The SWS-Bilang Pilipino Mobile “Surveys” are so seriously flawed that their results are highly questionable. Their operators can easily manipulate them to portray any candidate as the frontrunner. They should be stopped immediately.
An insult to us Filipinos, the subsidiaries of Philippine Long Distance Telephone Co. (PLDT) – controlled by Indonesian tycoon Anthoni Salim – solely run the technical operations of this “survey,” and would, therefore, be in a position to manipulate the project’s results if they chose to do so. This could mean that for the first time ever in our history – except perhaps the American CIA intervention in the 1950s – a foreign entity, an Indonesian, could play a big, even crucial, role in our presidential elections.
Voting preference surveys in this country have become crucial not really for convincing weak-minded voters to go for the poll leaders. Huge amounts of funds – at least P5 billion for presidential contest – are required for the expensive print and media ads and for the all-important mobilization of a grass-roots machinery to cover the archipelago before and especially on election day. Without a real political party system, candidates have to rely on donations by tycoons and businessmen.
The tyranny of polls in this country is that businessmen, being opportunistic, rely on polls to determine which candidate they should support financially. Many even allocate their contributions in accordance with each candidate’s percentage ratings in such polls.
I put the word “survey” in quotation marks when referring to this SWS project, as it is not really an opinion poll as we know it, but involves getting the views of the same permanent panel of about 750 people.
Yet, a big part of the “survey”’s deception is in its name itself, as most people could be misled to think it is the same kind of surveys SWS and other pollsters have been doing, which is to pick randomly – for each survey – respondents, from 1,200 to 3,000 people, weighted as to their geographical residence to reflect the population in the case of a national survey.
The SWS mobile “survey,” though, is totally different from legitimate polls as we know them.
Instead of taking a random sample for each survey run, or those taken every week, the SWS mobile “survey” organized a panel of 1,200 people, who were given cellphones as an incentive to participate. The cellphone service is free for use in the duration of the project and would be owned by the users after the elections. With its response rate, according to SWS President Mahar Mangahas, declining from the time it started to just 63 percent, the panel consists of just 750 people. At most it is an oversized “focus group,” whose views are collected – cheaply – through SMS messaging.
Views of same people
What this SWS project has done and will be reporting are the views of the same 750 people initially picked, with a few changes as one panelist may not respond in one survey but does so in another. This is not what legitimate opinion polls do, which is to get the views of different sets of 1,200 people for each “run” of a survey. Getting different sets of respondents, in fact, is crucial to a normal poll’s validity, as this corrects the biases of just one sample.
Yet SWS and its media partners, – which are all part of Indonesian tycoon Salim’s media conglomerate in the country – report the “surveys’” results as if they were no different from the usual, legitimate SWS polls. For instance, the Philippine Star’s (the biggest newspaper in Salim’s conglomerate) recent headlines: April 2 – “More than half in Bilang Pilipino SWS poll have ‘much trust’ in Ombudsman;” April 3 – “Poe, Duterte in statistical tie for top spot in Bilang Pilipino SWS poll; April 5: Escudero tops Bilang Pilipino-SWS poll; Marcos least liked VP bet to win.” It is not reported as an SWS Mobile Poll, just an SWS Poll.
The validity of the usual opinion polls is derived from natural sciences’ logic, in which, for instance, a quality-control engineer takes a pint of beer from a huge vat, and, assuming consistency in that vat, concludes that the quality of all the beer in that vat is so-so. However, to make sure that the result of that pint’s examination is not a fluke, the engineer takes different samples and examines them also to come up with a reasonable judgment on the beer quality of the whole production run.
Using this analogy to describe the SWS mobile “survey,” it is the same pint of beer that had been taken once, which the “quality-control” engineer analyzes every day to check the quality of the beer the factory produces. What if, as the SWS mobile “survey” does really, the original pint was taken from some part in the vat that was so different from the rest of it?
I’m quite sure Mangahas knows this is not what “mobile opinion polls” in the US and elsewhere mean, which takes different samples each time a survey is taken, with respondents interviewed not face-to-face by the pollsters, but through cellphones. This actually has become the prominent method in polls in the US and elsewhere because of the decline in the use of fixed-line phones, their former means of communicating with respondents.
Such polls in the US, however, require much expensive manpower as cellphone owners hate getting unsolicited texts or calls, and telephone companies are barred by law from giving out to third parties private cellphone numbers. Hence, in the US, pollsters have to dial more than 20,000 random numbers just to complete a 1,000-person survey. The response rate has also declined steeply, with one leading pollster, Pew Research Center, reporting by 2014 a measly 8 percent response rate.
Mangahas and his partners invented a cheaper form of mobile “surveying” – simply giving a cellphone to a permanent panel of respondents for them to send their responses. It gives us the false impression that SWS is merely doing the kind of mobile-phone polling most often used in the US, but which is actually an entirely different thing.
Not a poll, but a panel
Mangahas’ mobile “survey” is obviously so totally different, and if there were a professional organization of pollsters here, it would likely demand the dropping of the term “survey,” and replacing it with something like “What our especially-picked 750-person panel thinks today.”
There are fatal flaws in Mangahas’ project that will not qualify it to be called legitimately as an “opinion poll:”
First, legitimate opinion polls must take different samples every time a survey is run, to counter the possible bias of one sample. The SWS “survey” has just one permanent sample.
Can we even trust that his panel is representative of 50 million Filipinos, the estimated number of voters? Mangahas claims the SWS picked the panelists randomly. But why should we trust him, when he lends his institution’s imprimatur to what aren’t really “opinion polls” as we know them? Can he provide us with some documentation to show how the panel was selected? Was he manipulated into picking a panel that consists of partisans for a certain candidate?
Somebody, in fact, had sent me a Facebook message April 2: “A friend of mine, a supporter of Poe and a friend of the Aquinos proudly announced to our circle of friends “dalawang beses na ako na survey.”
I thought at that time that this was preposterous since the odds of being picked twice in two surveys are close to the chances of one person winning the lotto two consecutive times. But after studying the SWS Bilang Pilipino Mobile “Survey,” I realized this was because the person, being included in its panel, has been therefore, already surveyed twice.
Second, the reports on the SWS mobile “surveys” do not even explain that only about 60 percent of its panel members have been responding. For instance, in its press release entitled: “Poe, Duterte in statistical tie for top spot in Bilang Pilipino SWS poll, it reported that 34 percent of respondents” preferred Duterte, 31 percent Poe, 17 percent each for Binay and Roxas, and 1 percent, Santiago. That means a total 100 percent response. How does Mangahas reconcile that then with his own disclosure that the panel’s response rate has just been 60 percent?
And third, the panel itself and its results are so vulnerable to manipulation by the entities that run these fake “surveys.” Unlike the case of Smartmatic, Mangahas’ mobile “survey” involves technology and processes that haven’t been made open to scrutiny by any other outside party. What if its technology partners are able to manipulate the way the responses of its 750-man panel are transmitted to the SWS and summed up?
Panel’s views can be manipulated
Not only that. Here’s the big danger.
Only the SWS’ telecom and technology operators know the cellphone numbers of the 1,200 persons picked to make up the panel.
What if they manage to send text messages to this panel denigrating certain candidates and praising the candidates they prefer, disguised as TV5 “newsbreaks.” What if, just before the panel members are asked to respond as to their preferences, leading questions are asked that increase the chances of them picking a preferred candidate? If you believe these are preposterous, read a comprehensive article by Bloomberg on such tech-based black-ops, which it claimed helped Mexico’s president elected in 2012, http://www.bloomberg.com/features/2016-how-to-hack-an-election/.
What are the companies involved in this SWS mobile “survey?” Mangahas disclosed in his column that the project’s software is by Voyager Innovations and the cellphone service provided free by Smart Communications. Its media partners are TV5 Network (including its website interaksyon.com) and the Philippine Star group of publications.
This is what makes this project pretending to be a survey so anomalous. Voyager Innovations and Smart Communications are 100 percent subsidiaries of PLDT. TV5 and Philippine Star are firms that were officially set up by the telecom giant’s Beneficial Trust Fund – the employees’ pension fund, but controlled by its management – and which have received altogether P18 billion in funding from PLDT directly or its units.
And who controls PLDT? Since 1998 when President Estrada helped him take over the telco giant, it is the Indonesian tycoon Anthoni Salim, who is also the controlling stockholder now of the biggest public-utility conglomerate in the Philippines, with the famous Manuel V. Pangilinan actually only as his chief executive in the country with minimal shares and providing a public face to the corporate management. Don’t we have laws that ban a foreigner’s participation in any way in our elections?
Salim’s conglomerate consists almost entirely of public utility firms that are vulnerable to government regulations, and therefore, to that government’s President. This is really the first time ever that one tycoon — and a foreign one — is the major player in strategic public utilities: telecoms (PLDT), power (Meralco), water distribution (Maynilad Water Services), and expressway operations and construction. Would you believe Salim isn’t interested in the outcome of the presidential elections?
I had pointed out this clear and present danger to our sovereignty in several of my columns in the past two years on the Salim conglomerate. Salim now has in place a well-oiled machine for controlling a population’s mind, and even to convince them which candidate to vote as President. First, a content generator made up of his news enterprises in print, broadcast, and the internet; and second, a content disseminator consisting of his cellphone firm Smart Communications, the biggest in the country, and his cable news network. And now another content generator: the SWS Mobile “Survey.”
What has become of our country?
On Wednesday, facts – from the reports of his companies themselves — that Salim controls PLDT and its media conglomerate.