A 3% margin of error at a 90% level of confidence requires a sample size of about 750. For a 95% level of confidence, the sample size would be about 1,000.

One quick note: Most polls report margins of error alongside another technical term, “ confidence interval.” In the most rigorous reporting of polls, you might see a sentence near the end that says ...

Virtually always the reported margin of error assumes the maximum MOE for that sample size, ie, p = 0.5. It also nearly always assumes that a 95% confidence interval is being used.

Hold your breath and make sure you vote, because the result is likely to be a squeaker.

President Trump says negative polls are fake. They're not, but here's how to be an informed reader of polling.

And since pollsters generally use a confidence interval of 95 percent, one out of 20 polls will be a statistical outlier – that is, outside the margin of error – just by chance.

An election poll's point estimate is stabbing for the right answer, but the margin of error usually includes the true figure. Sometimes even that can be wrong.