Climate Change and the Community

Sarah Gray Erickson

Abstract: Climate change is arguably the most important and time sensitive problem in existence today. Immediate actions must be taken to curtail its devastating effects. While climate scientists are 98% in agreement that humans are causing climate change, the public is nowhere near this consensus. Change will not be possible until the public is closer to scientific agreement. The Washington and Lee University community provides an interesting subpopulation for exploring beliefs in climate change. This paper uses survey data to examine how the Washington and Lee community compares to national averages, how their beliefs have changed over time, and how political affiliation, ideology, media use, attention and education all influence belief. Additionally, it analyzes how students who completed a course on climate science changed their beliefs after the course, as well as what they learned about climate science. There are several policy implications from these findings that can help guide future decisions within the university as well as nationally.  


According to a January 2015 Pew Research Center study, 87% of scientists associated with the American Association for the Advancement of Science (AAAS) agree that climate change is caused by human activity (Funk et al. 2015:47). Further, there is a 98% consensus on anthropogenic climate change among climate scientists who actively publish in their field (Anderegg et al. 2010:12107). The Intergovernmental Panel on Climate Change (IPCC) predicts between a 2°-11° Celsius increase across the entire globe by 2100, with the northern hemisphere experiencing the most dramatic effects (Field et al 2014:11). Yet, even with such clear scientific consensus, society is often disagreeing with science. In 2014, only 40% of Americans believe humans are causing climate change, a 47 point gap from AAAS scientist consensus (Dimock et al. 2014:69). From Fall 2012 to Fall 2013, the percentage of Americans believing in anthropogenic climate change dropped 7% from 54% to 47% (Leiserowitz et al, 2014:7). Increasing the divide is the fact that only 12% of Americans correctly estimate that over 90% of scientists agree on anthropogenic climate change (Leiserowitz et al, 2014:8).A 2013 University of Michigan study found that only 40% of Americans believe in human-caused climate change (Lachapelle et al. 2014:6). The average of these three studies reveals that 62% of Americans believe climate change is happening, and only 44% believe it is human-caused.

There are several proposed causes for this disconnect between science and society. The first is that political affiliation influences whether or not an individual believes in climate change. This idea has been explored by many researchers. A Yale report entitled Politics and Global Warming found that 93% of liberal democrats think climate change is happening, only 28% of conservative republicans think similarly (Leiserowitz et al, 2014:4) Conservative Republicans are not only likely to not believe in climate change, but they actively make political decisions against candidates and legislation that attempt to improve the environment (Leiserowitz et al, 2014:5). Even towards the middle of the spectrum, there is still a divide between Democrat and Republican views on climate change and party association is a likely explanatory factor for belief.

Partisan media use is another proposed cause of belief in climate change. Cable news networks are often cited as examples. Content analysis of Fox News, CNN, and MSNBC found that Fox has a less urgent attitude towards climate change (Feldman et al. 2011:1). They are also more likely to interview more doubters of climate change than believers of climate change (Feldman et al. 2011:1). Individuals who view Fox are less likely to believe in climate change than individuals who view CNN and MSNBC (Feldman et al. 2011:1). Moreover, Republicans are strongly linked to their news source and are often predisposed to climate change skepticism while Democrats opinions do not vary with which cable news outlet they watch (Feldman et al. 2011:1). Building on this trend is the issue of using media to confirm pre-existing opinions about issues. Those who do not believe in climate change insulate themselves from adverse opinions by selecting news sources which they know share their same belief (Feldman et al. 2014:1). Whether or not these actions are intentional, they are a possible cause of the divide between scientific and society consensus.   

Another issue to consider is the argument that those with the most vocal opinions about climate change, especially in the media, are those with relatively low understanding of climate science. The proportion of scientists who disagree about climate change is extremely small compared to the 98% consensus. However, these scientists tend to be more outspoken and unwavering in their opinions (Oreskes and Conway, 2010:6).The motivations behind these fact-fighting, industry funded scientists are to confuse society about major scientific issues in order to further their agendas, and they have always been proven to be on the wrong side of scientific consensus (Oreskes and Conway, 2010:7).Additionally, climate skeptics have accused climate scientists of alarmism (i.e. overstating the effect of humans on the climate system), but it has been proven that the published articles of climate scientists actually understate the effects of humans on climate change (Brysse et al. 2012:1).  Climate scientists do so to avoid being attacked by skeptics, often through partisan media outlets.

Understanding the gap between scientific consensus and societal beliefs of climate change has important policy implications. Progress on climate change mitigation is not possible unless society believes in its importance. Issues that are considered “controversial” are rarely considered “urgent”. Passing legislation is much more inefficient when the issue is controversial. To make progress against climate change in the most efficient way possible, privately or publically, society must have a consensus on climate change. Although scientists already agree, closing the gap in public opinion is crucial.

This paper aims to investigate where a highly educated subset of Americans in an elite academic setting fall on the spectrum of climate change belief. Washington and Lee University (W&L) in Lexington, Virginia is a top-ranked liberal arts school drawing students from all over the country and world. Because of the W&L community’s intelligence, it is expected that they are able to understand climate science enough to believe in anthropogenic climate change. However, there is also a change that W&L’s large Republican constituency could skew this if they are using Republican media sources for news on climate science. A survey of the Washington and Lee community in 2011, 2013 and 2014 attempts to compare W&L’s belief in climate change to both national averages and scientific consensus. Further, it examines possible factors driving these beliefs such as political affiliation, ideology, media use, and scientific understanding. The final aim of this paper is to better understand this demographic to learn what can be done to bring them together with scientific consensus on climate change.


The survey was originally written by Professor Lisa Greer in 2011. It originally contained 23 questions. The survey was written to closely resemble a 2008 survey used in a study by Barry Rabe and Christopher Borick of the University of Virginia’s Miller Center of Public Affairs. The questions were worded carefully to avoid biases. The survey was modified in 2013 add 6 more questions but remained largely the same in content and wording. It was distributed in 2011, 2013 and 2014 using the same methods. First, it was given to students in Geology 141- Global Climate Change during the first week of class. There were 46, 55, and 63 students in years 2011, 2013, and 2014, respectively. Midway through the semester, the students were all trained with the Institutional Review Board and certified to conduct the survey. They were instructed to distribute and collect the survey to four W&L students and one member of the W&L community who was not a student. This resulted in 226, 289, and 301 surveys collected in 2011, 2013 and 2014 respectively. These surveys were given back to Professor Greer and stored for future tabulation. Finally, the students took the same survey again during the last week of the semester.

After each round of surveys was collected, the results were coded into number form. This means converting any answer choices that were given in “word” form and assigning each answer a number (i.e. no answer=0, yes=1, no=2, not sure=3). Four questions asked the individual to describe their understanding of certain aspects of climate science. These four questions were coded using the scale 0=no answer, 1=little understanding, 2=some understanding, 3=excellent understanding. They were read and coded by the same person to maintain consistency. After all the surveys were properly coded, they were input into Microsoft Excel. Each different year or surveys were kept in different files and a new tab within each file was kept for pre-course, post-course and community data.

Once the three years of data were collected and inputted, the results were analyzed. Most of this analysis took place in Microsoft Excel. The 2014 file contained tabs for each of the questions asked in the study. The 2011 and 2013 files contained tabs for questions (we wished) to be compared over time. The formulas were kept consistent throughout the process. Lastly, Excel was used to generate graphic representations of the data, including bar and pie charts.

To prove statistical significance of the data, Professor Robert Humston assisted in creating a combination of chi-squared tests and confidence intervals for proportions of data. A 95% confidence level was used, meaning that all p-values below 0.05 are considered statistically significant. Statistical analysis was run on the most important questions, but not every result to minimize the risk of a Type-1 error. The statistical results were then interpreted and included in the findings.


Because of the breadth of data, analysis lead to a large amount of results. Highlights are listed below.

1. Washington and Lee Community versus National Data Sets

Using the 2014 survey data, W&L was compared to three other national data sets with similar survey questions. The three data sets are Pew Research Center’s 2014 Report “Beyond Red vs. Blue: The Political Typology”, (“Pew”) (Dimock et al, 2014), U. Michigan’s CLOSEUP Study from 2014 (“CLOSEUP”) (Lachapelle et al, 2014), and Yale Project on Climate Change Communications 2014 “Climate Change in the American Mind” report (“Yale”) (Leiserowitz et al, 2014).

Figure 1

Figure 2

It is important to note that the W&L surveys were given choices that included “Yes”, “Somewhat Yes, “Somewhat No”, “No”, and “Unsure”. For these results, “Yes” and “Somewhat Yes” were all labeled “Yes” and “No” and “Somewhat No” are all labeled “No”. The data that did not exactly fit into the categories of the W&L survey was reclassified appropriately. Also, not every variable was included in each study so each point of comparison is not entirely complete. Because of the different phrasing and answer choices for each different set, statistical analysis was not run on these results.

2. Changes in the Washington and Lee Community Over Time

Respondent’s beliefs in climate change were determined by the question “There is not enough scientific evidence to prove that the earth is getting warmer.”

Figure 3

 3. Selected Variables for 2014 Data

Several variables were analyzed to look for results, focusing on the areas of political affiliation, political ideology, media use, attention to the issue and knowledge about climate science.

Political Affiliation:

Figure 4

Each affiliation, Republicans, Democrats, Independents, and Others, had a statistically significant distribution of proportions, with p-values of 5.74E-10, 5.14684E-20, 8.25163E-09 and 1.82027E-05, respectively, at a 95% threshold. In terms of confidence intervals, Republicans versus Democrats had the strongest statistical differences, while Independents and Others tended to overlap both parties.

Political Ideology:

Figure 5

Media Use:

Figure 6

Figure 7

Statistical analysis was conducted on Fox, New York Times, and NBC. Fox and NBC proportions were found to be statistically significant with a Chi-Squared test with 95% confidence, with p-values of 3.3156E-05 and 0.005516564, respectively, while NYT was not significant, with a p-value of 0.291736393.

Attention Paid:

Figure 8

Climate Science Knowledge:

Figure 9

4. Geology 141 Students Pre and Post Course

Students in the course were given the same survey in the first and last week of the semester. Pre and post course data was analyzed to understand how taking the course influenced students.

Figure 10

These results proved to be statistically significant under a Chi-Squared, with a p-value of 2.43893E-08. Confidence interval analysis shows statistically significant proportions for all categories except for strongly agree. Also included in the Appendix is belief in anthropogenic cause of climate change pre and post course, and comparison of the post-course data to the community data.

Figure 11

5. Potential Policy Implications

Belief in climate change versus belief in immediate government and personal action was analyzed for 2014 community data.

Figure 12

Figure 13


The results of the survey data analysis offer important insights into the Washington and Lee community’s views on climate change.

In general, Washington and Lee believes in climate change slightly more than national data sets. In 2014, 70% of the community said either “strongly disagree” or “somewhat disagree” when responding to the question “There is not enough scientific evidence to prove that the earth is getting warmer.” Comparatively, 61% of the Pew sample, 61% of the CLOSEUP sample, and 64% of the Yale sample answered similarly, as seen in Figure 1. However, there is an interesting issue with this figure. The surveys given to the W&L sample includes the options “strongly disagree” and “somewhat disagree” while the national surveys were only given a decisive option. The “somewhat disagree” answers could also be considered unsure. If the “strongly disagree” answers were considered to be the only decisive responses, then only 32% of the W&L community believes in climate change.

In 2014, 61% of the W&L community believed that humans were causing climate change, compared to 40%, 40% and 52% of the Pew, CLOSEUP and Yale studies, respectively (Figure 2). The same survey answer issue remains, however, and only 25% of the sample gave answers that they were certain that humans were causing climate change. While it is positive that it seems W&L is performing above national averages, the W&L results are still 37 points below the consensus of climate scientists, and 26 points below the AAAS consensus in terms of the anthropogenic cause of climate change. The most recent incoming class at W&L posted a 2104 average SAT score and a 32 average ACT score (“First Year Class Profile”). It is ranked as the 14th best liberal arts school in the nation (“Washington and Lee”). Clearly, the W&L community outperforms national data sets in terms of intelligence, education and knowledge. The proportion of those believing in the anthropogenic cause of climate change should me much closer to scientific consensus than it is.

Republicans at Washington and Lee are slightly more likely to believe in climate change than the Pew or CLOSEUP Republicans. In 2014, 58% of Republicans believed in climate change, while 38% of Pew and 52% of CLOSEUP Republicans responded similarly. About 45% of Republicans believed in anthropogenic cause, while only 19% of Pew Republicans believed the same. On the other end of the spectrum, 92% of W&L Democrats believe in climate change, compared to 84% and 71% of Pew and CLOSEUP Democrats. Moreover, 85% of W&L Democrats believe in the anthropogenic cause of climate change while only 53% of the Pew Democrats agree. There is an interesting almost 30-point gap in belief of climate change and belief in anthropogenic climate change in the Pew data that shrinks to only 8 points in the W&L data.

Those who classify themselves as independents varies by data set. The W&L data set gives an option for “Independent” and an option for “Other”, while the other surveys just gave an “Independent” option if they were neither Democrat nor Republican. The Independents at W&L fell in the middle of the other parties, at 72% agreement on climate change, compared to 61% and 59% of Pew and CLOSEUP Independents. In the middle of the parties again, 67% of independents believed in the anthropogenic cause. This data point was not available in the Pew data set. It is important to note that the political party data still considers “Somewhat” as affirmative answer, while it could really be considered unsure, thus greatly decreasing the percentage of believers in both climate change and its anthropogenic cause.

Interestingly, the W&L community contains more Republicans and Independents than Democrats. The Pew and Yale data sets have more Democrats than Republicans and far fewer Independents. This suggests that members of the W&L community are more likely to classify themselves as Independents than the national average.

Overall, in 2014, the Washington and Lee community believes in climate change and its anthropogenic cause more than the national average. This holds strong through each political affiliation examined. However, even at the highest examination point of the anthropogenic cause of climate change – 84% of Democrats – it is still not at scientific consensus.

 In 2011, 2013 and 2014, the Washington and Lee community remained consistent in their beliefs in climate change, with the 2013 data set having the highest percentages of believers (Figure 3). The Republicans and Democrats proportions stayed relatively constant as well, and the biggest movement happened within the Independent group, shown in Figures 20-22. In this time period, there were a few milestones that could have influenced the community’s opinions. In 2009, the controversial “Climate-gate” scandal released emails from climate scientists that contributed to public skepticism about the merits of climate change (Weart). In 2012, controversial studies were released attributing extreme weather events like heat waves, droughts, precipitation extremes, and floods to global warming (Weart). A year later, a study was released questioning a pause in warming of atmospheric temperatures since 1998, but was later explained (Weart). Also in 2013, global mean temperatures and atmospheric carbon dioxide reached new records (Weart). Although these events are not directly reflected in any part of the surveys, it is possible that they had an effect on the community.

There are several variables to explore that are potential reasons for the results of the W&L community data. The first is political affiliation. As expected, less Republicans believe in climate change than democrats. Looking deeper into this statistic, almost four times as many Democrats are entirely certain (“yes”) that climate change is happening than Republicans (Figure 4). This is statistically significant. Moving into the respondents that answered “somewhat yes”, there are almost twice as many Republicans. However, here, there is somewhat of an overlap in the confidence interval. Independents lay in the middle of the Republicans and Democrats in all responses except for “somewhat yes”, where they are the largest group (Figure 4). This, combined with their relatively high proportion of “somewhat no” answers could mean that Independents are more unsure of their views on climate change than the other two parties. These findings are consistent with the results reported in the Yale study Politics and Global Warming (Leiserowitz et al, 2014:4). This could mean that individuals whom associate with political parties are quicker to choose a stance on climate change based on where their party generally stands.

Political ideology provides another interesting variable. The survey asks users to rank their political ideology using the options “Very Active”, “Active”, “Intermediate”, “Laid Back”, or “Laissez-Faire”. Every respondent who claimed climate change was “not a problem” had an intermediate, laid back, or laissez-faire ideology (Figure 5). Inversely, very few respondents who said climate change was a “very serious” problem claimed these ideologies (Figure 5). When political ideology was crossed with belief in personal and governmental actions against climate change, the results were consistent with expectations. However, in every category except for “Very active”, more respondents believed personal action was more necessary than government action. 

The way climate change is portrayed on certain media outlets has been increasingly blamed for intensifying the climate change beliefs of those who select that outlet (Feldman et al. 2011:1). Individuals with certain political and issue-specific beliefs select media sources that they know will present news that will not challenge their beliefs. This effect is present within the Washington and Lee community. When asked which news sources they use to get news about climate change, over 10 times as many Republicans as Democrats listed Fox (Figure 6). The New York Times and NBC networks had roughly twice as many Democrats than Republicans, while most other major networks were more evenly split (Figure 6). Respondents were also asked if they believed the media was overstating or understating climate change. Those who listed Fox as a news source were much more likely to believe that the media was overstating climate change, while the New York Times and NBC users were more likely to believe that the media was understating climate change (Figure 7). Under a Chi-Squared test, both the Fox and NBC findings were statistically significant, while the NYT findings were not. The same was found when analyzing confidence intervals. Partisan media sources certainly have an effect on how the Washington and Lee community sees climate change.

Another interesting point for examination is how much attention the individual has paid to climate change versus their belief in climate change. Many more respondents claimed to have paid low attention to the issue (Figure 8). Those who paid low attention had a higher proportion of those who do not believe in climate change, versus those who paid high attention. This suggests that the more a person learns about the issue, the more likely they are to believe in it.

The last variable examined in the community data set was the understanding of climate science. This was measured from four questions: 1) can you explain the carbon cycle 2) how many greenhouses gases can you name 3) what is solar variability and 4) by what mechanisms and on what timescales does solar variability affect climate? The answers were scored for accuracy using a consistent scale and then averaged into one final score with respondents falling into one of four categories. The proportions of respondents who did not believe in climate change declined as their understanding increased (Figure 9). Only 8% of respondents who had a decent understanding of climate science believed it was not a problem, while the majority of those with a decent understanding believed it was a very serious problem (Figure 9). However, an overwhelming majority of the population has very little understanding of climate science. This leads to an important potential policy implication within the university to be addressed in the policy section.

Because of the pre- and post-course data, there are important findings suggesting the improvement of belief in climate change with improved education in climate science. Figure 10 shows the proportions of the class and their beliefs in climate change pre and post course. While most of the course believed in climate change before the course, they were “somewhat” believers and not “strong” believers. After the course, most everyone in the class agreed strongly that climate change was a problem. These results were statistically significant. Similar results are found in proportions of believers in anthropogenic climate change. A statistically significant increase in the proportion of those who “strongly” believe in the anthropogenic cause is seen after the course. Compared to the rest of the community, the post-course population believes in climate change much more, especially in those who believe it is a very serious problem. Every category in this finding is statistically significant. While it seems like students constantly hearing about the perils of climate change for a semester could cause these results, they are supplemented by a clear increase in scientific understanding in all categories. Again, these results provide an opportunity for a potential policy implication.

The gap between scientific and public consensus on climate change must be closed to create any progress in combatting arguably the most important issue of this generation. If climate change is still considered a “controversial” issue, there will be significantly less action taken against it. The time constraints created by the nature of climate change itself add pressure to policy. Results from the surveys indicate that as people’s belief in climate change increases, so does their belief in the necessity of government and personal action to combat it, as seen in Figure 12-13.

As mentioned, belief in climate change improves with both attention paid and education on climate science. The belief in personal and government action improves with education as well. Improved education about climate change is clearly an effective policy for improving a university community’s opinions on climate change. Washington and Lee University, an institution with the motto Non incautious futuri, meaning “not unmindful of the future”, should seriously consider making education on the importance of climate change a more integral part of undergraduate education. The group of students at Washington and Lee has the potential to make a large impact on the future and should leave the university understanding the issues of climate change and their pertinence on an individual and global level.

Outside of the University, these findings have implications for both politics and media use. Clearly, climate change is a partisan issue. There are deep divides on the political spectrum on belief in climate change. For an issue with 98% scientific consensus, this is unacceptable. The effects of climate change are felt on all people, regardless of their political affiliation. The media must stop acting as an echo chamber for climate change skeptics. The results show that media has an impact on individual’s views on climate change and for an issue that should not be partisan in the first place; partisan reporting is only harming the situation.


Climate change has one of the biggest gaps between scientific and public consensus of any issue. While the Washington and Lee community has a higher portion of believers than national data, they are still far from reaching scientific consensus levels. Moreover, much of the Washington and Lee community states that they are “somewhat” believers in climate change, which could move much of the data into the unsure area. Although there have been several new findings in the area of climate science, the community’s opinions have remained generally unchanged since 2011. Political affiliation, political ideology, media use, attention, and education all affect the community’s views on climate change. It is extremely important that the gap between scientific consensus and public belief is closed, as the future depends on it.


I would like to thank Professor Lisa Greer, Professor Jim Kahn, and Professor Robert Humston for their help in completing this project. Professor Greer provided the original idea and data for the project, as well as guidance and insight throughout the entire project. Professor Kahn was in charge of the capstone class and helped oversee the entire process and offered feedback after each step. Professor Humston was extremely helpful in teaching statistical analysis and formulating an analysis plan for each part of the data.


Thanks for reading. Check out New Zealand, New Furniture 



Anderegg, William R. L., James W. Prall, Jacob Harold, and Stephen H. Schneider. "Expert Credibility In Climate Change." Proceedings of the National Academy of Sciences 107.27 (2010): 12107-2109. Print.

Brysse, Keynyn, Naomi Oreskes, Jessica O’Reilly, and Michael Oppenheimer. "Climate Change Prediction: Erring on the Side of Least Drama?" Global Environmental Change (2012): 327-37. Print.

Dimock, Michael, Jocelyn Kiley, Scott Keeter, Carroll Doherty, and Alec Tyson. "Beyond Red vs. Blue: The Political Typology." Pew Research Center (2014): 1-176. Print.

Feldman, Lauren, Edward Maibach, Connie Roser-Renouf, and Anthony Leiserowitz. "Climate on Cable: The Nature and Impact of Global Warming Coverage on Fox News, CNN and MSNBC." The International Journal of Press/Politics (2011): 1-29. Print.

Feldman, Lauren, Teresa Myers, Jay Hmielowski, and Anthony Leiserowitz. "The Mutual Reinforcement of Media Selectivity and Effects: Testing the Reinforing Spirals Framework in the Context of Global Warming." Journal of Communication (2014): 1-22. Print.

Field, Christopher, Vicente Barros, Michael Mastrandrea, and Katherine Mach. "Climate Change 2014 Impacts, Adaption and Vulnerability - Summary for Policymakers." Intergovernmental Panel on Climate Change (2014): 1-32. Print.

"First Year Class Profile (Class of 2018)." <i>Washington and Lee University</i>. Web. 4 Mar. 2015. &lt;;.

Funk, Cary, Lee Rainie, Aaron Smith, Kenneth Olmstead, Maeve Duggan, and Dana Page. "Public and Scientists' Views on Science and Society." (2015): 47. Pew Internet. Pew Research Center. Web. 15 Feb. 2015. <>.

Lachapelle, Erick, Christopher Borick, and Barry Rabe. "Public Opinion on Climate Change and Support for." Issues in Energy and Environemntal Policy 11 (2014): 1-21. Print.

Leiserowitz, A., Maibach, E., Roser-Renouf, C., Feinberg, G., & Rosenthal, S. (2014) Climate change in the American mind: April, 2014. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change Communication.

Leiserowitz, Anthony, Edward Maibach, Connie Roser-Renouf, Geoff Feinberg, and Seth Rosenthal. "Politics and Global Warming, Spring 2014." Yale Project on Climate Change Communication (2014): 2-98. Print.

Oreskes, Naomi, and Erik M. Conway. Merchants of Doubt: How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warming. New York: Bloomsbury, 2010. Print.

"Washington and Lee University." <i>US News and World Report</i>. Web. 4 Mar. 2015. &lt;;.

Weart, Spencer. "Global Warming Timeline." <i>American Institute of Physics</i>. Web. 4 Mar. 2015. &lt;;.