What Do 44% of Americans Believe?: Newspapers and websites around the world are reporting what purport to be the results of a Cornell University poll on attitudes of Americans towards Muslims in the United States. The newspapers and websites are reporting that 44% of Americans say that they want to curtail the civil liberties of Muslim Americans. Here is the summary from the Associated Press:
  Nearly half of all Americans surveyed said they think the US government should restrict the civil liberties of Muslim Americans, according to a nationwide poll.
  . . .
  The survey indicated that 44 percent of those surveyed said they favored at least some restrictions on the civil liberties of Muslim Americans. Forty-eight percent said liberties should not be restricted.
  These are extremely disturbing figures that will be accepted by hundreds of millions or even billions of people around the world. But there is something important that the press reports overlook: the 44% of people polled did not actually say that they wanted to curtail the civil liberties of Muslim Americans. Rather, 44% of peple reported views that the Cornell University pollsters themselves categorize as being support for the curtailment of the civil liberties of Muslim Americans.

  I found the report on the poll here. It turns out that the pollsters asked people to agree or disagree with four statements:
1) Muslim civic and volunteer organizations should be infiltrated by undercover law enforcement agents to keep watch on their activities and fundraising.
2) U.S. government agencies should profile citizens as potential threats based on being Muslim or having Middle Eastern heritage.
3) Mosques should be closely monitored and surveilled by U.S. law enforcement agencies.
4) All Muslim Americans should be required to register their whereabouts with the federal government.
For each of these statements, between 20 and 30 percent of the subjects agreed; most disagreed. Overall, the study reports, 29% of the subjects agreed with 2 or more of these statements, and 15% agreed with one of them. (Some of these numbers don't quite add up, I think, but see page 6 of the report for the figures.)

  I don't want to be nitpicky, but am I right in thinking that a certain amount of spin is involved in how this poll is being reported? The pollsters made a judgment call that if you agree with any one of these statements, you are in favor of curtailing the civil liberties of Muslim Americans. Thus, the pollsters are claiming, and advocacy groups such as CAIR are trumpeting, that 44% of Americans are in favor of curtailing the civil liberties of Muslims.

  But is that really what the poll shows? Most of the questions are quite vague, and use lots of buzzwords. Take the statement: "Muslim civic and volunteer organizations should be infiltrated by undercover law enforcement agents to keep watch on their activities and fundraising." There have been many press reports of Muslim civil and volunteer organizations being used as fronts for terrorist financing schemes. If you believe these reports are probably true, or just may be true, you might reasonably want the FBI to investigate the organizations. You would then answer that you agree with the statement. Does that really mean you want to curtail the civil liberties of Muslims?

  Of course, this is not to say that the poll results are heartening. In particular, it is very disturbing that 29% of Americans would agree that "All Muslim Americans should be required to register their whereabouts with the federal government." I can imagine less damning explanations for this figure, but it is on the whole quite troubling. Nonetheless, the press reports around the world suggesting that 44% of Americans want to curtail civil liberties of Muslim Americans would appear to be at least misleading.

  UPDATE: I made some slight corrections to the original.

Related Posts (on one page):

  1. All or Nothing:
  2. What Do 44% of Americans Believe?:
All or Nothing:

I much appreciated Orin's post criticizing the "Nearly Half in U.S. Say Restrict Muslims" reporting of a recent survey. As Orin pointed out, here's what the survey actually measured, quoting this report:

1) Muslim civic and volunteer organizations should be infiltrated by undercover law enforcement agents to keep watch on their activities and fundraising.

2) U.S. government agencies should profile citizens as potential threats based on being Muslim or having Middle Eastern heritage.

3) Mosques should be closely monitored and surveilled by U.S. law enforcement agencies.

4) All Muslim Americans should be required to register their whereabouts with the federal government.

44% of respondents said yes to at least one of these questions.

To Orin's criticisms, let me add this one: Options 1 through 3 say nothing about under what conditions these procedures are to happen. People can have lots of views on them. Consider, for instance, option 1. Some people might say that all Muslim organizations should be infiltrated. Others might say that most should be. Others might say that the government should infiltrate those that it has some reason to believe are being used as recruiting centers for jihadism (that's my view). Still others might say that the government should never infiltrate any religious groups.

But the question lets people choose either "yes" or "no." So the count of those who would "restrict[] . . . civil liberties of Muslim Americans" would include those who would infiltrate all Muslim organizations, as well as those who would simply reject the extreme opposite position that any religious or political organization must be immune from surveillance. (Plus, of course, it's a judgment call whether one's "civil liberties" include immunity from government infiltration of groups to which one belongs -- there are arguments on both sides, but the Supreme Court has generally held that such infiltration doesn't violate either the First Amendment or the Fourth Amendment.)

The hypothetical proposal in "All Muslim Americans should be required to register their whereabouts with the federal government" would be pretty clearly a restriction on civil liberties; I'd reject it myself, and I share Orin's regret that it polled 29%. But it's also the only option that specifies "all," and that would clearly be unconstitutional.

Related Posts (on one page):

  1. All or Nothing:
  2. What Do 44% of Americans Believe?: