In his column, The Ateneo and public opinion polling , Mahar Mangahas kept referring to “the spiral of silence phenomenon,” without every actually explaining what it is:

In planning for the first Ateneo-SWS poll, of May 1986, the issue was raised on whether to risk asking whom the respondents voted for in the snap election – suppose most said they had voted for Marcos? I appreciate the Ateneo for agreeing to ask it; the result was that 64 percent said they had voted for Cory Aquino. (The risk was actually small; we didn’t know about the “spiral of silence” phenomenon yet.)

In early 1987, we had to decide whether to do an Ateneo-SWS poll just before the May election, to maximize its potential to predict the outcome, or much earlier, to enhance its value to campaigners. We took the second option. The March 1987 poll found only half of Cory’s senatorial candidates in the winning column; her campaign manager Paul Aquino told his staff that they could not afford to sleep any more. Eventually, with the help of “Cory magic,” 22 of her 24 candidates won. (But critics claimed that the survey failed, because the election outcome was different.)

After the joint project expired, Ateneo and SWS shared the briefing revenues 50-50 as pre-arranged, and then did polls separately. With funding from various foundations, Ateneo did at least six national polls over 1988-1992. In 1992, its post-election poll found some 40 percent saying they had voted for Fidel Ramos, even though he had won with only some 25 percent of the official count – but it was again the “spiral of silence” at work.

Here’s a handy-dandy definition,

The spiral of silence is a political science and mass communication theory propounded by the German political scientist Elisabeth Noelle-Neumann. The theory asserts that a person is less likely to voice an opinion on a topic if one feels that one is in the minority for fear of reprisal or isolation from the majority.

In a speech he made on May 18, 2000 (the keynote address during the Annual Conference of the World Association for Public Opinion Research, 17 - 21 May 2000, Portland, Oregon, USA), he tackled how not knowing “the spiral of silence” made some survey results curious:

Some of you, perhaps, may recall that Ferdinand Marcos was virtual dictator in the Philippines from September 1972 to February 1986. Opinion polling was very uncommon then. For instance, in 1983 a book of survey indicators, The 1982 Philippine Social Weather Report, by myself and others, was suppressed from publication. In November 1985, when Marcos unexpectedly announced over American TV (the David Brinkley show) that he would hold a so-called ’snap’ presidential election, he waved on-camera a national opinion poll, done the previous July and publicly reported in August, by the Bishops-Businessmens’ Conference (BBC 1998), an independent civic group, as his basis for expecting to win. He was alluding to a survey item that asked, “How many in this locality would vote for Ferdinand Marcos if he runs for President again?”, to which 53% answered Many or Very Many, and 37% answered Few or Very Few.

No amount of clarification could persuade Marcos loyalists, and even some anti-Marcos elements (to my frustration as the BBC survey director), that the score of 53-37 was NOT a prediction of the vote for Marcos versus whoever. Perhaps Fate decreed that this portion of the poll be misinterpreted so much. More significant survey findings, such as the opposition to legislation by presidential decree, and opposition to detention of persons by presidential fiat, both by 2-to-1, were ignored by the Marcos-controlled media.

Three weeks before the snap election on February 7th, a professional poll commissioned by the TV networks showed a score of 45% for Marcos, 26% for Corazon Aquino, and 29% undecided. In the final week, a poll by Asia Research Organization (Henares 1991), affiliated to Gallup International, found 42% for Aquino and 41% for Marcos, and assigned the 17% undecided to Aquino on account of the fear-factor; but this was not revealed by ARO for 5 years, and the sponsor is still unknown today. In the quick-count of the vote by the National Movement for Free Elections, the winner was Aquino, by 53% to 47%, while in the slow-count by the National Legislature the winner was Marcos, by 54% to 46%. The issue was politically settled by the People Power Revolution and the Marcoses’ flight to Hawaii on February 25. The following May, a joint survey by Social Weather Stations and Ateneo de Manila University asked respondents — after some discussion of the merits of ‘letting well enough alone’ — for whom they had voted in the snap election, obtaining 64% for Aquino, 27% for Marcos, and 9% refusals (Ateneo and SWS, 1986). At that time none of us knew of The Spiral of Silence yet.

Does it exist, now, and is it reflected in the “undecided” in survey results today? And if so, who do the “undecideds” fear? My view is, they reflect tacit but not explicit support for the administration -the fear, in this case, being fear of the majority that opposes the administration and castigates its public defenders.