Did SWS flirt with its client?
By Mortz C. Ortigoza
Polls like what two major Philippines survey outfits’ Social Weather Station (SWS) and Pulse Asia (PA) have been doing became a major factor for the strategy of an elective candidate to map out how he /she improves his /her weakness or strength in an area or stop his/her endeavor to save time, efforts, and monies as election nears.
Earlier, some political kibitzers criticized SWS how it came with 137% total sampling respond on its September 2 to 5 2016 presidential polls where it found 47% for Senator Grace Poe, 39% for Local Government Secretary Mar Roxas, 35% for Vice President Jejomar Binay, and 16% for Davao City Mayor Rodrigo Duterte as against the standard practiced of a 100% based percentile poll.
The credibility of SWS had been questioned again by some sectors when it came days later with Survey No. 2 that was polled in the same September 2 to 5, 2015 period of Survey No.1 where a presidential aspirant Roxas landed No. 2 in the first survey while he landed No.3 in the second poll.
In No.1 poll, the top four presidential wannabes were sparsely distributed from each other while they were densely distributed in the No.2 poll. How these happen my dear SWS President Mahar Mangahas when those who were asked to answer were the same 1,200 people?
With 1,200 respondents divided by the 100% based percentile, the No.2 survey says 26% Poe, 24% Binay, 20% Roxas, 11% Duterte, 4% Roxas, 4% Senator Francis Escudero, 3% Manila Mayor Joseph Estrada, 2% Senator Merriam Santiago, 1% former Senator Manuel Villar, and 5% undecided and others.
Was there a manipulation that ensued in the Ranks 2 and 3 (between Roxas and Binay) to appease a customer after he saw how close the race for the presidential office?
Why the distance, in Survey No. 2, of Binay to Grace became statistically tied and that of Roxas to Binay became almost tied as based on the margin of error (MoE) of plus 3 or minus 3?
Why survey No. 2 differed to survey No. 1 when, again, they have the same interviewees?
From the projection of 60 million voters that would go at the voting precincts in 2016, each plus or minus is 600,000 votes. Does your math how the distances of each candidate to those behind or above them count.
In Survey No.1, SWS said that Poe dusted off by 8% or 4.8 million votes Binay. If you add the 3 plus or 3 minus MoE that 4.8 million could be 6.6 million votes or 3 million votes.
Why the wide discrepancy on Survey No.1 versus Survey No.2?
The first question in Survey No. 1: "Ayon sa Konstitusyon, ang termino ni Pang. Noynoy Aquino ay hanggang sa taong 2016 lamang at magkakaroon ng halalan para sa pagka-pangulo sa Mayo 2016. Sinu-sino sa palagay ninyo ang mga magagaling na lider na dapat pumalit kay Pang. Aquino bilang Presidente? Maaari po kayong magbanggit ng hanggang tatlong sagot." ("According to the Constitution, the term of Pres. Noynoy Aquino is up to 2016 only, and there will be an election for a new President in May 2016. Who do you think are the good leaders who should succeed Pres. Aquino as President? You may give up to three names"
In Survey No. 2 the same number of respondents were asked whom their bet for the 2016 presidential election by providing them with the list of the candidates and could even cite one that was not provided on the questionnaire they think to be their preferred candidate in the next year’s election.
In that two surveys, the economic and social backgrounds and area estimates of the respondents were taken from the list of the 62 million voters 2015 projections by the National Statistics Office.
I used 60 million votes as my projection because not all Filipinos go to the precincts every election day.
The 1,200 respondents were spread all over the country by 300 each in Metro Manila, Balance of Luzon, Visayas and Mindanao. It had MoE of 3% for national percentages, and 6% each for Metro Manila, Balance of Luzon, Visayas and Mindanao.
Margin of Error, to the uninitiated, is a confidence interval of plus-or-minus 3 percent, which equals an error range of 6 percent. The MoE can be lessened say by plus 2 or minus 2 if the respondents would be bigger by, say, 2,400.
SWS, according to its website, employs its own staff for questionnaire design, sampling, fieldwork, data-processing, and analysis, and does not outsource any of its survey operations.
With Roxas eclipsing Binay in Survey No. 1 and the latter overtaking the former in Survey No.2, did SWS muddle its two surveys before the eyes of the confused if not sceptical political spectators?
When asked what was more credible, the first or the second poll, SWS’s Survey Data Library Director Leo Laroza told CNN Philippines, the results from the second report “will have to be the one closer to that scenario.”
Let’s turn to Pulse AsiaWith these confusion and scepticism spawned by the SWS, the Filipinos can turn to the pollster’s rival Pulse Asia’s 1,200 respondents (with 3% plus and minus MoE) conducted in August 27 to September 3 survey.
The poll was commissioned by ABS-CBN TV.
The respondents were asked: “Of the people on this list, whom would you vote for as President of the Philippines if the 2016 elections were held today and they were candidates. You may mention others not included in this list.”
The poll showed 27% Poe, 21% Binay, 18% Roxas, 15% Duterte, 5% Marcos, 5% Estrada, 4% Senator Merriam Defensor-Santiago, 1% Cayetano, 1% former Senator Panfilo Lacson, 1% Don’t Know, 1% Refused,1% None.
Dr. Ana Maria L. Tabunda, Chief Research Fellow of Pulse, defended anew the survey’s using population sample size of 1,200 respondents out of more than 62 million registered voters in the Philippines. “As long as sampling is random, and representative of population, more or less, the results are reliable.”
In September 8 to 14 Pulse Asia’s asked 2,400 crowd with the following: “Of the people on this list, whom would you vote for as President of the Philippines if the 2016 elections were held today and they were candidates? You may mention others not included in this list.”
It polled 26% for Poe, 20% for Roxas, 19% for Binay, and 16% for Duterte.
Although Roxas and Binay interchanged ranks in Numbers 2 and 3 which statistically tied, it was understandable because the period of the surveys were different unlike those of the two surveys of SWS that were done in the same period.
If P.S can have two surveys conducted in the same period with almost the same result, why did SWS come out with two polls conducted in the same period with different results?
Did SWS flirt with its clients?
(You can read my selected columns at http://mortzortigoza.blogspot.com and articles at Pangasinan News Aro. You can send comments too at firstname.lastname@example.org)