Juiced (II)

Back in the spring, VOCM brought in tech.changes to its often-abused and occasionally ridiculed Question of the Day feature, making it harder to "juice" the numbers.

Harder, mind you. Not unpossible.

The last of the old-style QotD's in the handy labradore virtual archive is from April 23rd. On that day, the question was, "Should the Chief Electoral Officer investigate the 2001 St. Barbe by-election?" 31,476 "votes" were logged - the eleventh highest daily total on (labradore) record.

Unsurprisingly for such an obvious juice, 63% said no.

However, it's interesting to see which questions do still generate an inordinate number of "votes" — and which way the votes end up going.

For the sake of argument, let's define a "juiced" VOCM QotD as one in which the total number of votes is twice the daily average. Before VOCM put in controls, as noted above, the votes could easily and quickly skyrocket into the tens of thousands.

(Interestingly, the new controls haven't greatly depressed the QotD participation on a typical day. The average daily vote count under the old system was about 2900. Since the new system, with verification, was brought in, it has averaged 2600. Disregarding "juice" days, participation is 2260 votes on a typical day now, vs. 2470 during a comparable span of time in the old system.)

Under the new regime, the cutoff for "juicing" is thus 2×2608, or 5216. By that definition, there have been four "juiced" polls under the new system. In addition, there are another five which had a 50% or more boost in participation compared to the average. On this column graph, showing the total number of votes by date, these nine are highlighted in blue. (No partisan implications with the colour scheme; these are Excel's default settings.)

The four juiced polls in question were, giving date, total number of votes and result:

September 16th: Do you think Wade Verge should resign? (5880, 79% no)

September 22nd, Is it time for the provincial government to reconsider the issue of indexing pensions for its public sector pensioners? (6696, 73% no)

October 3rd: Would you like to see another election for mayor in Paradise? (7439, 90% yes)

October 8th: Do you like the new cabinet? (7718, 67% yes)

It's striking that three of them bear directly on the performance of the Provincial Government. The fourth is a hot-button question emanating from the recent municipal election.

A similar pattern obtains in the next tier of lesser-juice polls:

June 29th: Do you think the economy is bringing more people home to this province? (4118, 62% yes)

July 23rd: Do you think marijuana should be legalized and taxed like tobacco and alcohol? (4188, 65% yes)

July 27th: Do you think water meters should be installed in every home? (3934, 59% no)

August 10th: Do you agree with government's decision to appeal the court ruling over the highway depots? (4756, 71% yes)

September 28th: Should we roll out the welcome mat for Moammar Gadhafi? (3923, 77% no)

The economic question has a close relationship to the political narrative that Danny Williams-Government has attempted to weave for itself. And really, are there that many people with such a firmly pro-government opinion on the issue of highway depots and the associated legal proceeding?

That leaves social hot-button marijuana (which could have easily been the target of a national or international juicing effort), a political hot button in the aborted Gadhafi visit, and another municipal hot-button issue from VOCM's main listening area in greater Danabad.

Of course, it would be ludicrous to suggest that there was any political or government effort to juice questions so that it would appear, when the announcer read the results of the "poll", that 73% of ordinary people were opposed to indexing civil service pensions, that 79% were clamouring for the hapless Wade Verge (hey bud, how's it going?) to stay in office no matter what, and that two thirds were positively rapturous in their praise for the latest iteration of the Danny Williams-Government Cabinet.

It would be almost as ludicrous to read anything into the fact that the next-biggest poll was on the question of highway depots and the government's appeal in the related labour case. And it would be beyond the realm of plausibility to make any inference from the marked upward trend in the outlying high-participation polls.

There was, and has never been, any organized effort to manipulate the Question of the Day for partisan propaganda purposes. Nothing could be further from the truth.