Tuesday, July 07, 2020

Popular elections and danger of opinion polls

Nov 11. 2016
Facebook Twitter

By The Nation

Political surveys can be counter-productive to democracy

After the US presidential election, many analyses were focused on why the opinion polls were utterly wrong in predicting the eventual winner. There was a lot less talk about whether the polls were relevant, and absolutely no talk on whether publicising polled decisions on such an important matter is a good or bad thing.

America is a polling country, with dozens of credible groups conducting their own surveys on the showdown between Donald Trump and Hillary Clinton. In addition, there were “polls of polls”, which were analyses or summaries of polls conducted by others. Needless to say, even the “polls of polls” failed to forecast that the next president of the United States would be Donald Trump.

As usual, when the polls were blatantly snubbed by the actual results, people scrambled to find reasons. Trump’s supporters were probably too shy to state their political preference on the phone, some say. Women might have hesitated to tell pollsters they were going to vote for him. 

Moreover, pollsters were probably ignorant of the fact that the electoral-college system is a winner-takes-all arrangement that can actually help the loser of the popular vote win the presidency. This observation was boosted by the fact that, while Trump ran away with electoral-college “votes” by winning in big, key states, Clinton was actually ahead in the popular vote. However, the real issue is not whether the opinion polls are accurate, but whether they should be allowed at all, due to their possible influences on the electorate’s decisions. In countries where voters are responsible and always exercise their right, opinion polls might not have a lot of impact, but in countries were absence from balloting is normal, publicised opinion polls could really swing the results one way or the other.

Simply put, in the US, where the public generally feels less compelled to vote, opinion polls that were overwhelmingly in Clinton’s favour could have convinced many to stay home, thinking she was sure to win. When that happens, the margin can be drastically narrowed or enlarged, and even ultimately belie the reality.

If democracy is really about allowing every person to make his or her own judgement, opinion polls help little. They only carry an “excitement” value. Knowing what the majority thinks might be good on certain occasions, but it can be counter-productive where democracy is concerned. If someone was adamant about voting for Clinton or Trump, he or she didn’t need to know what the majority felt. A separate issue altogether is media endorsement of political candidates. The practice makes sense, since it lets the public know what “informed” elements of society think. Editorial endorsement is controversial only when the media groups making them are too close to those they support. When media outlets make endorsement sincerely based on what they know, they help society with their useful input. Unlike opinion polls, media endorsement is accompanied by argument for or against a candidate.

The US election results, however, showed that editorial endorsements can be grossly defied. Trump won the presidency despite a majority of media groups and television networks recommending that their readers and viewers vote for Clinton. Combine the media endorsements with the opinion polls and Trump’s victory becomes an even more astonishing phenomenon.

American democracy will lumber on. Argument for a ban on pre-vote surveys will again be met with an outcry. Trump has exposed the flaws in opinion polls, but his election is unlikely to stop them or trigger a real debate about their pros and cons.

Facebook Twitter
More in Opinion
Editor’s Picks
Top News