Interpret polls responsibly
When National Bureau of Investigation Director Melvin Matibag said, as an impeachment trial witness the other day, that an opinion poll is “only a snapshot,” he surely meant that it was applicable to the time when it was fielded. In the case of the Social Weather Stations (SWS) survey for the second quarter this year, the fieldwork was done last June 20-29, i.e., before the trial began on July 3. Thus, nothing that has happened at the trial could have affected the outcome of that survey.
The latest SWS media release was about some other snapshots taken in the June 2026 survey: “Social Weather Report: 74% of adult Filipinos knew about the impeachment case vs Vice Pres. Sara Duterte even before the survey; 70% said she must address the impeachment charges,” www.sws.org.ph, 7/23/26. The findings mentioned in the title were elicited by questions prefaced by a long description of how the impeachment process got derailed in 2025, but was renewed in 2026, with the number of charges raised from three to four.
Then the survey respondents were asked (a) if they knew already (i.e., before being interviewed) of the forthcoming impeachment trial, and (b) if they agreed or disagreed that “kailangang harapin ni Bise Presidente Sara Duterte ang kasong impeachment upang masagot at maipaliwanag niya ang lahat ng paratang laban sa kanya.” (The full questionnaire, in Filipino and English, is in the SWS webpage; it’s too lengthy for ordinary newspaper stories.)
The finding for question (b) begs another question of whether, in the people’s minds, the Vice President is presently addressing the charges sufficiently if she continues to be absent from the trial hearings. Thus, there were many snapshots, not just one, taken last June. They show various aspects of public opinion at that point, and not the changes in public opinion over time. The latter would require not just one poll but a series of several polls.
The sentiments expressed in June 2026 on knowledge of the charges and on Duterte’s needing to face them are not new. SWS did similar probes in June 2025, September 2025, and March 2026, with similar findings.
Weeks ago, I explained that SWS has the financial capacity to do four surveys per year on its own account, or one per quarter (see “No polls on the Senate yet,” 6/6/26). The SWS first quarter survey in 2026 was fielded on March 24-31. Every report states the exact fieldwork period, which is normally in the last week of a quarter, except in the fourth quarter when it’s important to finish before the holidays start. It will be in its third quarter survey round, set for fielding in September, that SWS will be able to discern the effects of the trial itself on public opinion. By then, the trial may be over. Or maybe not; let us see.
Public satisfaction with how a specific official does her/his job is distinct from public trust in that person as an individual. The SWS satisfaction ratings of specific high officials at their jobs are distinct from its trust ratings of those persons as individuals.
“Satisfaction” refers to nasisiyahan sa paggawa ni XXX (name) ng kanyang tungkulin bilang YYY (position). “Trust,” on the other hand, refers to pagtitiwala kay ZZZ (name) without mentioning the person’s status in any way. (Still another rating is that of “approval” [sang-ayon in Filipino], which is common in the United States but not used by SWS, as we already began using “satisfaction” 40 years ago.)
The ratings of trust and satisfaction are correlated but not identical. SWS routinely rates public satisfaction with a high official’s job performance; it can rate public trust in that person if commissioned to do so—as in “SWS confirms June 2026 survey items for Stratbase Consultancy on public trust in Ferdinand Marcos Jr. and Sara Duterte” (www.sws.org.ph, 7/20/26), where the questionnaire item is silent as to these persons’ positions. The satisfaction item and the trust item are sequenced far away from each other in the questionnaire.
To me, it is proper to compare the satisfaction rating of a current president or vice president to the satisfaction ratings of previous presidents and vice presidents, at similar points in their terms. For both officials, see the SWS tables and time charts across all past administrations, starting from Cory Aquino, which clearly show their honeymoon periods.
Survey competition is welcome. There are more data for the public to see when multiple, independent pollsters take up a given topic and publish the results. As a market-oriented economist, that’s fine with me. The findings of scientific polls will converge toward each other.
SWS surveys are NEVER manipulated. The one bit of Matibag’s testimony that was uncalled for was his remark that surveys can be manipulated. Of course, one should be wary about fakes; one should know how to spot them, so as not to buy them. Use the market-test: bad quality gets driven out by good quality. Cheaters in research don’t get repeat-work. Rely on the tried and true.
—————–
mahar.mangahas@sws.org.ph.
******
Get real-time news updates: inqnews.net/inqviber
