Pollsters call it the “margin of error,” that small but statistically important percentage that indicates how reliable a survey’s results really are.
The smaller the number, the more likely it is that the poll represents the true feelings of the entire population.
During elections, “poll fatigue” quickly sets in.
It’s often more educational to look at how the media chooses to report on certain polls, rather than focus on the results alone.
A new survey based on a hypothetical race between Hillary Clinton and Donald Trump is a good example.
According to NBCNews.com:
In this week’s poll, Americans are nearly split between their choice of Trump or Clinton; her margin over Trump narrows from 5 points last week to 3 points this week to 48 percent to 45 percent.
The NBCNews.com headline reads, “Hillary Clinton Holds Slim National Lead Over Donald Trump: Poll.” Which is true. But the accompanying graphic reveals that while Clinton has dropped one percentage point since the previous week, Trump’s support has increased by the same amount.
So it would have been just as accurate, if not as politically correct, to report: “Hillary Clinton’s Support Drops As Donald Trump Continues to Rise.”
For months, Americans have been told again and again that Donald Trump had no chance of defeating Hillary Clinton “in the general.”
This poll, with its notably low 1.2% margin of error, is just another example of how wrong received wisdom has turned out to be in this election cycle.