01 September 2006

The perils of polling

[This is the third and final instalment of a three part series looking at public opinion polling and provincial government media relations. In the first installment - Playing the numbers - we looked at the broad issue of news releases output and content. In the second - The media and the message - we looked at the media relations activity of the Williams administration. In the last installment, we'll pull the whole thing together and take a look at why government does what it does.]

The grass isn't always greener; sometimes it isn't even grass

In politics, astroturfing is the use of orchestrated activities designed to create the illusion of popular support for or opposition to a candidate, a political party or a government program.

The word is a play on grassroots - genuinely spontaneous political action - and derives from the artificial cover used on sportsfields around the world.

Danny Williams' administration, like the two Liberal administrations before it, uses astroturfing to plant letters to the editor but more typically callers to radio talk shows. The plants get e-mails or telephone calls giving them the approved comments for the day or for the topic. More often than not they turn up alongside a raft of cabinet ministers, parliamentary assistants and backbench members of the provincial legislature extolling the virtues of their party and condemning its critics.

What makes this type of astroturfing effective is a combination of several factors. First, call-in radio shows are all found on a single outlet and occupy eight and half hours - that's right, a full workday - from Sunday to Friday.

Second that outlet - VOCM - has a perfectly valid approach to its news programming that simply presents what it is given without much filtering.

Third, VOCM has such an audience that its news and current affairs programming is heard by one of the largest radio audiences in the province.

Fourth - and no less important - is the near complete absence of an effective opposition. The New Democrats under former leader Jack Harris spent more time agreeing with Danny Williams than disagreeing despite the fact that theoretically a conservative party and a social democratic party should be ideological polar opposites.

The Liberals under both Roger Grimes and now Gerry Reid have been consistently unable to mount a coherent and sustained round of political attacks that would be normal for the party with the second largest standings in the legislature. Even though the Liberals under first Tobin and then Grimes used the astroturfing technique extensively in government, the Liberals now on the opposition benches seem unable to grasp the simple notion that they can influence public opinion just as easily as the government party can.

Managing the information flow

In the second part of this series, we looked at the Williams administration's approach to media relations. It encompasses astroturfing the radio shows and newspapers' letters pages. It also includes orienting their interviews to isolate news organizations one from the other. There is nothing illegal or unethical about it, but there is no question that on some issues - as in the Ruelokke matter - Williams will give special attention to the other major news outlet in the province, NTV, that also follows the approach of presenting its news with little filter or interpretation.

Make no mistake, the major news organizations in the province are universally professional, competent and ethical. There is absolutely nothing improper about the way NTV and VOCM report news, nor is there any explicit or implicit value judgment in the difference between the way NTV and CBC, for example, or NTV and The Telegram handle the news.

That said, sophisticated news consumers should be aware that the provincial government can and does vary its approach to certain media outlets. They show a marked preference for the two outlets with the largest audiences which also tend not to generate news on their own.

The provincial government also goes to considerable lengths to frustrate efforts to find out things that the public has every right to know but which government - for whatever reason - doesn't think it should have. The Telegram, for example, submits so-called access to information requests for government information that is supposed to be readily available under the province's Access to Information and Protection of Privacy Act.

Consider the number of times however, The Telegram has had to resort to every possible avenue of appeal to squeeze information from the current administration. One of the most infamous examples was a request in 2005 for polls commissioned by the Williams administration in 2004. Government did relent - six months after the privacy commissioner ruled against it, but it did its very best to stonewall and flat-out refuse to release information that would have given The Telegram something other than news on government generated by the government itself.

The perils of polling

One of the reasons Danny Williams tried to hang onto those polls is that they are extremely important clues as to how he governs. There wouldn't be any secret information in them but knowing what was researched and when might just provide some clues as to how Williams goes about his business. There were quite a few of them - almost one a month for a while - and they were all conducted by Ryan Research, the Premier's favourite pollster.

Bond Papers already discussed one of Williams' early polls, during the flag-flap in late December and early January 2005. Having pulled down Canadian flags, Williams was keen to know if he had made the right move especially after he started to receive tons of e-mails and letters in a genuine grassroots campaign aimed at expressing displeasure at his use of a national symbol for a political purpose.

Ryan's questionnaire was laid out in a way that would tend to produce support for Williams. But interestingly enough, despite what appear to be best efforts at skewing the results on the key question - about pulling down the Maple Leaf - respondents took the extreme negative choice. By the way, Williams likely got the results late on the 9th of January or early on the 10th before he went into a news conference on the Lower Churchill. In the newser Williams suddenly and unexpectedly announced the flags were headed back to the top of the flagpoles on government buildings.

As this series has demonstrated, polls are at the heart of much of Danny Williams' political communications. The poll in this case is the quarterly omnibus by Corporate Research Associates (CRA). There is simply no doubt that the current administration plays to the polls in an effort to goose the results. Those results - reported publicly by CRA as part of its own marketing - and covered throughout the province have become a well-established sign of political health for a government. It looks like the results are scientific and devoid of any undue influence, but the amount of energy Danny Williams expends in goosing the numbers suggests that they aren't.

CRA isn't doing anything wrong here any more than NTV or VOCM. In fact, CRA has no control over how its publicly-released results are used. CRA uses the results of their data collection as a marketing tool, nothing more.

That said, it should be understood that these polling results - like most publicly available polling results - don't necessarily reflect the actual mood in the province. Not only is the reported margin of error sometimes appallingly high - pushing 5% in some instances - we simply don't know how the data was collected or how the sample broke down with respect to age, geographical location, sex and other demographics all of which can influence results. Since there isn't another set of publicly-available polling with which to compare it, the CRA results are, in some respects, nothing more than a bunch of numbers.

All that is solid becomes air

The Williams administration plays to the CRA polls in order to use the polling results as evidence that Williams is politically unassailable. Not only is he politically unassailable, but his supporters often use his supposed popularity as a means of stifling dissent within the province and foreclosing any debate of substantive public policy issues. It doesn't matter if you don't agree with his policy, so the line goes, everyone else does: "See the poll results?"

The Williams administration supports its poll goosing by controlling the flow of information to the extent that it can. This makes informed public debate - and genuine informed consent - almost impossible to obtain. It also means that government is, on many levels, fundamentally unaccountable

As in the case of the recent release from finance minister Loyola Sullivan, information on something as fundamental as the province's financial health can be presented in a misleading way as part of the poll goosing efforts. So effective is the effort that even the opposition finance critic will swallow Sullivan's claim from the news release that he wound up with an extra $68 million in the bank last year. The reality - buried in an attachment to the release - shows that government actually wound up with over $700 million in surplus on current account and a net surplus of over $500 million for government operations as a whole.

Beyond the implications for the health of democracy in the province, playing to the polls puts the Williams administration in a precarious position. So much of its energy is devoted to poll playing that it seems unable to make difficult decisions for fear of the impact on the poll results. Promises are made, in Harbour Breton, on the fishery, in Stephenville that are popular in the short-term but which are not kept. The Premier's fiscal plan based on supposedly reliable financial information and announced on January 5 2004 is abandoned less than six months after it is announced simply because his polling numbers suffered too drastic a decline.

This failure to adopt a course for good reasons and stick to it is abandoned on public opinion polls which themselves are subject to statistical variation anyway and which are measuring something which is inherently variable from moment to moment. The result is a government which can develop a politically deadly credibility problem and which, in the minds of potential financial investors in the province - for example becomes simply too unpredictable and unreliable.

For the politicians, their problem is masked by attention to the polls. The polls become a self- imposed delusion.

Ultimately, the Williams administration's addiction to good poll results makes it vulnerable to any slipping of the numbers. A slip in the next poll can easily become the crack in the dyke that simply continues to spill in the months leading up to the next election.

The story for many will become the implications of a decline when it occurs. If it is a 10% decline, Williams will likely find himself dealing not only with the real issues of economic slowdowns themselves the result of government decisions, a scandal in the legislature which is far from over and a general lack of progress on any major economic files. He will also be facing the very real implications as the illusions he has worked to create vanish before his eyes.