NSA spying, Edward Snowden, and the Polls: Misrepresenting public opinion

(This article was originally published on August 24, 2014 in iPolitics under the title “Spying blind: How polls provide cover for domestic espionage”)

Using inappropriately vague and misleading questions, polls have found an American public evenly divided in their support of NSA domestic espionage — and on whether Edward Snowden’s role in revealing the breadth and depth of it makes him a patriot or a traitor. Closer scrutiny indicates these divisions are more likely the result of systemic methodological biases in the polls than an expression of genuine opinion. This points to a far more troubling problem: Bad polls subvert a fair and balanced public debate on mass government spying, resulting in potentially anti-democratic remedies.

And Canadians shouldn’t tell themselves this is just an American problem. Testifying before a Senate committee, Canadian intelligence agencies seem to feel that mass spying falls nicely in their bailiwick, legal constraints be damned. Regrettably, there is no Canadian Edward Snowden to blow the whistle on these operations.

A poll conducted by Pew Research in July of last year is a good example of this built-in bias. It asked respondents if they “approve or disapprove of the government’s collection of telephone and Internet data as part of anti-terrorism efforts”. The results showed 50 per cent approved while 44 per cent disapproved.

If a respondent didn’t know much about the issue, spying on some phone calls that could lead to exposing terrorist plots would seem like a reasonable trade-off between protecting privacy rights and fighting terrorism.

In fact, many knew very little about the issue. When Pew asked if respondents had heard about this government spying program, about 50 per cent indicated they heard little or nothing at all.

But what if, before answering the question, respondents were given the background to make an informed choice? What if they were told that it wasn’t just “some phone calls” but all phone calls? And what if they were told this monitored phone traffic wasn’t just among legitimate terror suspects but included family members, friends and associates? And not just phone calls but all emails, texts and other Internet communications, including website visits, over a period of years?

And what if they were told also that for all the years that this spying system has been in place, it was almost completely ineffective in thwarting terrorism? By the NSA’s own admission, of the 54 terrorist activities that were intercepted, only one or at most two relied on mass telephone spying (most intercepts involved non-telephone spying such as the PRISM Program). Would the average American see any economic sense in the government continuing to spy on the telephone calls of all of its citizens?

That mass telephone spying is ineffective in deterring terrorism was, in fact, the conclusion of President Obama’s own review panel on NSA spying. It recommended that the telephone spying program be significantly curtailed and provided with much more oversight than before. It also was sharply criticized by the Privacy and Civil Liberties Oversight Board, an independent federal agency which concluded that NSA telephone spying had “minimal” counterterrorism benefits, is illegal and should be shut down. In another rebuke, a Federal Court judge also questioned the constitutional legitimacy of the program.

Prior to all these revelations, a Pew poll reported that 53 per cent of Americans said they believed that “the government’s collection of telephone and Internet data has helped prevent terrorist attacks.” That was the administration line. But the evidence does not support the claim.

Had the public known of the shortcomings of the spying program — and had poll respondents been more fully apprised of its invasiveness and ineffectiveness in the wording of the questions — those poll results suggest that support for the program likely would  have dropped significantly.

To be fair, it wasn’t just Pew using vague and misleading wording. Other pollsters were equally to blame. A CBS News poll used phrases like “… to reduce the threat of terrorism … (by) collecting phone records of ordinary Americans”. ABC News/Washington Post used the phrase “…extensive records of phone calls … to try to identify possible terrorist threats”. A Time poll mentioned the need “… to prevent terrorist attacks by collecting data on telephone dialing records of U.S. citizens”.

Ironically, the sensitivity of the response to question wording was revealed by a parallel methodological study conducted by Pew Research. It found that support for NSA spying was very much dependent on the wording of poll questions. For example, if the NSA spying program was described as collecting “data such as the date, time, phone numbers and emails … with court approval as part of anti-terrorism efforts”, the study found the program had 41 per cent in favor and 49 per cent opposed. However, if the program was described as “collecting recordings of phone calls or the text of emails (of nearly all communications in the U.S.), with no mention of either courts with the goal of fighting terrorism”, the level of support dropped to 16 per cent in favor and 76 per cent opposed — an increase in opposition of 27 per cent.

Armed with this information, why wasn’t Pew Research more careful in its choice of wording for the questions?

Perhaps the central reason was that the program’s lack of an impact on terrorist activity only emerged clearly after the Pew survey was conducted. Prior to Snowden’s revelations about mass telephone spying, Director of National Intelligence James Clapper lied to Congress under oath when he stated that U.S. intelligence agencies did not collect telephone data on millions of Americans. After the revelations, the chief of the NSA, General Keith Alexander, assured Congress that this mass spying program helped thwart 54 terrorist operations. Later, he was forced to admit that the actual number of terrorist operations disrupted by the program was closer to … one.

The other important revelation that has emerged since the survey was taken is that most of the program operated in a de facto warrantless search environment. NSA analysts could spy on anyone, whenever they wanted.

This is crucial in understanding Snowden’s concern with the spying program. Spying was never the issue; the question was and is whether it is done in accordance with the law. Commentary from independent judges and constitutional lawyers casts the program’s constitutionality in doubt, and questions whether the role of FISA, the secretive oversight court, was to be nothing more than a rubber stamp for all forms of NSA snooping.

That said, it’s hardly surprising that a large segment of the American public regards Snowden as a traitor (49 per cent, according to an Angus Reid poll). The U.S. has been subjected to a massive vilification campaign against Snowden for his leaking of NSA spying documents — a campaign driven by the establishment in both the Republican and Democratic parties, as well as the White House. It’s one of those rare national issues that unites both traditional Democrat and Republican voters but finds less support in an odd mix that includes Independents, Tea Party types, conservatives and voters under 30.

This campaign portrays Snowden as a traitorous criminal bent on endangering national security by revealing confidential state secrets in a way that aids terrorists. The reality is that, in the absence of warrants and in an environment of secrecy and deceit, the truly dangerous crimes were committed by NSA. This is an organization that both parties had a hand in creating. It has become a rogue operation that now needs their protection.

In light of this, the ongoing debate as to whether Snowden is a patriot or a traitor seems silly. It simply demonstrates how easy it is to manipulate public opinion when the electorate is ill-informed, the subject matter is complex and terrorism is on the table. That word seems to elicit the most irrational fears — a sort of national hysteria — among Americans. Polls show that year after year (CBS News poll, June 9-10 2013 in PollingReport.com), between 40 per cent and 70 per cent of Americans believe the country will “likely” suffer a terrorist attack in the next few months.

Snowden’s decision was clearly motivated by a desire to see the NSA function within constitutional constraints.  Its failure to do so, and its willingness to have its officials lie under oath before Congress, make the organization liable to criminal prosecution.

The wise men who drafted the American Constitution understood the purpose of the Fourth Amendment — which guarantees privacy rights — was not to protect individuals who may be doing something illegal. Its intention was to limit the powers of government and the potential for abuse. When a government decides to override Fourth Amendment rights, it is the government that is behaving in a criminal manner, not the citizen. That’s where Snowden and the U.S. government parted company.

An ill-informed public is easily swayed by propaganda. Pollsters need to help information-challenged respondents. One traditional approach is to provide a short but accurate preamble that provides crucial and relevant information a respondent would need to know in order to meaningfully answer a question. Pollsters don’t like to do this. It makes questions wordy, slows down the survey, and reduces response rates. Nevertheless, it’s the price they need to pay to get reliable data.

Given what we know today about NSA spying, these polls simply cannot be relied upon. Polling firms need to go back into the field with properly tested questions — and do it right this time. Otherwise, it’s just a question of whose propaganda is more compelling.

About Oleh Iwanyshyn

Oleh Iwanyshyn has been involved professionally with surveys from the mid-70s when he started as a methodologist at the Institute for Behavioral Research at York University. Later, while at the Canadian Broadcasting Corporation his focus shifted to media and election surveys. He now runs his own survey research company, ViewStats Research, established in 1997 and specializing in online surveys. Public opinion surveys is an essential communication tool between a democratic society and its leaders. Unfortunately, surveys can be very easily manipulated. Revealing such manipulations and their consequences is the raison d'être of the poll stuff blog.
This entry was posted in Polling and tagged , , , , , , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.