How Americans View Climate Change:
An Interview with Anthony Leiserowitz of Yale University

Anthony LeiserowitzDr. Anthony Leiserowitz is Director of the Yale Project on Climate Change Communication and a Research Scientist at the School of Forestry and Environmental Studies at Yale University. His body of research has made him a respected expert on public views of global warming. Under Dr. Leiserowitz' leadership, the Yale Project on Climate Change Communication, together with the George Mason University Center for Climate Change Communication, has conducted five national surveys in an ongoing study of American climate change knowledge, attitudes, and behavior. A:I/R interviewed Dr. Leiserowitz in August 2010, and this account refers to data collected in May through June 1, 2010. For all surveys, KnowledgePanel® was the sample source.

Funding for the first three waves came from the Surdna Foundation, the 11th Hour Project, the Pacific Foundation, and the Robert Wood Johnson Foundation. The most recent survey was funded by the National Science Foundation, as part of the Communicating Climate Change Initiative (C3) in collaboration with the Association of Science & Technology Centers and Cornell University.

What was the purpose of your study that yielded the report, "Global Warming's Six Americas"?

We've been studying the American public's response to climate change. What do people understand and misunderstand about the causes, the consequences, and potential solutions? How do they perceive the risks, including the likelihood and severity of climate impacts? What policies do they support or oppose? What actions are they willing to take or not and what barriers do they confront? We investigate four main categories of behavior. Energy use at home and on the road is one; a second is consumer behavior. A third category is our behavior as members of social networks – including families and friendships. What are the social expectations for how people are supposed to act in society? We've seen complete sea changes in the way Americans have changed the unwritten rules of how, for instance, one is supposed to be a good person and not throw trash out the car window. That wasn't always the case. Last is our behavior as citizens. Will Americans support the political leaders who will vote for policies that reduce carbon emissions? Finally, as scientists, we're interested in knowing why. What are the underlying psychological, cultural, political, or structural factors that shape public responses to climate change?

Of course, the American public does not speak with a single voice on this issue. One of the first rules of effective communication is to "know thy audience." There are very different publics on this issue within the United States, and this study tries to identify and profile these different groups.

Why did you decide to work with Knowledge Networks?

There were many reasons, including the advantages of the online platform, but a major one was that Knowledge Networks (KN) introduced me to a new segmentation methodology called latent class analysis. KN has been a real partner in helping us think through new and creative ways to get at scientific questions. In this case, we tried a variety of segmentation solutions and ultimately found that a six-segment solution was by far the best, both statistically and in terms of what we would call "face validity." It was interpretable.

You identified six distinct segments: Alarmed, Concerned, Cautious, Disengaged, Doubtful and Dismissive. What are some of their characteristics?

lightningThe "Alarmed" are convinced that climate change is happening and human caused, and consider it a serious and urgent threat. They're much more likely to be taking some action in their lives and eager to know what else they can do individually or we can do as a country. The Alarmed are the most interested, motivated, and engaged with this particular issue; as of June, they comprised about 13% of the country – a sizable "issue public." But unlike other "issue publics," such as the pro or anti-gun lobbies, or pro-choice or anti-abortion movements, the "alarmed" are relatively unorganized.

The largest segment is the "Concerned" – 28%. These are people who think climate change is happening, human caused and a serious threat; but see it as more distant — that the impacts won't be felt for a generation and are primarily going to affect other places. Then comes a group we call the "Cautious" – about 24%. They're not sure if global warming is happening, human caused, or a serious threat. They don't pay much attention to the issue and haven't really made up their minds. Then comes an interesting group (about 10%) that we call the "Disengaged," who have heard the term "global warming" but literally know almost nothing about it. For this particular group, the primary barrier is simply basic awareness – quite different than other groups.

For the last two groups – the "Doubtful" and the "Dismissive," it's an entirely different issue. The "Doubtful" – around 12% – aren't convinced that climate change is happening, but if it is, they feel that it's just part of the natural process of nature and not a threat, so they don't worry about it. Last is the group that we call the "Dismissive" (also about 12%). They are convinced that climate change is either not happening or that any changes are not human caused; in fact, a sizeable proportion goes even further and says it's a conspiracy – scientists making up data or a plot by socialists or others. This group often views the potential policy response to climate change as a threat to individual freedom – such as more government control of markets and regulations.

What are some advantages of using the KnowledgePanel online approach as compared to other modes?

Because we are trying to understand what's happening in an entire country, scientific rigor and methodological quality are of utmost importance. That is a primary reason why we started working with Knowledge Networks years ago, and why we continue. The sample is of the highest quality and comparable to the best telephone surveys. A colleague, Jon Krosnick, did a nice study on this a year or so ago, where he compared the quality of KnowledgePanel with other modes and found exactly that.

Second is the platform. Online offers so many advantages over telephone surveys, especially in today's environment when telephone response rates are dropping. Potential respondents are increasingly moving away from landlines to cell phones, so you must invest ever more, now, to get a response rate greater than 30% using telephone. Having a panel of pre-recruited people, who already have agreed to be part of a survey system, has a huge advantage; we get a participation rate of over 70% with KnowledgePanel on a regular basis. Plus we don't have to ask basic demographic questions about race or marital status. Were I surveying by telephone, I'd have to devote a substantial proportion of my questionnaire to such questions, but not with Knowledge Networks – which means that I'm far more efficient in using my limited survey response time.

The Web interface also enables us to ask more questions. In a phone interview, participants are often asked a series of similar questions with slight variations; online, many questions can be asked using an efficient matrix format. We can also imbed pictures or graphics, and even video and audio. Plus there is cleanliness of the data. Collecting it digitally removes a potential source of error – that of an interviewer entering responses into a CATI-based system. And because people are interacting with the computer instead of a human being, I think that in some cases, respondents are more willing to say what they really think when the issue is controversial, or there's some sort of social implication – especially if you're asking questions that perhaps have racial or gender overtones. With a computer as the "interviewer," everything is standard across all respondents.

You add up all of these things, and I was a very early adopter of the Knowledge Networks approach, as soon as I was convinced it had the scientific integrity on which we could rely.

How important do you feel polling accuracy is to policy makers?

Unfortunately, I think that too many policy makers – not just elected officials or government policy makers, but business leaders – do not understand the importance of polling accuracy. As a result, they're willing to use lesser quality companies, survey methods, and questionnaires that are providing them unsound data on which they're basing huge decisions. These include where to invest billions of dollars and new programs or policy choices, often at the national or even international level. If you're not paying attention to data quality, you can be led astray, which is a real danger. I wish it were more important to policy makers, frankly.

Had you not conducted your surveys on KnowledgePanel, do you feel that some segments would have been overlooked or under-represented?

If we'd done our study using a scientific quality telephone survey, we would get pretty much the same answer. However, it probably would have made a big difference had we used one of the hundreds of online survey companies that rely entirely on opt-in voluntary panels. A large percentage of the American public does not have a computer or Internet access, so the opt-in approach immediately cuts them out of the sample. And when you're using an opt-in panel, you're typically reaching only people who basically live online, who are aware of the opportunity to join, including the small minority of those people who are willing to become professional survey takers. You can build a big panel of 400,000 people using that approach, and weight the hell out of the data, but from a scientific standpoint, it's much less likely that you're going to have a truly representative view of the country.

You saw a 10% point increase - from 24% in January 2010 to 34% in June 2010 – in the idea that "developing sources of clean energy should be a very high priority for the President and for congress." What factors do you think influenced this shift?

We certainly can speculate. First, there was the Gulf oil spill, which happened during this time period. Every day, people were confronted those graphic images of the underwater "spill cam," along with the oil washing ashore and coating – a visual reminder of the downside of our reliance on fossil fuels. It's possible that these images increased support for developing sources of clean energy. But clean, renewable energy development has been highly supported for years – across the board by Democrats, Republicans, liberals and conservatives. The key question is one of priority – not only saying that we support it, but that it should be a higher priority.

In addition, the state of the economy may have played into it. I think that people are increasingly aware that clean energy is likely to become the largest single market of this century; some estimate that it will dwarf the size of information technology. People around the world want clean energy, not just because of climate change but all of the other negative consequences of fossil fuels – air pollution; respiratory diseases; the environmental effects of things like mountain top removal; mining; drilling; oil spills; reliance on foreign sources of energy; unstable prices; national security concerns; etc. There is widespread support around the world for this transition, and climate change just adds yet another driving motivation. Whoever wins the race to develop the new clean energy technologies, products, and services, and sell them to the rest of the world, is set up to dominate the 21st century.

You're quoted as saying, "If it's not in the media, it's out of sight and out of mind." Whose job is to educate the media about global warming?

First, it's the media's job to educate themselves about this complex issue. But it's also the job of climate scientists to work with the media, to help them understand these issues and to figure out ways to simply and accurately convey the science. The news media is also in the midst of a complete restructuring and we don't know what it will look like in five years. Many science and environmental reporters, for example, have been downsized, and the business model of the news has changed fundamentally. And while the Internet has played an enormous role in making information available, the average person who isn't self interested is now much less likely to at least encounter scientific information – as compared to when it came through a newspaper.

You mentioned the confusion between weather and climate. This is a consistent problem when it comes to understanding climate change as well as its coverage. Last winter, climate change Dismissives were saying, "record snow storms – there's your proof that it's not happening," which is ridiculous from a climate science perspective.

You can't take any one individual weather event and say, "that was caused by climate change" or alternatively "that's proof it's not happening". Climate change is something that you can only observe through multiple flips of the coin or rolls of the dice over time. If you flip a coin a 100 times, and it comes up heads 80 times, you know something is weird about that coin. Climate change essentially loads the dice over time, changing the odds of particular extreme weather events – for example more frequent or intense heat waves than in the past. There will still be winters and record cold snaps in the future, just warmer and fewer of them than now, respectively.

Mistakes in the 2007 IPCC report lessened confidence in the science supporting climate change. Was any one of the Six Americas impacted more than another?

The other event, which I think had more of an impact than the mistakes in the IPCC report, was "climategate" – the unauthorized release of emails among scientists at the University of East Anglia, in the U.K. Particular phrases were taken out of context and used by some as "proof" that climate change is all a big hoax. We've written a working paper on this, which hopefully will be published soon, that investigated the impact of this event using KnowledgePanel. We found that "climategate" had a significant influence – fewer Americans believed that global warming is happening and human caused and some lost trust in scientists – around a nine-percentage point drop in trust. That's significant and substantial, however it's important to put this drop into context. Even with that drop, scientists are overwhelmingly the most trusted source of information about climate change.

The other question to ask is "who lost trust?" This gets us back to our Six Americas. By far, the groups who lost the most trust in scientists were the "Doubtful" and "Dismissive." These groups are much more likely to be politically conservative, or to have an individualistic worldview or value system most predisposed to being skeptical about climate science. These scandals were generated, seized upon, and used as ammunition by Dismissives in their own communication networks. This is a nice example of what psychologists call motivated reasoning – the tendency to interpret new facts and information in line with one's current beliefs. However, it wasn't just the people that were the most "Dismissive" who were affected. Some people on the opposite end of the spectrum – among the "Alarmed" – actually became more convinced that climate change is happening and to have more trust in scientists as a result of the story. They saw the exact same "Rorschach test" and arrived at the exact opposite interpretation. We think this is a nice of example of how facts do not exist in a vacuum. They are always interpreted in particular ways, by particular people.

Learn more by visiting the "Six Americas" video link:

For more information, contact:

Joe Garrett
703 830-0613

Download PDF

Bookmark and Share