It seems that the most convenient way to predict results in the politics scenario is to follow polls. But can we actually trust them? Now this has become THE awkward questioning for pollsters.
In a previous post we tried to explain how I should choose the sample size in order to ensure that the obtained result is within a controlled margin of error. Today we are going to deal with a related topic: when I observe a difference in a result obtained through a survey, how can I determine if that difference is statistically significant?