The Social Weather Station warned the public against dubious online election surveys because they do not adhere to international standards on the conduct of surveys.
Online Facebook page Silent No More Ph posted the results of a social media survey supposedly done through Facebook on January 5, 2019 and mockingly labeled it as “Joke of the day.”
The alleged results did not mention who conducted the survey, how many the respondents were, where they were from, and even what questions were asked of the respondents.
Dear Fellow Filipinos,JOKE OF THE DAY! HAHAHAHAHAHAHA!#Joke#NgaNga#AlamNaThis#SilentNoMorePH
Posted by Silent No More PH on Monday, January 7, 2019
The next day, Facebook page Atty. Glenn Chong Supporters, posted a similar survey allegedly conducted on January 2, 2019.
Chong is a former congressman of Biliran province running for a seat in the Senate this 2019 midterm elections.
Pakikalat at report natin. Huwag basta maniniwala sa mga biased surveys.CTTO
Posted by Atty. Glenn Chong Supporters on Saturday, January 5, 2019
This survey is from the Facebook page of CinEmotion Digital Films, a video production firm that has existed since 2014.
SOCIAL MEDIA 2019 SENATORIAL SURVEY TOTAL NUMBERS FROM JANUARY 1-9 2019
Posted by CinEmotion Digital Films_Official on Tuesday, January 8, 2019
The results of these online surveys differ from those conducted by the two top polling firms last December, wherein senators Cynthia Villar and Grace Poe landed in the top two spots of voter preference.
According to journalism organization Vera Files, these surveys do not conform to the mandatory requirements developed by international organizations World Association for Public Opinion Research (WAPOR) and World Association for Social, Opinion and Market Research (ESOMAR). Local polling firm Social Weather Station is a member of WAPOR.
These requirements, also called disclosure requirements, include corresponding values of percent figures and sampling method used to attain the results.
Leo Laroza, communications and information technology director of Social Weather Station, told Vera Files that the absence of these details in surveys is already a red flag.
“Everyone is enjoined to rely only on the SWS website (www.sws.org.ph) for its advisories, announcements, and survey reports released to the public,” it said.
“SWS also recognizes that with the coming May 2019 elections, many surveys on preferences for national and local positions are being conducted by various groups. Some of these groups are legitimate, while others are not,” it added.
Previously, Pulse Asia President Ronald Holmes emphasized that sharing the sentiments of the electorate is part of its service to the public.
Standard in pre-election polls
According to the ESOMAR-WAPOR guidelines, disclosure requirements are important to help journalists and other stakeholders differentiate between professional and unprofessional polls.
“Because all surveys have become more complex and diverse, and the difficulties of conducting polls have grown in recent years, researchers need to provide a higher level of methodological disclosure,” it said.
Some of the requirements include the sampling method used, the name of the organization who conducted it, the participants or respondents, the survey method, the weighting used to adjust the results and the relevant questions asked.
Meanwhile, the American Association for Public Opinion Research or AAPOR explained the two types of sampling methods in political surveys, probability and non-probability sampling, which should be transparent to the public.
“In a probability sample, all persons in the target population have a chance of being selected for the survey sample and we know what that chance is,” it said.
Non-probability sampling is the opposite. “Non-probability sampling methods do not share this feature that everyone in a population has a chance of being selected and researchers know exactly what that chance is,” AAPOR said.
Moreover, professional surveys also normally do not contain typographical or grammatical errors, news website Axios said.
“Check for sloppy mistakes like typos. It likely means the authors of the poll aren’t familiar with the content or weren’t paying attention like a professional would,” Shannon Vavra wrote.
