Loud on all you survey
There you go—the latest SWS political survey showed that the three prospective presidential candidates for 2016 were, statistically speaking, tied.
They were of course Sen. Grace Poe, Vice President Jejomar Binay, and former Interior Secretary Manuel Roxas, who were so far the only ones who declared their intention to seek the presidency in the 2016 elections.
This particular survey result, showing a neck-and-neck race, essentially asked a clear simple question: “Who will you probably vote for as president of the Philippines if elections were held today?”
In comparison, various media outfits earlier proclaimed—rather loudly—that another survey showed Poe as the runaway leader with 47 percent, Roxas with 39 percent and Binay with 35 percent.
The wonder of it all was that the surveys with extensively divergent results were done by the same outfit, the Social Weather Station, covering exactly the same dates, which were Sept. 2 to 5 this year.
Of course, most media reports went to town, accentuated by screaming headlines, proclaiming that, based on the “other” SWS survey, Roxas overtook Binay in “second place.”
Well, really now, second place for what?
As it turned out, the “other” SWS survey was rather a tricky one, in which SWS told some 1,200 respondents to give three names as their preferred successor to our leader Benigno Simeon, aka BS, and they had to pick the names from out of nowhere.
Take note that only politicos appeared in the list of preferred successors to our leader BS, but if only I was a respondent in the “other” SWS poll, I would give the names of those with confirmed integrity, such as Gaudencio Cardinal Rosales.
Anyway, there was nothing inherently wrong in the “other” SWS survey asking for three names. What was abhorrently wrong was that media took the liberty to mangle the statistics.
Look, news media pronounced Poe as the pulling away presidential contender and Roxas as having overtaken Binay. In a seriously skewed way, news media simply did a ranking of the top three prospective candidates from first to third.
And such a ranking, ladies and gentlemen, could never be determined even with just a hint of guesswork in the survey methodology. The “other” survey only showed that Poe was perhaps the most preferred leader, taken together with two other choices.
Well, for all we know, Poe could have been only the “third” choice, but she would still be counted as “number one” in the results, owing to the lack of ranking in the results.
So much so that, if you simply add the total percentage points garnered by all the prospective replacement to our leader BS, the total would hit more than 163 percent.
If such were the actual results of the elections, it would already mean millions of flying voters.
Thus, the “other” SWS survey that hit the news media like the giant waves of the tsunami would be what some honest professional pollsters would call a “compromised survey.”
For its part, the SWS of course said the “other” survey was simply meant to give the “undecided” among the respondents a real chance to express their choices by giving three names. Sure—and so it would now be up to the pollster to help voters decide?
It was just that ruling Liberal Party, which happened to be the party of our leader BS and, thus, the party of his chosen one, none other than Roxas, also recently used the same spin in a media release: That Roxas already overtook Binay in the polls.
Of course the LP claimed that its claim was based on a “secret” survey, even refusing to give the name of the survey company.
The wonder of it all was that the supposedly “secret” survey, by some luck, reached media outfits that, without much ado, fed the LP spin to the unsuspecting public.
In other countries, they already recognized some big problems with political polls in the past two years: They were spectacular disasters. There has been in fact a crisis in election polls.
Ultimately, in this country where elections could easily turn into bloodletting, the question regarding the 2016 elections remain: Could we really trust the pollsters?
