Libertarian push-polling at CPAC

A poll of participants at the just concluded CPAC gathering in Washington, D.C. asked, among other things, “Do you favor or oppose the National Security Agency’s (NSA) use of data collection such as phone-tapping and reading of emails to combat global terrorism?” 78 percent of respondents said they oppose; only 19 percent said they are in favor.

The question is ambiguous, I think. It could be construed to mean, “Do you favor the NSA using data collection such as phone-tapping and reading of emails to combat global terrorism?” I doubt that any non-terrorist American would want the NSA not to tap the phones or read the emails of known foreign terrorists, for example Osama bin Laden when he was still around.

As Patrick Brennan points out, even the “Amash amendment,” the failed attempt of liberal Democrats and libertarian-leaning Republicans to take away the NSA’s power to do what Edward Snowden revealed it did — gather metadata — would not have altered its ability to tap phones abroad, for example.

Thus, the question is better construed as a referendum on NSA’s approach to data collection that occurs in the U.S. as it pertains to American citizens. I assume this is the way those who responded to the question understood it.

The problem with CPAC’s question construed this way is that, as framed by CPAC, it presents a misleading picture of what the NSA is authorized to do with respect to American citizens, and a misleading picture of the controversial policies revealed by Edward Snowden. Brennan explains:

Th[e] controversy is over the fact that the NSA collects large amounts of data about Americans’ Internet activity and cell-phone usage without having access to the content of the communications. Obviously the NSA does a great deal of phone-tapping and reading of emails, but it isn’t authorized to do so inside the United States or with the communications of American citizens (without very specific authorization involving a warrant, etc.).

If CPAC wanted to conduct a fair referendum on NSA’s data collection practices with respect to Americans, it would have described the Agency’s practices more accurately, as Gallup (for example) did. It would not have described them, without more, in terms of wire tapping and reading emails.

Brennan pursued the matter with the pollster who conducted the survey, Tony Fabrizio. He asked Fabrizio whether he thinks the question gives an accurate impression of what the NSA does. Fabrizio responded that it was “an accurate description of what Americans are perceiving from the media.” Let’s take that as a “No.”

Brennan followed up by asking whether the impression Americans are getting is factually accurate, given that a lot of people mistakenly believe the NSA’s cell-phone programs include collecting the content of phone calls. Fabrizio stuck to his story, saying that the description he used — “phone-tapping and reading of emails” — is “what [Americans] are reading in the media.” So again, “No.”

Fabrizio then admitted that he had “dumbed down” the poll. But that’s an inaccurate description of what he did. The poll doesn’t oversimplify its description of NSA’s practices with respect to collecting data from Americans; it affirmatively misleads.

In short, CPAC engaged in push-polling. And when called on this, its pollster relied on “what Americans are perceiving from the media.”

Like Brennan, I find it ironic, after listening for three days to conservatives complain about how the mendacious mainstream media distorts the facts, to hear CPAC’s pollster rely on the mainstream media’s version of the facts pertaining to NSA data collection.