Fall 2001, Vol. 4, No. 3

Table of Contents
Fall 2001

Quick Links:
Other articles
in this issue

From the Editor: The Civil Religion Goes to War

Religion After 9-11:

Good for What Ails Us

Falwell and Robertson Stumble

Islam is Everywhere

Pacifism on the Record

When Our Allies Persecute

No Bad Sects in France

The Stem Cell Conundrum

Gain, No Pain

On the Beat: Covering Religion in Hard Times


Letter to the Editor

To the editor:

In an article entitled "The Perils of Polling" in the summer 2001 issue of Religion in the News, Dennis Hoover claims that we "misreport[ed]" our findings from a poll we conducted in March 2001 on public reactions to President Bush’s faith-based initiative and that various journalists followed our lead, "rubb[ing] some undeserved salt" in the wounds of the faith-based initiative. In this assertion, Hoover not only misunderstands the survey’s results, but also ignores key evidence that contradicts his arguments.

Hoover’s fundamental criticism is that we overstate the level of public opposition to the involvement of certain non-Judeo-Christian groups in the provision of government-funded social services by asking respondents whether they favor or oppose different groups "applying" for such funding (which he calls a "highly ambiguous" and therefore meaningless question), as opposed to asking respondents if these groups should be "allowed to apply." This critique misunderstands the intention of the question, which was designed to measure whether individuals favor or oppose government funding of certain religious social service providers, regardless of whether it is an issue of allowing groups to apply, wishing they wouldn’t apply or somehow hoping they would not succeed in their application. Hoover may be correct in his assumption that some Americans might oppose Buddhist temples applying for government funding, while at the same time saying they support a policy allowing temples to apply. We believe, however, that our formulation of the question is a more thorough and inclusive measure of the public’s openness to certain religious groups’ participation in tax-funded programs.

More importantly, Hoover fails to address the fact that our survey found that a majority of Americans favor "giving government funding to churches and other religious groups" when there is no mention of denomination or faith, while at the same time pluralities oppose Muslim mosques and Buddhist temples, and majorities oppose the Nation of Islam, the Church of Scientology, and "groups that encourage religious conversion as part of the services they provide" even applying for such money. If that does not suggest that the public expresses strong support for the idea of faith-based groups receiving government funding while there are many reservations about extending that right to non-Judeo-Christian religious groups, we are not sure what would.

At the same time, when the evidence pointed in favor of government funding for faith-based organizations, we set it forth just as emphatically. Our report states, for example, that "[w]hen the arguments for and against allowing churches and other houses of worship to use government money to provide social services are pitted against each other, the positive arguments clearly outweigh the negative ones. The power of religion and efficiency arguments stand out as the most important predictors of support for faith-based funding, even when the five arguments against this approach are factored in."

We are also troubled by Hoover’s statement that "[t]he budget line suggested by the survey report—‘poll shows that public rejects specifics of charitable choice’—encouraged journalists to accentuate the negative." Our survey report, entitled "Faith-based Funding Backed, But Church-State Doubts Abound" did not suggest this "budget line"; a news service did. Indeed, because the survey’s findings were decidedly mixed on these matters, we were careful to avoid any statement that the survey provided wholesale support or rejection of the faith-based initiative generally or charitable choice particularly. In any case, we cannot be held responsible for the budget line suggested by a news service and it is unfair to suggest otherwise.

Furthermore, if one reads Hoover’s article, one would think that the poll was almost exclusively reported as a setback for the Bush faith-based initiative. In truth, however, there were a variety of stories with a variety of perspectives on the poll. On April 11, 2001, for example, the Washington Times story on our poll carried the following headline: "Bush’s faith-based initiative has public’s blessing." The article quoted Senator Rick Santorum stating: "There’s broad support for these healing institutions….We’ve got some details to work out, but the public support is there." Rep. Tony Hall "agreed with Mr. Santorum that the new survey gives the legislation a boost. [Hall stated:] ‘I don’t think it will pass the legislation, but it will help….That’s a substantial margin.’" Robert Woodson, president of the National Center for Neighborhood Enterprise, said of the survey: "It gives [the faith-based initiative] tremendous momentum….Politicians can read. It means [Americans] overwhelmingly support the president’s initiative." Hoover briefly mentions the fact that the Washington Times wrote a story on our survey, but he does not mention any of this information.

In short, we remain confident that our interpretation and reporting of the survey results were neutral and unbiased. However, we always make a point to provide the actual questionnaire and technical information so that journalists and academics can come to their own conclusions. For more information about the survey, please visit our websites at and

Melissa Rogers
Executive Director
Pew Forum on Religion and Public Life

Michael Dimock
Research Director
Pew Center for Research on People and the Press

Dennis Hoover replies:

I thank Melissa Rogers and Michael Dimock for their letter, but I am disappointed in their unwillingness to acknowledge the error that marred their survey report—a report that otherwise provided, as I said, "a bevy of interesting, timely data." I did not say the survey asked any "meaningless" questions nor did I suggest that the authors of the survey report intended to overstate public opposition to charitable choice. My central point was, and remains, that the authors in all probability did overstate such opposition by comparing generic apples to specific oranges.

The questionnaire first asked respondents whether they favored the idea of "allowing churches and other houses of worship to apply," along with other organizations, for government funds to provide social services. Later it asked about a number of specific religious bodies (including some non-Judeo-Christian ones), but this time it asked for opinion about such groups actually "applying," not about "allowing" them to apply. The survey report makes these substantively different questions out to be perfectly parallel, saying on page one that while most Americans support the right of religious institutions in general to apply, they "would not extend that right to non-Judeo-Christian religious groups."

Rogers and Dimock concede the problem yet somehow ignore its implications. They admit in their letter that, "some Americans might oppose Buddhist temples applying for government funding, while at the same time saying they support a policy allowing temples to apply." Exactly. For Buddhists and other specific groups, public opinion would very likely have been more favorable if the question had really been asked the same way as the question about generic houses of worship. Conceivably, some non-Judeo-Christian groups might have received a plurality, perhaps even a majority, of support. A swing of only 5 percentage points in the balance of Buddhists’ numbers would have given them a plurality.

Most of the news coverage, including the Washington Times’, repeated the claim that Americans oppose allowing non-Judeo-Christian groups to apply. This "finding" was widely reported as a significant bit of bad news for charitable choice—understandably, since open eligibility is virtually the essence of the policy. As it stands, journalists need to know that when the survey report makes claims like, "46 percent wouldn’t allow Buddhists to apply," all that can safely be reported is that 46 percent would rather they didn’t. Some proportion of this 46 percent might be willing to go so far as to deny Buddhists the right to even apply, but we must await a future survey to give us this number.

Hit Counter