Kalye surveys

With every kulas and kulasa conducting political surveys these days, statisticians are coming out to enlighten the public on the science of opinion research.

Professor Francisco de los Reyes, former acting dean of the University of the Philippines School of Statistics, raised the following points:

Reputable surveys must not have respondents who are randomly picked on the street; the residence address is critical for proper intervals in the sampling that should be representative of the different areas and income classes covered.

The interviewer must be as neutral as possible when posing questions. This requires special training. The face must be professionally impassive – no frowning or smirking in reaction to certain answers, or to try to influence the respondent. Even the voice and manner of asking questions must not trigger any particular feeling from the respondent, such as intimidation, fear, embarrassment about being clueless about issues, or a feeling that the respondent is expected to answer in a particular way.

If the interviewer is the one writing down the answers, the respondent must see the questionnaire at every step of the way, to ensure that the answers are accurately recorded. The interviewer must guarantee the confidentiality of the answers and the data privacy of the respondent.

The interviewer, in reporting to the polling firm the results before publication, must also write down observations about matters such as the demeanor of the respondents while answering as well as the environment during the face-to-face interviews.

Telephone interviews can therefore be problematic when compared with face-to-face questioning, and online polls such as those on Facebook, where identities and demographic information can be impossible to verify, are the least reliable.

Polling is expensive; even an accurately representative nationwide sample size of 1,200 can cost millions. Costs and precision must be balanced in sampling.

*      *      *

In an unprecedented move, faculty of the UP School of Statistics issued a statement of concern last Monday about the “pervasive abuse of survey methodologies.”

The UP statisticians also decried the proliferation of “kalye surveys” carried out with a “cavalier disregard for the principles of data gathering,” and which are meant to sway rather than accurately reflect public opinion.

“We have been observing these practices, including surveys not based on random sampling, such as videos of ‘kalye surveys’ being done by some vloggers and suspicious online surveys, especially on Twitter and Facebook,” the faculty said. “We also witness PR companies, private individuals, and even some media organizations publishing results of surveys with unclear methodologies.”

Supporters of the current survey frontrunners need not fret; De los Reyes told us on One News’ “The Chiefs” last Monday that two pollsters in the country have a track record for credibility – Social Weather Stations Inc. and Pulse Asia. And the latest surveys of the two pollsters showed the UniTeam of Bongbong Marcos and Sara Duterte-Carpio still enjoying a comfortable lead.

All that the UP statisticians want, according to their statement, is to promote “the ethical practice of statistical science,” especially as kalye surveys, De los Reyes said, are becoming part of disinformation.

*      *      *

Polling is an unregulated enterprise in this country, De los Reyes told us. But he said there is the Market Opinion Research Society of the Philippines where concerns about polling can be raised.

He said statisticians are drawing up infographics that will be released on social media and disseminated to mainstream media, which will demand more details from the pollsters that release results related to the May elections.

Among the questions people should ask, as the UP statisticians detailed in their statement: “How was the sample selected? Which sectors of the population, if any, are being represented? What events surround the period of data gathering? When facing an interviewer, was the respondent interviewed in a neutral yet professional tone? What are the control mechanisms implemented to ensure accuracy of the protocols in data collection?”

In my university days I conducted a survey for one of my classes. It was murder designing the survey, and when I finally conducted it, I was fully aware of what an amateur undertaking it was.

My college polling was literally a kalye survey, taken in the streets, with the respondents any adult I ran into who was willing to be disturbed for a few minutes of questioning by a teenager.

But this is exactly what some groups are actually doing in this election season, and then releasing the results of their efforts. Many are blatantly partisan.

“Biased methodologies will only give biased results,” the UP statisticians pointed out.

*      *      *

I have previously written about the important role played by polling in government policy making in a democratic society.

De los Reyes points out that apart from election surveys, professional and accurate polling – such as when conducting a census and similar statistics-gathering activities – is indispensable in efficient development planning.

In the UK, when their surveys botched the results of the 2015 elections, the British Polling Council immediately announced an independent inquiry into what went wrong in the polling.

A detailed report on the post-mortem was released a year later, blaming the error on faulty sampling. The report also released recommendations on how to further improve transparency in the country’s political polling industry, which the Brits say is “streets ahead” of polling transparency in the United States.

Polling plays a critical role in the UK’s policy making, such as its decision to exit from the European Union, so you can understand why there was such concern over the erroneous surveys in their May 2015 elections.

You wonder how the Brits would describe the polling industry in the Philippines.

As in previous elections, many people have formed their opinions about this year’s candidates long before the formal campaign started. But there is still a significant undecided segment of the population that those behind partisan “kalye surveys” could be trying to drag into their bandwagon.

Their spread like toxic mushrooms in this election season should be seen as an opportunity to implement measures that will boost professionalism in the country’s polling industry.

The industry’s long-term viability depends on public trust. In confronting these current developments that could undermine that trust, the industry should emerge stronger.