Conflicting Polls Show an Uncertain Public on ANWR

Three recent polls on the public's reaction to possible oil drilling in Alaska's Arctic National Wildlife Refuge (ANWR) found conflicting results. One poll found clear majority opposition, another clear majority support, and the third modest majority support. Some political partisans on both sides of the issue have attempted to point to one or another poll as the "true" measure of what the public wants. However, as I have argued in previous columns, the truth about public opinion is often elusive. While all three polls provide some insights into what the public is thinking, no one poll definitively reflects public opinion on the matter.

A Zogby telephone poll, conducted Dec. 13-15, 2004, of 1,203 likely voters found majority opposition by a 17-point margin.

Do you think oil companies should be allowed to drill for oil in America's Arctic National Wildlife Refuge?

A poll of 800 registered voters conducted and released in January by the Luntz group, found majority support for pursuing oil and natural gas in ANWR, though the margin depended on whether the interviewer said "ANWR" or said "the Arctic National Wildlife Refuge." Half the sample was read one version, the other half the second version. If the whole name was read, then the margin in support is 12 points, 53% to 41%. If "ANWR" was read, then the margin is 17 points, 51% to 34%.

And right now, if you had to choose do you strongly support, somewhat support, somewhat oppose, or strongly oppose the exploration, development and production of oil and natural gas from [The Arctic National Wildlife Refuge/ANWR] as one of the ways to increase our nation's energy supply?

Finally, Harris Interactive reported the results of their online poll of 2,209 adults, conducted Jan. 11-16, 2005, which also found majority support -- but by a six-point margin -- 53% to 47%.

How much would you support or oppose the following items that might be on the agenda of the new Congress? -- Energy reform to allow companies the ability to drill for oil in certain areas such as the Alaska National Wildlife Refuge (ANWR) to decrease our reliance on foreign oil?

According to these results, the public could support oil drilling in ANWR by up to a 17-point margin, or it could oppose such drilling by an identical 17-point margin. Can opinion be that vague?

The answer is almost certainly yes, given the low level of knowledge that most people have about this subject. In its survey, the Luntz group asked respondents what ANWR stood for and where it was located. Eighty-seven percent could not place the location in Alaska, and the same percentage could not identify any word of the acronym ANWR. In addition, only 8% said they knew either "a lot" (3%) or "a good amount" (5%) about the area.

Despite this lack of knowledge, the number of people who were unwilling to express an opinion on the matter was 7% in the Zogby poll, 7% in the half sample of the Luntz poll that referred to the "Arctic National Wildlife Refuge," 15% in the other half sample in which the acronym "ANWR" was mentioned, and 0% in the Harris poll.

If the vast majority of people don't know anything about ANWR, then on what basis are they expressing an opinion about oil drilling there? Answer: Mostly, the questionnaire itself. Respondents usually play the "interviewing game" -- if we ask people questions and don't offer them an explicit option of "don't know," most will go ahead and offer an opinion. What little they know often comes from the questionnaire itself.

Questionnaire Context

The Harris question, for example, frames oil drilling in ANWR as "reform," implying that to allow oil drilling would be better than the current situation. The question also gives a reason for oil drilling in addition to reform -- "to decrease our reliance on foreign oil." It offers no reason why such drilling might be objectionable. Thus, in this framework, the public shows a narrow (six-point) majority in favor of drilling.

The Zogby question on oil drilling is fairly neutral, though it does not offer a "no" option. It just asks if drilling should be allowed, leaving the "no" response implicit rather than explicit. In a previous survey, conducted May 10-13, 2004, Zogby asked a more balanced question:

Do you think Congress should or should not open up the Arctic National Wildlife Refuge in Alaska for oil drilling?

In this case, Zogby explicitly included the "should not" option, which should have increased the percentage of respondents choosing it. Instead, the margin in opposition is just eight points, compared with the 17-point margin of opposition in the other survey. How could that happen?

As it turns out, the oil drilling question in the May 2004 survey was the first question asked about ANWR (or any related subject), which means it probably elicited the most "pure" response from the public. The question itself suggested no reason for or against such drilling, and it was not asked in the context of energy or the environment or a related issue. In the more recent survey, in which the question was slightly "unbalanced", the basic oil drilling question followed a question that asked the best way to reduce U.S. dependence of foreign oil:

Which of the following options do you think is the best way to reduce U.S. dependence on foreign oil?

This question presents two ways for reducing dependence on oil that essentially rely on conservation, and one option that talks about more oil drilling. Also, the order of presentation favors the last one, given the "recency effect" that could occur when long options are read on the phone. All other things being equal, in telephone surveys, people are more likely to choose the last option in a series than they are to choose the first option.

The net impact of this preceding question appears to be one of emphasizing conservation, and in that context, support for oil drilling in ANWR is considerably lower than in the more "pure" context of Zogby's May 2004 survey, or in the context of "reform" and reducing our dependence on foreign oil that is found in the Harris survey.

Finally, the Luntz survey's basic questions on oil drilling also appear quite neutral as well as balanced. However, they are both preceded by 13 questions that ask about the high price of oil and gas and the best ways of reducing America's dependence on foreign oil. Then the questionnaire presents three knowledge questions about ANWR (if people know where it is, what the initials stand for, and how much people think they know about it), followed by the basic support/oppose questions on the "exploration, development, and production of oil and natural gas" in the Arctic National Wildlife Refuge/ANWR.

With the issue implicitly framed as one of reducing U.S. dependence on foreign oil and dealing with the high price of oil and gas, Americans express support for oil drilling in ANWR by a decisive margin -- although less so if they hear "wildlife refuge" than if they hear only the lesser known acronym, ANWR.

What is the "real" status of public opinion on this issue?

My sense is that for most people, the issue is so obscure that they have no opinion on it at all. When asked totally out of the blue whether they support or oppose oil drilling in the "Arctic National Wildlife Refuge," many people probably oppose oil drilling simply because we have just told them this is a "wildlife refuge," implicitly suggesting it is designed not to be developed in any way. If other questions in the survey further bolster that environmental context, then opposition increases -- as we see in the Zogby poll.

On the other hand, when the issue is presented as a way to reduce dependence on foreign oil, then many people decide that oil drilling would be good thing. That is what we apparently saw in the Harris poll, and more extensively in the Luntz poll. The latter poll also bolsters the argument that reading "wildlife refuge" elicits greater opposition than reading the more obscure term, "ANWR."

Opinion polls have their uses. But they cannot create real public opinion where none exists. These three polls all provide insights as to what Americans might think about the issue, depending on the context within which it is framed. Right now, however, we don't know what most Americans really think about oil drilling in the Arctic National Wildlife Refuge, because they don't either.