While Americans wait election resultsMany look to surveys to find answers.

Dr. from the University of New Hampshire Research Center. Andy Smith joined FOX’s LiveNOW on election day to give tips on deciphering the polls and how to better understand them.

What is the margin of error?

The margin of error describes how close a survey result can reasonably be to the true representative of the entire population. In other words, even the purest random sample of pollsters will not be an exact match of the entire American population.

Smith tells us that calculating the margin of error works through a complex mathematical theory known as the central limit theorem, but he gives us some clues to make it easier to understand.

He said the biggest confusion about sampling error is that people think it applies to a percentage difference between candidates, but that’s not the case.

“What you need to do is apply a margin of sampling error to both estimates,” he said. “So Donald Trump at 46%, in this example (with a margin error of 3%), could be as high as 49%; could be as low as 43%. Harris at 50% could be as low as 47%, or as much as 53%.” high.”

“If you want to understand whether the difference between these two candidates is statistically significant, you really have to double the margin of sampling error,” he added.

RELATING TO: Presidential election polls: Who’s ahead now?

Telephone and digital surveys

FILE – Close-up photo of a man holding a mobile phone.

“The voting industry is going through what I call a paradigm shift; we’re moving away from the types of methods that worked 30, 40 years ago that really don’t work anymore,” Smith said.

He said response rates to phone calls have dropped significantly in recent years, with only 5% of people contacted for phone surveys completing them.

“This results in very, very expensive phone surveys and (as we saw) inaccurate phone surveys in 2016,” he said.

He said more polls were conducted digitally this election, but there are also several different methodologies available for digital methods.

“After this election, we will be doing quite a bit of research to understand which of the methodologies used perform best and use the information we gain going forward,” he said.

“Right now I’m very cautious about saying one survey is better than another survey because frankly we don’t know. We’re in the middle of developing our best practices and we’re not at the end yet.”

RELATING TO: When will we know the results in swing situations?

Contrarian polls

Every election cycle, a handful of polls publish shocking results that attract attention.

Latest The highly anticipated survey From Iowa “gold standard” pollster J. Ann Selzer shows Kamala Harris beating Donald Trump by 3 points in the Hawkeye State.

READ MORE: New Iowa poll shows Harris up 3 points in Hawkeye State

So how can some survey results be so different?

Smith says this may have to do with how the survey was conducted or that an unusual sampling group was used that may have excluded a certain demographic.