A poll predicting the winner of this year's presidential election conducted by USC Dornsife and the Los Angeles Times is getting criticism for its unusual approach and what has been an unusual result: consistently showing Donald Trump in the lead while most other polls show him losing to Hillary Clinton.
The Daybreak Poll is a probability survey whose methodology aims to provide a best estimate of how Americans plan to vote in the November election based on the level of voters' enthusiasm toward a candidate. It is one of only a few such daily probability polls that exist in the country.
One reason why it has received backlash is because there is so much weight being put on one individual – a 19-year-old, black, Illinois, male Trump supporter. According to an article in the New York Times, "he's weighted as much as 30 times more than the average respondent, and as much as 300 times more than the least-weighted respondent."
Dan Schnur, director of USC's Jesse M. Unruh Institute of Politics, described the Dornsife/LA Times poll as "an experiment in polling the electorate in a different way than most traditional surveys do." Unlike most polls it surveys the same 3,200 people over and over again. Everyday more than 400 people of that group are asked to quantify their likelihood to vote, their support for a specific candidate and their guess of which candidate will win.
Since the poll began in July, Trump led Clinton for the most part. Clinton took the lead briefly right after their second debate on Oct. 9. As of Sunday night, the poll has Trump ahead at 44.8 percent and Clinton 43.7 percent.
The LA Times responded to the New York Times' article, defending the methodology behind the poll and explaining that at times, weighting in favor of a certain group to make sure the poll sample is representative of the diversity of the population, especially a minority group, is necessary in order to get a point across – sometimes it works and sometimes it doesn't.
"Typically, they almost never come up with a sample that is truly representative right off the bat, so almost every poll has to do some kind of weighting," said David Lauter, Washington Bureau chief for the LA Times.
"You have to increase the weight for those voters (typically small demographic groups) in order to have a representative sample, but if you increase the weight too much, you run the risk that it could have one or two individuals that are going to skew your poll. There is no perfect answer for it. What you have to do is figure out a methodology that has good grounding to it that makes sense and then be transparent about it," he said.
Lauter believes that the New York Times calling out the poll was fair but "thought the headline was a little misleading because the one 19-year-old, black, male Trump supporter that they refer to hasn't actually participated for a while – he alone isn't having so much impact on the poll."
David Andersen, assistant professor of political science at Iowa State University, is not necessarily sure whether the methodology behind this poll was the best, but says it is making a name for itself because of the fact that it is so different from any other polls that have been conducted.
"They kind of have an innovative take on methodology and the only way we are going to know whether it's accurate is on election day," he said.
Anderson believes that measuring enthusiasm is good, but he doesn't know how accurate the poll will be and believes it will lose credibility if it turns out that it's wrong on election day.
USC freshman Rachel Dabora feels like the media has had too much influence on the election. She agrees with Andersen and says that because of the media's involvement, she doesn't "really seem to notice the predictions or necessarily trust that they're accurate."
Reach Staff Reporter Juana Lopez here, or follow her on twitter here.
Annenberg Media