How Many Muslims Are There in the U.S.? New Pew Figure Sharpens Debate

(Below: Muslim American high school students gather in October, 2006, for a community iftar, or fast-breaking, during the month of Ramadan. PHOTO BY ANDREA USEEM)

A new study, released yesterday (May 22, 2007) by the Pew Research Center, estimates a total Muslim population in the U.S. of 2.4 million – about 0.6 percent of the total U.S. population.

This 2.4 million number, which includes children, is hugely smaller than the number estimated by the Council on American-Islamic Relations, which says on its website, “There is no scientific count of Muslims in the U.S. Six to seven million is the most commonly cited figure.”

Ibrahim Hooper stood by the 6-7 million figure today, saying the Pew number was “extremely low,” and that he believed a number of factors contributed to undercounting.

“I wouldn’t appear on that survey, with the last name of Hooper, and neither would any number of converts who don’t change their names. We also have a substantial immigrant population, and it’s more difficult for them to pop up in survey data,” Hooper told ReligionWriter by phone.

But the Pew figure cannot be easily dismissed on either of those bases.

The Pew survey did not attempt to find Muslims in the United States by searching for Muslim-sounding last names. Rather, it screened nearly 60,000 households, some of which were likely to contain Muslims, and weighted its results accordingly.

The set-up of the Pew study also made several attempts to increase response rates. Because immigrant experts estimate that a quarter of recent immigrants have limited or no English, the researchers made available foreign language interviewers who spoke Arabic, Farsi or Urdu. For this reason, the study said, researchers found a higher number of Muslims than had been discovered in previous Pew national surveys.

But given the level of suspicion among Muslims about unfair harassment and government surveillance – a finding confirmed in the report itself – wouldn’t many Muslims be reluctant to identify their religion to a stranger on the phone and answer detailed questions about their religious and political views?

The Pew study may not have been entirely successful, but they did take two steps to remove this barrier. First, the interviewers revealed the purpose of the study early on in the conversation, saying, “We have some questions about the views and experiences of Muslims living in the United States. I think you will find these questions very interesting.” In addition, Muslims were offered $50 for completing the half-hour interview.

Revealing a study’s purpose is “not common in survey research,” the study said, but the Pew researchers said they hoped this transparency would put Muslims ease and have “a greater chance of establishing a bond of trust,” according to the report.

Secondly, male interviewers were available to interview male subjects, and females for females, “a practice common among survey researchers conducting face-to-face interviews in majority Muslim nations,” the report said.

Andrew Kohut, president of the Pew Research Center, said at yesterday’s press conference that Pew’s estimate of the population – 2.4 million – was actually somewhat higher than other survey-based studies.

These estimates ranged from 0.2 percent of the population (found by the National Election Study conducted by the University of Michigan in 2000 and by Baylor University in 2006) to 0.5 percent (found by a compilation of Pew Research Center surveys between 2000-2007 and by the General Social Survey from the University of Chicago, which has been conducted every other year since 1998.)

Two Muslim social scientists, Ilyas Ba-Yunus, an emeritus professor of sociology at the State University of New York at Cortland, and Kassim Kone, a professor of anthropology at the same university, wrote that survey-based data are not a good basis for population estimates. In their chapter on Muslim demography in the 2004 book, Muslims’ Place in the American Public Square, they decried “the myth in American culture about the scientific validity of surveys based on national polling,” pointing to cases were survey data has, for example, wrong predicted election outcomes, including Al Gore’s initial win in Florida in the 2000 presidential election. Ba-Yunus and Kone estimated a 2004 Muslim American population of 5.7 million, based on data collected from Muslim organizations across the country.

Hooper agreed with Ba-Yunus and Kone that accurate information can only be obtained through the community itself. “We have the most diverse religious community, and it’s hard to get a handle on it unless you have access to it, as we do,” said Hooper.

Would CAIR be open to looking more closely at the Pew survey?

Hooper said an on-staff research expert would have to examine the data closely.

He said, “There have been innumerable estimates over the years, some higher, some lower. It doesn’t change our work in serving the Muslim American community.”