[Note – I have included some translations because I don’t want to be misunderstood or dismissed as an elitist using data.]
Most citizens have moved on beyond the issue of the election polls. If pollsters really failed in this case in contrast to their accuracy in such a high majority of previous cases, those who work in such fields really need to understand, consider, and adjust.
Here is why? Republican pundits have begun explaining away the abysmal Trump approval ratings as due to the same failed pollster performance that resulted in the inaccurate election prediction.
Were the pollsters wrong? Well, I remember the final prediction was that Hillary would win by 4%. She won by 2%. The statisticians would say this was an accurate prediction as the difference was within the margin of error [translation – this is like horse shoes and close does count]. It gets far more complicated when it comes to combining the predictions from multiple states.
Whatever. Still, the overall pollster prediction was accurate. Hence, claims that an overall assessment of Trump performance as the lowest for any new president is somehow inaccurate because the previous election prediction was inaccurate is simply not true (translate – it is a lie). Neither predicted election vote totals nor approval ratings were measured inaccurately.
Translation – The people have concluded that Trump sucks no matter how you try to spin it.