The polls for the 2016 election were showing Hillary Clinton winning the presidential election. Election polls rely on people giving honest answers. In this case we had many individuals for whatever reason not wanting to let pollsters know that they were voting for Donald Trump, so they said they were voting for Hillary Clinton. Other research showed that a relatively small polling bias which saw Republicans underrepresented in several key states which caused the polls to tip in Hillary Clinton favor, predicting a Hillary Clinton win. Before the election Hillary Clinton had a 70 percent probability to win the 2016 presidential election. This prediction turned out to be incorrect because of a real change in voters’ preferences just before the election, an overrepresentation of college graduates in some poll samples, and late-revealing Trump voters. As the election night went on, the numbers turned in Donald Trump’s favor and was elected the 45th president. We see polls like this on many bills, likability of congress and senate members and the how a mayor, governor or president is doing with the according to those who answered the questioners when called. This is improper sample of individual because not everyone is polled, only those who are called and have answered. Yes, you are getting a percentage but not from the entire voting population, just a small sample.