The Numbers Guy on Domestic Eavesdropping Polls

Ah the travails of the not-yet-full-time blogger.  Expecting an unusually busy day, I woke early this morning to write up a post on this week’s column by Carl Bialik, the Wall Street Journal‘s "Numbers Guy."  Then I somehow managed to leave home without copying the draft to my thumb drive.

So with the day-job pressing, let me quickly recommend Bialik’s column, free and available to all, which takes an MP-like look at recent polling on the Bush administrations controversial domestic eavesdropping program.  Bialik’s bottom line: 

A half dozen polls on the issue have turned up different conclusions, and a key distinction appears to be the way pollsters identify the people who might have their emails and phone calls monitored as part of an effort to fight terrorism. Recent poll questions have referred to "suspected terrorists," "people in the United States" and "American citizens."

As always, Bialik’s work is worth reading on it’s merits, although the fact that Bialik quotes me and includes multiple links to this site doesn’t hurt.  Again, I have some additional thoughts that I’ll post later tonight.

UPDATE: I want to reiterate a point I made to Bialik about "non-attitudes."  Most Americans, by their own report, are not following the NSA wiretapping story closely.  As Bialik notes, Gallup found 31% have been following news about the story very closely (for complete results and wording on this and other questions in this post, see the Polling Report’s compilation of results on the terrorism and the wiretapping issue).

The problem is not that Americans lack attitudes about wiretapping or how far the government should go to pursue terrorists.  They do.  The problem is that the Bush/NSA wiretapping story involves both attitudes, plus views on wiretapping without a court order, partisanship and general views about President Bush. Moreover, most Americans have not been following that story closely.  So when a poll question describes the various elements of the Bush/NSA wiretapping program, respondents are likely to draw cues from the language of the question and answer the question accordingly. 

Bialik also cited two questions from the most recent CBS New York Times survey that, to MP’s eye, seem to tap attitudes that are more likely pre-existing.  They get very different results: 

In order to reduce the threat of terrorism, would you be willing or not willing to allow government agencies to monitor the telephone calls and e-mails of ordinary Americans on a regular basis?
28% willing, 70% not willing, 2% unsure

In order to reduce the threat of terrorism, would you be willing or not willing to allow government agencies to monitor the telephone calls and e-mails of Americans that the government is suspicious of?
68% willing, 29% not willing, 3% unsure

So 68% of Americans are willing to allow wiretapping of "Americans the government is suspicious of," but only 28% are willing to allow such monitoring "ordinary Americans on a regular basis."  Since the NSA wiretapping program does both, and also involves the issue of doing so without a court order, survey questions about will get results somewhere in the middle.

Consider the language of three survey questions that try to put it all together: 

NBC News/Wall Street Journal – As you may know, since 2002, the Bush Administration has been using wiretaps to listen to telephone calls between suspected terrorists in other countries and American citizens in the United States without getting a court order to do so. Do you approve or disapprove of the Bush Administration’s approach on this issue?
51% approve, 46% disapprove, 3% unsure

ABC News/Washington Post – As you may know, the National Security Agency has been investigating people suspected of involvement with terrorism by secretly listening in on telephone calls and reading e-mails between some people in the United States and other countries, without first getting court approval to do so. Would you consider this wiretapping of telephone calls and e-mails without court approval as an acceptable or unacceptable way for the federal government to investigate terrorism?
53% acceptable, 43% unacceptable, 1% unsure

CNN/USAToday/Gallup – As you may know, the Bush Administration has been wiretapping telephone conversations between U.S. citizens living in the United States and suspected terrorists living in other countries without getting a court order allowing it to do so… [Continues on subsequent question] Do you think the Bush Administration was right or wrong in wiretapping these conversations without obtaining a court order?
46% right, 51% wrong, 3% unsure.

Not surprisingly, Americans are more closely divided, with support for the program falling somewhere in between support for routine eavesdropping on "ordinary Americans" and wiretapping of those "that the government is suspicious of."   

The CBS/New York Times poll added a split sample experiment showing that an added rationale regarding terrorism – "this was necessary in order to reduce the threat of terrorism" – increased support for the program by seven percentage points:

After 9/11, President Bush authorized government wiretaps on some phone calls in the U.S. without getting court warrants, saying this was necessary in order to reduce the threat of terrorism. Do you approve or disapprove of the President doing this? [Asked of random half sample]
53% approve, 46% disapprove, 1% unsure

After 9/11, George W. Bush authorized government wiretaps on some phone calls in the U.S. without getting court warrants. Do you approve or disapprove of George W. Bush doing this?" [Asked of random half sample]
46% approve, 50% disapprove, 4% unsure

Bialik’s piece also includes comments from Jeff Jones of Gallup and Scott Rasmussen on why they worded their questions the way they did.  Again, it’s worth reading in full.