With a spotlight cast on American Muslims since Sept. 11, one seemingly simple question has defied a clear answer and become the focus of a politically charged dispute: What is the size of the U.S. Muslim population? Four major Muslim organizations released a study in April that estimated the population at 6 million to 7 million. Based in part on that report, most media organizations, as well as the White House and the State Department, have said in recent weeks that there are at least 6 million Muslims in the country.
But two studies released last month, including one commissioned by the American Jewish Committee, concluded that the total is much lower: no more than 3.4 million and perhaps as few as 1.5 million.
Nihad Awad, executive director of the Council on American-Islamic Relations, one of the sponsors of the April report, condemned the AJC-sponsored study, calling it part of an effort by the Jewish community to "marginalize" Islam in the United States.
"Why are they worried about our numbers? What's it triggering?" Awad asked. "We have never misrepresented our figures and have never been interested in competing with any other faith or ethnic community."
David Harris, the AJC's executive director, said his organization had long suspected that the U.S. Muslim community was inflating its population figures. With the attention brought to Islam by the terrorist attacks and the "wildly divergent" figures quoted in the media, Harris said the AJC decided it was time to commission an analysis.
"It's not about numbers; it's about truth and accuracy," Harris said. "If a group born yesterday suddenly says it has 8 million members, that has societal consequences. If it's true, God bless them. If not true, do we go with the manufactured number?"
Religious denominations, like all interest groups, can gain or lose political clout based on perceptions of their size, said J. Gordon Melton, director of the Institute for the Study of American Religion in Santa Barbara, Calif. In the case of the U.S. Muslim community, Melton said, its efforts to influence policy in the Middle East would get a boost if it were viewed as being larger than the country's Jewish population, which is estimated at 6 million.
"It's a political question: How does it sway votes?" he said.
Awad said the council's interest in U.S. Muslim population estimates has more to do with its desire to have a voice on domestic issues such as health care, education, crime and drug abuse.
The conflicting studies are not simply a case of sponsors with different political agendas, however. The gap in the numbers also illustrates the problems that demographers have long faced when trying to count religious populations.
Definitive numbers don't exist in part because the U.S. Census, the most extensive survey of American society, is prohibited from asking about religious affiliation. Religious groups, when contributing population figures for reference books, most often rely on self-reported membership figures from houses of worship.
Even then, numbers often are not comparable. Some denominations count anyone on the rolls, including babies, while others consider only baptized adults.
Islam presents a particular challenge, because mosques typically do not maintain membership lists.
The April report co-sponsored by CAIR, titled "The Mosque in America: A National Portrait," was the Muslim portion of the largest U.S. denominational survey ever, a project coordinated by the Hartford Institute for Religious Research.
Researchers called the nation's 1,209 known mosques and interviewed leaders at 416 of them. Respondents were asked to estimate the number of people involved in their mosque in any way. The average response was 1,625 participants. Multiplying that figure by the 1,209 mosques, lead researcher Ihsan Bagby determined there were 2 million "mosqued Muslims" in the United States.
Bagby, a professor of international relations at Shaw University in Raleigh, N.C., multiplied that number by three to account for people who identify themselves as Muslims but might not participate in mosque activities. He calls this multiplier an educated guess based on years of observation of the Islamic community.
Paul Perl, a research assistant at the Center for Applied Research in the Apostolate at Georgetown University, analyzed the data for Bagby and wrote a preliminary draft of the report. But Perl said he did not see the final version before publication, including Bagby's population estimate of 6 million to 7 million.
"I don't think there is an easy way to go from the number of people at mosques to a total population figure," Perl said. He also said the average figure Bagby used on mosque participation might have been too high, noting that two imams in the survey estimated that their mosques had 50,000 participants.
Carl Dudley, co-director of the Hartford denominational project, said it is not uncommon for religious groups to multiply worship attendance figures by three, five or even seven to obtain an estimate of total adherents. "The whole thing is a little slippery," he said.
CAIR's Awad, asked why his group settled on an estimate of 7 million in its press statements rather than Bagby's range of 6 million to 7 million, said the organization had used 6 million for six years. "If we still used the number six," he said, "people would say, ãHaven't we grown?' "
The American Jewish Committee believed the number was wrong, so it hired Tom Smith, director of the General Social Survey at the National Opinion Research Center at the University of Chicago.
Smith analyzed 45 documents and papers, including public opinion surveys, worship attendance studies and immigration statistics. He concluded that his "best survey estimate" of American Muslims was 1.9 million but allowed for a range of 1.5 million to 3.4 million.
The AJC's Harris said his group knew it would be criticized for commissioning Smith to do the study. Smith was known to be skeptical about the figure of 6 million U.S. Muslims; last year, he told the Los Angeles Times that the number was "completely invalid" and that Muslim groups were "inventing an estimate."
But Harris said: "If Smith had said, ãOops, I made a mistake. The estimate is 10 million,' we would have published it."
Smith said he had been looking at the Muslim population issue since 1995 and planned to publish a report by the end of the year even before the AJC approached him.
On Oct. 24, a day after the AJC published Smith's report, the Graduate Center of the City University of New York released its American Religious Identification Survey 2001.
The university study was based on a random telephone survey. Members of more than 50,000 households -- a sample 25 to 50 times larger than in most national surveys -- were asked, "What is your religion, if any?"
Researchers adjusted the figures to account for such factors as nonparticipation by immigrants who did not speak English or who were afraid to respond because they were from countries where publicizing one's religion can result in reprisals.
Projecting the numbers in the sample to the overall U.S. population, the study's authors put the number of Muslims at 2.8 million.
David Barrett, a demographer whose staff provides annual U.S. and world religion estimates for Encyclopaedia Britannica and 20 other yearbooks, last year estimated the U.S. Muslim population at 4.1 million.
Barrett said he has reservations about each of the other three studies. He said he was "not all that impressed" by Smith's study because it relied on old material. He said household telephone surveys, even one as large as CUNY's, miss people without telephones and do not account for teen-agers who have different beliefs from their parents. The mosque study, he said, relies too heavily on self-reported figures.
Barrett said his own method is to analyze population data from a variety of sources, including United Nations reports. For U.S. Muslims, many of whom are immigrants, he looks at the percentage of Sunnis, Shiites and other sects in their country of origin and projects that onto the immigrant population here.
The large gaps between the various estimates suggest that more research needs to be done and that demographers would benefit from sharing data and variables specific to the Muslim community, said Bryan Froehle, executive director of the Georgetown research center, which was hired to analyze the mosque figures in Bagby's report.
"What their study does is force us to look more sharply, to think of other ways to get at this issue," Bagby said of the CUNY report. "I just wish it wasn't a political hot potato."