

151

dioxide in the atmosphere has been since 1850. A majority were found to correctly understand
that burning of fossil fuels contributes to climate change, but the study also finds that majorities
of Americans. .. incorrectly believe that the hole in the ozone layer, toxic wastes, aerosol spray
cans, volcanic eruptions, the sun, and acid rain contribute to global warming (2010, p. 11). The
results of their understanding could be seen in the respondents listed solutions. These included
both correct (reducing driving, switching to renewable energy sources) and incorrect (reducing
toxic waste, banning aerosol spray cans) actions.
The Yale study also documents the pervasiveness of climate skepticism in the American
publics understanding of climate change. Many of those surveyed did not believe that scientists
can predict future weather or that computer models are reliable. A third of those surveyed stated
that they did not believe that humans cause climate change. More than half of the respondents said
that they arent sure if climate change is happening, or that they do not believe it to be happening
at all. Over one third said that there is much disagreement among scientists about whether climate
change is happening. The respondents listed television, printed media, and the Internet as their
main sources of information about climate change, and a majority stated that they needed more
and better information on the subject (Leiserowitz et al., 2010).
Overall, the literature reveals a lack of scientific understanding of the processes of climate
change within the general public. The consequences of this lack of understanding tend to be
misinformation on the possible effects and severity of the issue, and the adoption of ineffective
mitigation actions, as well as possibly weakened support for governmental initiative on the issue.
RESEARCH QUESTIONS
1. What levels of knowledge, as measured by a calculated Knowledge Score, do respondents
at these two universities hold to explain climate change?
2. What knowledge elements of mental models are represented, and how do the respondents
models compare to the scientific model of climate change?
3. How does the participants knowledge of climate change, as defined by the Knowledge Score, differ between respondents in climate-related science majors, respondents
belonging to environmental groups, and respondents in neither category?
MENTAL MODEL THEORY AND CLIMATE CHANGE
Mental models are an important component of this study, as they are one of the tools by which
respondents understanding will be described and measured. Mental models, also called cultural
models or cognitive models, come from a subfield of anthropology known as cognitive anthropology. DAndrade (1995, p. 151) defines cognitive models as a small-scale model of external
reality and possible personal actions that people carry around in their heads. It consists of an
interrelated set of elements which fit together to represent something and this model is used to
reason or to calculate by mentally manipulating its parts to solve a problem. In cognitive science,
mental models are commonly used to explain human reasoning. It is theorized, people do not
reason using abstract rules, but rather they construct and combine mental models and generate
inferences consistent with those models (Bly & Rumelhart, 1999, p. 156). People can run
their mental models to come up with conclusions or solutions to a problem. Since mental models

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

152

STUDENT UNDERSTANDING OF CLIMATE CHANGE

are constructions of human thought, they often include simplifying assumptions (Mudditt, 1996).
Errors in conclusions, therefore, will likely not be random, but will instead correspond with gaps
or errors in mental models. Studies show that mental models, complete with gaps or distortions,
can be passed on, taught and transferred (Bly & Rumelhart, 1999).
Mental models are not only seen as agents of reasoning, but also are linked to motivation and
action. DAndrade (1992, p. 2324), in an essay from the book Human Motives and Cultural
Models, describes mental models as having a motivational force because they describe the
world, set forth both conscious and unconscious goals, and elicit desires. Holland and Quinn
(1987, p. 9) specifically address how mental models motivate behavior, expressing that the basis
of mental models directive force is in the authority and expertise with which they are invested
and within the intrinsic persuasiveness the models hold for us. Indeed, the influence of mental
models on the actions and perceptions of their holders can be seen in a variety of studies (Gentner
& Stevens, 2014). The examples range from peoples interactions with their thermostat settings,
which is influenced by mental models informants hold for how the thermostat works (Kempton,
1987) to middle school students understanding of the greenhouse effect (Shepardson, Choi,
Niyogi, & Charusombat, 2011) to students models of their personal affect on the environment
and the behaviors those mental models produce (Liu & Lin, 2015).
This research will use respondent survey responses to reconstruct and examine the knowledge
elements of participant mental models of climate change. Mental and cultural model theory has
been used as the theoretical framework for numerous studies regarding public understanding
of climate change. In Kempton (1991) and Kempton et al. (1995), the studies most influencing
the methodology of the present research, semi-structured interviews and survey results were
used to infer mental models of climate change held by members of the general public. The
Kempton et al. (1995) study found that laypersons hold several overarching models to explain climate change. These same models were found in research conducted by Bostrom et al.
(1994) and in a retest by Reynolds et al. (2010) using a similar survey instrument. In 2010, Zia
and Todd used mental and cultural model theory to examine and map the belief networks of
members of the public as related to their ideological, religious, socioeconomic, and educational
background.
These and other studies set the methodological and conceptual precedent for the use of mental
model theory to explain understanding of climate change in this study. This article analyzes the
knowledge portion of a larger dissertation study (Huxster, 2013).

DESIGN AND METHODOLOGY

The methodology of this research contains elements of both ethnography (from cultural anthropology) and survey research, like the work done by Kemptonand colleagues in their 1995
book. Kempton et al. (1995) conducted semi-structured interviews of members of the general
public. The ideas, reasoning, and mental models expressed in these interviews were then used
to construct survey statements to be administered to a larger population. This methodology, also
described in Bernard (2002), allows the researcher to test the existence of the mental models
seen in semi-structured interviews across a larger population. The mixed methods analysis of
Kempton et al.s (1995) data included the examination of informant quotations and survey responses to infer the mental models of the sampled population regarding climate change. This

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

HUXSTER ET AL.

153

methodology is applied in the present study to illuminate the mental models of U.S. undergraduate
students, assess their knowledge, and understanding of climate change, and answer the research
questions.
Prior to the development of the survey instrument, a pilot study was conducted consisting
of semi-structured interviews of 18 university students. Semi-structured interviews include a
pre-determined set of questions, but the interviewer is able to ask further questions and probe
for more complete answers, depending on how the informants respond (e.g., Agar, 1980). The
interviews conducted for this study were constructed of open-ended questions aimed at gathering
information about students understanding, knowledge, and lines of reasoning regarding climate
change. Transcripts of these semi-structured interviews were reviewed, and the mental models
of these informants were inferred from the lines of reasoning displayed. This process involved
the identification and systematic counting of keywords, statements, and ideas, and reconstruction
of participant perceptions and models based on inferred logic. The comments and arguments of
the informant mental models were then condensed into one-sentence statements on the survey.1
Representative statements from all of the displayed mental models were included to insure that
even the models seen infrequently in the 18 interviews were represented and could be tested in
the overall student population. Survey respondents were asked to indicate their agreement to the
survey statements on a 5-point Likert scale.2
Using this methodology for survey creation results in some multi-part statement wording not
used in simple opinion surveys. Since the statements on this survey draw from the interviews
with the same population, the wording can be informal yet appropriate to the student interview
subjects. Also, since we are attempting to elicit mental models rather than isolated beliefs, some
survey statements include multipart ideas that attempt to capture the mental models and lines
of reasoning displayed in the pilot study interviews. As noted by Bernard (2002) and Kempton
et al. (1995), this survey item methodology is used for testing for lines of reasoning or multi-part
mental models; the items are not intended to elicit individual components facts. The included
survey questions can be seen in Table 1 and the full survey and interview instruments can be
found in Huxster (2013).
The informants in the semi-structured interviews held several different mental models, but
their reasoning and understanding could be categorized into three overarching models: predominantly scientific understanding, environmental issue confusion, and natural causes and cycles.
After the survey was administered, exploratory and confirmatory factor analyses were performed
to provide evidence of instrument and construct validity, and it was found that the survey did
measure the three main components of mental models observed in the pilot study. In the confirmatory factor analysis performed with varimax rotation, three factors were identified (Table 1).
These three factors fit closely with the models seen in the semi-structured interviews. The first
factor includes almost all of the scientifically correct statements about the mechanisms, causes,
scientific consensus, and severity of climate change and has been labeled Predominantly Scientific Understanding. The items loading into the second factor are statements that attribute the
causes or mechanisms of various environmental issues to climate change, but these causes are not
actually significant to the issue. This factor is labeled Issue Confusion. The statements in the
third factor all refer to skepticism about the severity and scientific certainty of climate change, as
well as statements expressing a belief that natural causes are responsible for climate change. This
factor has been labeled Natural Causes and Skepticism. One survey item was discarded after
the factor analyses owing to a misprint in the Likert scale wording. A more detailed explanation

154

STUDENT UNDERSTANDING OF CLIMATE CHANGE

TABLE 1
Factor Loadings of the Rotated Factors for Knowledge Variables

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

Sci Und.
2.40 We need to change our transportation system, like creating more public
transportation, in order to make a big impact on reducing climate change
2.37 Individuals can reduce climate change by not buying gas-guzzling cars
and instead maybe owning a more fuel-efficient car or hybrid
2.43 The government could put more resources into developing alternative
and renewable sources of energy in order to reduce climate change
2.42 Ive heard that using efficient light bulbs, turning off electric appliances
and insulating my house are all ways I can reduce my contribution to
climate change
2.25 Even though its sometimes debated, I feel that its pretty well
documented that humans have released too much carbon dioxide into the
atmosphere
2.32 The government could tax carbon emissions or introduce a system of cap
and trade in an effort to reduce climate change
2.30 The best way to deal with climate change would be to reduce or
eliminate carbon-based fuel sources
2.23 Burning fossil fuels releases carbon dioxide into the atmosphere, which
traps heat and causes climate change
2.31 Individuals can give political support to clean energy initiatives to help
deal with climate change
2.39 Cutting down on waste and pollution will help stop the melting of the ice
caps
2.74 Even if the science isnt clear to me, I think its better to act and be safe
rather than sorry
2.27 Ive heard that driving less and using bicycles and public transportation
are ways to reduce climate change
2.44 Recycling is essential to reducing climate change because it keeps
plastics out of landfills and reduces production
2.18 Climate change is caused by an increased amount of gases like carbon
dioxide and methane in the atmosphere
2.75 Its hard for scientists to be certain about anything, but I think the patterns
and trends that theyre seeing in climate change have them pretty convinced
2.68 I think that its generally pretty well accepted that climate change is
occurring
2.14 Gas and chemical emissions from our cars deplete the ozone and cause
climate change
2.11 Climate change is happening because we are depleting the ozone layer,
and that lets in more heat from the sun
2.16 Greenhouse gases make the hole in the ozone layer worse
2.15 Toxic pollution in the atmosphere causes climate change
2.21 Climate change happens because we release chemicals from aerosol cans
into the atmosphere
2.36 I can help prevent climate change by not buying or using aerosol cans
2.22 Nuclear power plants create pollution that contributes to climate change
2.35 The government can reduce climate change by regulating toxic pollution

Issue
Conf

Nat Cause
and Skep

.710
.687
.672
.624
.619
.616
.611
.600
.571
.512

.401

.508
.507
.506
.502
.443
.425
.782
.778
.724
.710
.681

.444

.625
.608
.561
(Continued on next page)

HUXSTER ET AL.

155

TABLE 1
Factor Loadings of the Rotated Factors for Knowledge Variables (Continued)

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

Sci Und.
2.19 Climate change involves magnetic field and fluctuations in the sun, and
other things beyond our control, and those things are bigger factors than
greenhouse gas emissions
2.70 The climate has changed a lot in the past, so I think we might be making
too big a deal about how its changing now
2.62 From my sources Ive learned that too much emphasis is put on the
human element of climate change, rather than the natural elements
2.9 I think that climate change concerns are exaggerated more than the science
supports
2.69 I think that scientists are certain that the climate is changing, but not
about who or what is responsible
2.20 Natural thing like volcanoes and water vapor have more of an impact on
climate change than human actions
2.73 I dont think that climate change is the most pressing environmental issue
we face today, and I think that it gets too much press
2.10 Scientists are highly certain that humans are definitely the cause of
current, rapid climate change
2.12 Climate change is caused by the greenhouse effect

Issue
Conf

Nat Cause
and Skep
.672
.667
.654
.637
.616
.613
.612
.535
.450

Extraction Method: Principal Component Analysis.

Rotation Method: Varimax with Kaiser Normalization.

of the confirmatory factor analyses can be found in the supplementary material accompanying
the online article.
The survey was administered to a sample of junior and senior students at two public, central
East Coast universities. These respondents were selected for the survey as they have spent three
or four years studying their individual majors and therefore better represent the influence that a
students major may have on their understanding of climate change. Systematic sampling was
used to gather a representative sample of students and the survey was conducted online using
Qualtrics survey software. Eight hundred fifty-three respondents from the two universities took
the survey, with 465 respondents from the two universities completing the entire survey. For
the comparison of climate-related science majors and non-science majors, a selection of majors
was made including majors relating to environmental science and policy, chemistry, biology, and
geology.
In order to acquire sufficient response rates from majors of interest with fewer students, the
geology majors and the Environmental Science and Policy Group were sampled 100%. For
all other majors, proportional stratified sampling was used. Within the departmental subgroups,
systematic sampling was used to select the final sample, with an interval of N = 4. In the statistical
analysis of the survey results, those majors that were sampled at 100% with high response rates
were weighted at .25 when used in conjunction with the other majors. Following the researchers
sampling, the registrars office of each University performed the selection of students from the
appropriate majors, and sent them e-mails asking them to take the online survey. The identities

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

156

STUDENT UNDERSTANDING OF CLIMATE CHANGE

of the individual respondents sampled were not revealed to the researcher and no incentives
were offered to the students for participation. To ensure higher numbers of participants in the
selected majors of interest, respondents from capstone courses in those respective majors were
also sampled based on identification of those students by their professors. Beyond the initial
survey invitation and reminder emails sent to the entire sample, the researcher did not contact
individual respondents.
The surveys Knowledge Variables, analyzed here, include both correct and incorrect statements about the mechanisms of climate change, as well as the actions that can be taken to
prevent or mitigate climate change. In a socioscientific issue such as climate change, the social components cannot be completely separated from the science (Klosterman & Sadler, 2009).
The survey Knowledge Variables also involve those social components most closely associated
with the scientific model and that can be coded as correct and incorrect in comparison with the
scientific model. The social aspects of these Knowledge Variables are most clearly represented
in statements dealing with scientific consensus and government action. A complete list of the
Knowledge Variables is included in the factor analysis in Table 1.
As part of the analysis of the Knowledge Variables, a Knowledge Score was generated
for each respondent based on respondents agreement and disagreement with scientifically correct and incorrect statements. The Knowledge Score was calculated from 35 survey statements.
Students received 1 point for each statement matching the scientific model of climate change
to which they agreed. Students also received 1 point for each statement not matching the
scientific model to which they disagreed. This created a score on a scale of 0 to 35. Only
the answers of the 465 participants who completed the survey in its entirety were included
in the calculation of the Knowledge Score. The mean Knowledge Scores of science and nonscience students were compared using a t-test, as were the mean knowledge scores of environmental group members and respondents not belonging to an environmental group. Two-Way
Analysis of Variance was used to compare mean Knowledge Scores between four Sets of
respondents:
1.
2.
3.
4.

Science major and Environmental group

Science major and Non-environmental group
Non-Science major and Environmental group
Non-science major and Non-environmental group.

Thus, we can compare the influence of major and environmental group on climate change
knowledge.
In addition to the calculation and analysis of the Knowledge Scores, responses to individual survey items dealing with the three overarching types of mental models seen in the
interviews are analyzed. The first set of statements examined are those similar to the scientific model, and are included as examples of respondents responses to individual parts of the
model. Statements showing confusion between climate change and other environmental issues
were included to illustrate ways in which respondents mental models differ from the scientific
model. Incorrect statements attributing current, rapid climate change3 to natural causes were
included in this analysis to show the role skepticism plays in respondents mental models. Finally, survey statements asking participants to self-report their knowledge and understanding
of climate change were included to examine how their self-assessment fits into their mental
models.

HUXSTER ET AL.

157

TABLE 2
Descriptive Statistics for Two Public, Central East Coast
Universities Knowledge Scores
N
Mean
Median
Mode
Std. Deviation
Variance

465
17.84
18.00
16.00
5.40
29.26

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

RESULTS AND ANALYSIS

The creation of the Knowledge Score allowed for the evaluation of climate change knowledge of
the respondents, and comparison of knowledge across specific groups of respondents. Incomplete
surveys were excluded from this analysis, as their scores skewed the results. The mean Knowledge
Score (scale 035) of the 465 respondents (both universities) was 17.84 with a standard deviation
of 5.40. The highest score received was a 35, indicating all answers matched the scientific model,
and the lowest was a 0, indicating that no answers matched the scientific model. The statistics for
the Knowledge scores can be seen in Table 2 and the score distribution can be seen in Figure 1.
Comparison of Knowledge Scores Between Groups
Descriptive statistics were performed and compared between groups of respondents based on
their science education and their environmental group membership. The first comparison was the
science majors as compared with non-science majors. An independent samples t-test was run and

FIGURE 1 Student respondent knowledge score distribution.

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

158

STUDENT UNDERSTANDING OF CLIMATE CHANGE

FIGURE 2 Mean knowledge scores for four core groups.

a significant difference was found between the means of the two groups (t(448) = 3.46, p < .05).
The mean Knowledge Score of the science majors was significantly higher (m = 19.60, sd =
5.98) than the non-science majors (m = 17.43, sd = 5.19). Based on these results, it can be said
that respondents in the selected science majors, unsurprisingly, have a greater understanding of
climate change science than respondents who were not science majors.
The next comparison was between participants who belong to an environmental group with
those who do not. An independent samples t-test comparing the two found a significant difference
between the means (t(456) = 4.44, p < .05). The mean Knowledge Score of environmental group
respondents was significantly higher (m = 20.11, sd = 5.04) than that of the non-environmental
group respondents (m = 17.36, sd = 5.35).
In order to examine the effects of both environmental group membership and science major, a
2 (Environmental Group Membership) X 2(Science Major) between-subjects factorial ANOVA
was calculated comparing the Knowledge Scores of the respondents. Environmental group membership and science major were each shown to have significant main effects. The respondent Sets
and their corresponding mean scores are as follows:
1.
2.
3.
4.

Science major and Environmental group (m = 21.38),

Science major and Non-environmental group (m = 18.36),
Non-Science major and Environmental group (m = 19.5),
Non-science major and Non-environmental group (m = 17.13).

A significant difference was found between the groups F(3, 447) = 9.60, p < .05) and
post hoc analysis shows that environmental group membership is a greater indicator of a high
Knowledge Score than enrollment in a science major. The interaction between environmental
group membership and science major was not significant (F(1, 447) = 6.36, p > .05). A graph
of the mean scores (Figure 2) shows an obvious difference between the mean of Core Group

HUXSTER ET AL.

159

1, Science major and Environmental group, and that of Core Group 4, Non-science major and
Non-environmental group.

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

Specific Elements of Respondent Mental Models

Specific statements from the survey were of particular interest in inferring elements of respondents mental models of climate change. The selection of these specific statements was based on
the mental models observed in the responses of students in the pilot study interviews. During
the interview process, students were able to describe correctly some aspects of climate change,
but also showed significant issue confusion, in which they conflated climate change with other
environmental issues such as ozone depletion. A few of the students interviewed expressed that
natural causes play a significant role in recent climate change, while others indicated that they
know little about how climate change occurs (Huxster, 2013). Representative statements from
these lines of reasoning are presented here to test the prevalence of these specific aspects of student
mental models across the student population. Separating and presenting these contradictory sets
of statements also illustrates some of the interesting overlaps in student agreement seen between
scientifically correct and scientifically incorrect statements. This serves to further explain and fill
out the outline of their mental models of climate change.
Statements Matching the Scientific Model
First, responses to survey statements matching different aspects of the scientific model are
examined to gain a clear picture of how respondents understand the causes of climate change
and the actions that can be taken. Next, responses to statements of different models of climate
change, specifically those that confuse climate change with other environmental issues and those
attributing climate change to natural causes, are listed to investigate respondents subscription
to those models. Finally, respondents responses to statements that self-identify confusion or
ignorance about climate change are considered.
Table 3a shows response percentages to survey items matching the scientific model of climate
change.4 The responses to statement 2.10 show that just over half of the respondents understand
the high level of scientific consensus that exists about climate change. Almost 30% of the
respondents are unsure about the scientific agreement, and almost 20% do not agree that scientific
agreement exists. This shows some lack of understanding of the scientific communitys views.
A more positive indicator for respondents understanding of climate change is that 77% of the
respondents agree with statement 2.23, that the burning of fossil fuels contributes to climate
change.
Complete comprehension of the scientific model of climate change also includes understanding
of the mitigation actions that can be taken. Responses to survey statements matching the scientific
model of actions to help prevent or lessen climate change are shown in Table 3b. The majority of
the respondents have heard of energy efficiency measures they can enact in their own homes, and
know that alternative and renewable energy sources are important parts to the solution to climate
change. A smaller majority was able to identify that the reduction or elimination of carbon-based
fuel sources is the best way to mitigate climate change. The difference in responses between
2.30 and 2.43 suggests that although respondents have retained the information that renewable

160

STUDENT UNDERSTANDING OF CLIMATE CHANGE

TABLE 3
Responses to Survey Statements Grouped by Model Element
3a. Responses to Statements Matching the Scientific Model of Climate Change
Item

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

2.10 Scientists are highly certain that humans are definitely

the cause of current, rapid climate change
2.18 Climate change is caused by an increased amount of
gases like carbon dioxide and methane in the atmosphere
2.23 Burning fossil fuels releases carbon dioxide into the
atmosphere, which traps heat and causes climate change

Agree

Neither

Disagree

53.2%

29.0%

17.8%

682

75.1%

21.5%

3.4%

610

77.0%

19.6%

3.3%

566

3b. Responses to Statements Matching the Scientific Model of Mitigation Actions

2.30 The best way to deal with climate change would be to
59.2%
32.0%
8.8%
reduce or eliminate carbon-based fuel sources
91.7%
4.9%
3.4%
2.42 Ive heard that using efficient light bulbs, turning off
electric appliances and insulating my house are all ways I
can reduce my contribution to climate change
84.9%
10.8%
4.2%
2.43 The government could put more resources into
developing alternative and renewable sources of energy in
order to reduce climate change
3c. Responses to Statements Confusing Climate Change with Ozone Depletion
2.11 Climate change is happening because we are depleting
43.0%
26.9%
30.1%
the ozone layer, and that lets in more heat from the sun.
2.14 Gas and chemical emissions from our cars deplete the
66.8%
17.4%
15.7%
ozone and cause climate change.
2.16 Greenhouse gases make the hole in the ozone layer worse
54.8%
23.2%
21.9%
2.36 I can help prevent climate change by not buying or using
50.1%
28.7%
21.1%
aerosol cans
3d. Responses to Statements Confusing Climate Change with Toxic Pollution
2.15 Toxic pollution in the atmosphere makes climate change
65.2%
24.0%
10.7%
worse
2.35 The government can reduce climate change by regulating
65.7%
21.7%
12.7%
toxic pollution
2.39 Cutting down on waste and pollution will help stop the
54.4%
31.8%
13.8%
melting of the ice caps
3e. Responses to Statements attributing Climate Change to Natural Causes
25.4%
47.1%
27.5%
2.19 Climate change involves magnetic field and fluctuations
in the sun, and other things beyond our control, and those
things are bigger factor than greenhouse gas emissions.
2.20 Natural things like volcanoes and water vapor have more
16.3%
41.7%
42.0%
of an impact on climate change than human action.
3f. Responses to Statements Expressing Confusion or Ignorance about Climate Change
45.3%
18.2%
36.5%
2.13 I know that changes in the atmosphere cause climate
change, but I dont know very much about the scientific
processes that make it happen
27.6%
11.6%
60.8%
2.24 No one has ever really explained to me how the
greenhouse effect works, or if they have I cant really
remember

522
494

489

672
648
633
505

641
507
500

594

588

658

562

HUXSTER ET AL.

161

resources are an important solution, fewer respondents understand why these alternatives are
important or why non-renewable energy sources exacerbate the problem.

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

Statements Displaying Issue Confusion

Despite the high number of responses indicating that fossil fuels play a role in climate change,
responses to several statements dealing with the confusion of climate change with other environmental issues reinforce the mental models observed in the pilot study interviews. Some of
the statements dealing with the confusion of climate change and ozone depletion are seen in
Table 3c. The percentages of agreement with these statements indicate significant confusion in
respondents mental models between climate change and ozone depletion. The responses were
relatively split on statement 2.11, with a slight majority agreeing that depletion of the ozone
layer causes climate change. However, more than half of the respondents incorrectly believed
that greenhouse gases contribute to the hole in the ozone layer. Two thirds of the respondents
have incorporated new information about the contribution of car emissions to climate change into
their ozone depletion model of climate change, as can be seen in statement 2.14. The confusion
of climate change and ozone depletion affects not only respondents understanding of causes of
climate change, but also understanding of prevention measures. Half of the respondents mistakenly believe that limiting their use of aerosol cans will help to prevent climate change. The issue
confusion between climate change and ozone depletion as seen in this study is similar to issues
encountered in previous studies of public understanding of climate change (Bostrom et al., 1994;
Kempton et al., 1995). The confusion surprisingly has persisted over 20 years, despite public
coverage and despite these respondents having taken science courses, presumably with lessons
to the contrary, (Reynolds et al., 2010).
Another environmental issue commonly confused with climate change in the pilot study and
in previous literature is that of toxic pollution or waste. In Table 3d, statements illustrating this
confusion are listed. For both statements 2.15 and 2.35, nearly two thirds of the respondents
agreed that toxic pollution is a cause of climate change and that the regulation of these products will help combat climate change. In a statement directly linking toxic waste and pollution
to a commonly understood effect of climate change, the melting of the ice caps, more than
half of the respondents expressed agreement. This mental model of climate change has been
noted, and called the Pollution model, in previous research (Bostrom et al., 1994; Kempton et al., 1995; Reynolds et al., 2010). The issue confusion seen with the ozone depletion
model and the pollution model of climate change in this study may be a case of lumping any
generally good environmental practice into the category of preventing climate change. In
1994, Bostrom et al. found that in the absence of sufficient information, people tend to agree
with all practices that they see as good for the environment to explain how to combat climate
change.
Statements of Climate Skepticism and Self-Reported Knowledge
Climate change skepticism promotes the mental model of natural occurrences, or cycles, as the
primary causes of climate change. Agreement with statements of this model can be seen in Table
3e. For the first of the two statements in this table, the majority of respondents indicated that
they neither agree nor disagree that natural elements beyond human control are bigger factors in

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

162

STUDENT UNDERSTANDING OF CLIMATE CHANGE

climate change than greenhouse gas emissions. In the second statement concerning water vapor
and volcanic eruptions, the response rates seen for neither and for disagree were almost
equal. These results may mean that there is some confusion among the respondents about the role
that natural occurrences play in climate change. For both statements, however, more respondents
disagreed than agreed with the statements that natural occurrences are more important factors in
climate change than human actions.
The final set of survey statements analyzed in this study, shown in Table 3f, are included to
illustrate respondents self-reported understanding of climate change. For statement 2.13, a slight
majority of respondents agreed that they do not feel they understand the scientific processes of
climate change. Larger majorities of respondents, however, disagreed with statements indicating
they had not learned about the greenhouse effect or that they did not know how they personally
affect climate change. These results show that a majority of respondents believe that they know the
correct ways in which to combat climate change. It has been shown, however, that self-reported
informedness does not correlate with objective measures of subject knowledge (Durant & Legge,
2005). Given the high number of respondents holding incorrect models of climate change as
indicated by responses to statements of issue confusion, this may be the case in this study.

DISCUSSION AND CONCLUSIONS

In the overall population at the two universities represented in this study, the respondents show
some basic understanding of the scientific processes of climate change. The mean knowledge
score falls at 17.84 on a scale of 035. In terms of college grading, this is close to 50%
on a grading scale. These scores indicate that the answer to the first research question is that
students knowledge leaves much to be desired. Further research might look at respondent levels
of understanding, learning progressions of university students, and what levels of understanding
are necessary for correct and informed action, as suggested in Mohan, Chen, and Andersons
(2009) study of high school students.
The second research question asks how student knowledge and mental models compare to
the scientific model of climate change. In looking at specific statements made in the pilot study
interviews and the responses to corresponding questions from the survey, further detail in the
knowledge elements of respondents mental models is revealed. Three quarters of the respondents were able to identify the burning of fossil fuels as a source of greenhouse gases and most
respondents knew that energy efficiency measures and renewable resources were potential solutions. However, only half of the respondents understood that there is a high level of scientific
agreement about human impact on the global climate. Also, subtle changes in the wording of
similar questions about fossil fuels (i.e., calling them carbon-based fuels) may have influenced
responses. Overall, these results indicate that gaps exist in the knowledge elements of students
mental models.
In addition to these gaps identified in respondent understanding of the scientific model, the
respondents also show a great deal of confusion between climate change and other environmental
issues. This confusion of issues is seen not only in respondents understanding of the causes
of climate change, but also in their understanding of the solutions and actions to be taken for
climate change. The areas of confusion, specifically in respondents ozone depletion model and
the pollution model, closely match those seen in previous studies of the public, dating back to

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

HUXSTER ET AL.

163

the early 1990s. These findings indicate that misconceptions in respondents mental models have
persisted and these misconceptions need attention to ensure that these young members of society
are prepared to tackle this serious issue effectively. One positive finding is that although the
role of natural causes has led to some confusion in respondent mental models, climate change
skepticism doesnt seem to play nearly as large a role in the undergraduates mental models as has
the persistence of environmental issue confusion. Understanding the specific nature of the gaps
and misconceptions in the knowledge elements of mental models can better inform education,
both formal and informal, on this matter.
The third research question asks about the differences in student mental models between college majors and environmental group participants. Predictably, when looking at different groups
of respondents, those belonging to the related science majors are shown to have a higher level
of knowledge than those students not belonging to science majors. The finding that those respondents belonging to environmental groups score higher on the Knowledge Score scale than
those not belonging to such groups could also have been predicted. What is more surprising,
however, is that environmental group membership is a greater indicator of climate change knowledge (as defined by the Knowledge Score) than enrollment in a science major. This research
does not cover how these respondents reached this higher level of knowledge, or what causes
respondents in environmental groups to have a greater understanding. It is possible that these
findings could be attributed to the education and information respondents receive from environmental groups, but we consider it unlikely that environmental groups are better sources of
information than science classes. Students in both a climate science-related major and an environmental group have the highest mean knowledge scores. Another possible explanation for the
differences in scores could be that respondents greater understanding is a function of a higher
level of interest and concern for climate change, which is manifested in the respondents choice
of major and environmental group participation. These hypothesized relationships could also
explain why environmental group membership is a greater predictor of high Knowledge Score
than enrollment in a science major. Further study of the information covered in specific classes
and disseminated through environmental groups, as well as in students motives in choosing
college majors and belonging to an environmental group, would be needed to test this possible
explanation.
Further research is needed to sort through the possible explanations for the understanding
of science students and environmental group members. More important, the respondents who
likely need the most attention are those not belonging to either of these groups. The non-science,
non-environmental students represent the majority of the respondents (64%). The findings of this
research indicate that these respondents are either not receiving or not retaining the information
necessary to have an adequate level of understanding of the issue. If they are to contribute
meaningfully to climate change solutions as citizens, they will likely need to have a greater
understanding than shown in this study.
ACKNOWLEDGMENTS
We thank Dr. John Madsen for assistance in acquiring partial funding for this research from
the National Science Foundation grant for Research on Learning in Formal and Informal
Settings.

164

STUDENT UNDERSTANDING OF CLIMATE CHANGE

NOTES

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

1. The work reported here draws on only the knowledge portion of the survey from Huxster 2013. Elsewhere
the survey is referred to as the Survey of Student Understanding of Climate Change.
2. For the purposes of reporting the results, statement responses for Agree and Strongly Agree and
Disagree and Strongly Disagree were condensed together in Table 3
3. Survey respondents were informed that the use of the word climate change was used in place of the term
global warming and represented the same phenomenon. In cases in which it was necessary to be more
specific, the phrase current, rapid climate change was used. The survey was framed to respondents as
an exploration of their opinions on environmental issues.
4. In reporting responses to individual survey statements, all students responding to that element are included
(even those who might not have completed the entire survey) to give the most complete analysis possible.
The n values in table 3 reflect the number of students responding to each survey item

REFERENCES
Agar, M. H. (1980). The professional stranger: An informal introduction to ethnography. New York, NY: Academic
Press.
Bernard, R. H. (2002). Research methods in anthropology: Qualitative and quantitative approaches, (3rd ed.). Lanham,
MD: Altamira Press.
Bly, B. M., & Rumelhart, D. E. (Eds.). (1999). Cognitive science. San Diego, CA: Academic Press.
Bostrom, A., Morgan, M. G., Fischoff, B., & Read, D. (1994). What do people know about climate change? 1. Mental
models, 2. Survey studies of educated laypeople. Risk Analysis, 14(6), 959982.
DAndrade, R. (1992). Schemas and motivation. In R. DAndrade & C. Strauss (Eds.), Human motives and cultural
models (pp. 2244). New York, NY: Cambridge University Press.
DAndrade, R. (1995). The development of cognitive anthropology. New York, NY: Cambridge University Press.
Durant, R. F., & Legge Jr., J. S. (2005). Public opinion, risk perceptions, and genetically modified food regulatory policy.
European Union Politics, 6(1), 181200.
Gentner, D., & Stevens, A. L. (2014). Mental models. New York, NY: Psychology Press.
Holland, D., & Quinn, N. (1987). Culture and cognition. In N. Quinn & D. Holland (Eds.), Cultural models in language
and thought (pp. 340). New York, NY: Cambridge University Press.
Huxster, Joanna K. (2013). Student understanding of climate change: Influences of college major and environmental
group membership on undergraduate knowledge and mental models (Doctoral dissertation). Retrieved from ProQuest.
(Publication Number 3594926).
IPCC. (2013). Climate change 2013: The physical science basis summary of policy makers. Contribution of working
group I to the fourth assessment report of the intergovernmental panel on climate change. [Stocker, T., Dahe, Q., &
Gian-Kasper, P. (Coordinating Lead Authors)] Retrieved from www.ipcc.ch
Kempton, W. (1987). Two theories of home heat control. In N. Quinn & D. Holland (Eds.), Cultural models in language
and thought (pp. 222242). New York, NY: Cambridge University Press.
Kempton, W. (1991). Public understanding of global warming. Society and Natural Resources, 4, 331345.
Kempton, W., Boster, J. S., & Hartley, J. A. (1995). Environmental values in American culture. Cambridge, MA: MIT
Press.
Klosterman, M. L., & Sadler, T. D. (2009). Multi-level assessment of scientific content knowledge gains associated with
socioscientific issues-based instruction. International Journal of Science Education, Research Report.
Leiserowitz, A., Smith, N., & Marlon, J. R. (2010). Americans knowledge of climate change. Yale University, New
Haven, CT: Yale Project on Climate Change Communication.
Liu, S. C., & Lin, H. (2015) Exploring undergraduate students mental models of the environment: Are they related to
environmental affect and behavior? Journal of Environmental Education, 46(1), 2340.
Markowitz, E., Hodge, C., & Harp, G. (2014). Connecting on climate: A guide: A guide to effective climate change
communication. Columbia University, New York, NY: Center for Research on Environmental Decisions.

HUXSTER ET AL.

165

Downloaded by [Joanna Huxster] at 15:13 29 June 2015

Mohan, L., Chen, J., & Anderson, W. A. (2009). Developing a multi-year learning progression for carbon cycling in
socio-ecological systems. Journal of Research in Science Teaching, 46, 675698.
Mudditt, A. (1996). Cognitive science: An introduction. Cambridge, MA: Blackwell Publishers.
Oreskes, N. (2004). Beyond the ivory tower: The scientific consensus on climate change. Science, 306(5702), 1686.
Reynolds, T. W., Bostrom, A., Read, D., & Morgan, M. G. (2010). Now what do people know about climate change?
Survey studies of educated laypeople. Risk Analysis, 30(10), 15201538.
Rosenberg, S., Vedlitz, A., Cowman, D. F., & Zahran, S. (2010). Climate change: A profile of U.S. climate perspectives.
Climatic Change, 101, 311329.
Shepardson, D. P., Choi, S., Niyogi, D., & Charusombat, U. (2011). Seventh grade students mental models of the
greenhouse effect. Environmental Education Research, 17(1) 117.
Zia, A., & Todd, A. M. (2010). Evaluating the effects of ideology on public understanding of climate change science:
How to improve communication across ideological divides? Public Understanding of Science, 19(6) 743761.