November 2, 2016 by wimvincken
A poll is nothing more than an attempt to predict an outcome based on what a relatively small sample of people say they are going to do. To suggest that they could be off by more than the statistical margin of error is just wishful thinking. Is that true? No. The polling industry is in the problems, because populations are simply getting sick of answering multiple times questions from pollsters. That’s why. So, does that mean that polls are or are not reliable?
The polling industry today is in a serious crisis. Ten of the thirteen polls conducted during the last week before the British referendum on EU membership said “remain” would win. Worse, they showed “remain” winning by up to 10 points, while the three that showed “leave” winning predicted only a 1-2% margin of victory and were online polls (generally considered less reliable). The majority of those polls were wrong. So far the British referendum.
The polls for the Israeli election in 2015 showed Netanyahu’s Likud Party losing to the Labor Party when, in fact, Likud ended up with more seats in the Knesset than anyone expected. All of the polls were wrong.
And while polls predicted that Scotland would vote to stay in the UK, they were wrong with the size of the victory by about six percentage points.
And now we are reaching the American elections and the role of the polling industry in this huge event. What are the problems the polling industry is having with the American elections?
- There are not enough people who want to answer all those endless questions from pollsters (on average, pollsters have to contact ten people to get one response (non-response bias)).
- With the smartphones, people can see exactly who’s calling and they refuse to answer calls from pollsters.
This is a big problem, especially right before the elections, where there is not much time.
- Pollsters are using smaller samples, because not many people respond to their calls or questions.
- Applying of pre-cooking percentages of Democrats, Republicans, and Independents, as it’s known what average percentages are for those three political affiliations are.
- Pre-cooking the percentages based on political affiliation may or likely not working for controversial candidates.
- Pre-cooking the polls rely heavily on assumptions, not actual facts or factual observations or actually on any reliable gathering of opinions.
And then there are those polls, who are plainly cheating. For example, a pollster is interviewing 345 Republicans, 880 Democrats and 23 Independents and guess who’s the most favorite from that poll? Or another one to measure how popular or unpopular Trump and Clinton is. They interview 309 Republicans, 389 Democrats and 276 Independents, 23% male and 77% females. Everyone knows that Hillary Clinton is popular with women, so we know what the outcome of this poll is. Those polls are plainly published in the mainstream data and nobody says a word, except Donald Trump, who’s protesting, but ignored. At the end he calls that rigged polls and he’s right.
I personally call this suicide of the polling industry and the mainstream media.