Informational Effects on Predicting Election Results
Before Election Day 2016, the Independent published a piece about pre-election polls. Nearly every single poll indicated that Hillary was going to win the presidency (by around 2-4 percentage points), and a select few polls showed Trump winning with 5 percentage points. These polls, however, were almost unanimously overlooked by the public. Even the LA Times rejected its own poll results that predicted a Trump win. The Independent hypothesized that many of these papers and pollsters were skewing their own data to fit in with the rest of pro-Hillary polls.
In class, we talked about how informational effects cause people to imitate what others are doing (Chapter 16). Prior to this election, pollsters were getting information from other reputable polls that were predicting a Hillary win. Informational effects were especially impactful during this election due to extra pressure for polls to yield correct predictions; this is because earlier in the election cycle, many poll predictions proved to be incorrect.
Now that the election is over, we can see that the outlier polls who were not as affected by informational effects (like the LA Times) made the correct prediction in the end. Trump won the election. This is particularly interesting because we actually discussed in class about how informational effects could lead to wrong guesses!
Reference: http://www.independent.co.uk/news/world/americas/us-elections/us-election-2016-latest-polls-hillary-clinton-donald-trump-who-will-win-accuracy-a7402846.html