US

Why the LA Times poll differs from the others

Most polls show Donald Trump losing big time, especially in the crucial battleground states. But the L.A. Times poll shows Trump ahead of Clinton by two points, a fact heralded by the Trump campaign. So why are that poll’s results so different?

Most polls ask people who they are going to vote for. The L.A. Times poll, designed by USC social scientists, is not so straightforward. It asks a pre-selected group, used for other research purposes, to rate on a scale from 0-100 their chances of voting for a particular candidate. Then the results are weighted for demographics, which is usual, but then also weighted for how the respondents voted in the 2012 election.

The Times admits that the result is that Republicans are probably over-represented. I would add that the 0-100 scale isn’t going to tell us much if respondents are ambivalent about both candidates. But I think this is also an example of social scientists overthinking their task and trying to come up with a methodology that is so sophisticated that it is unlikely to work. But maybe it will. We’ll know in November.

From Why the USC/L.A. Times tracking poll differs from other surveys – LA Times:

Typically, polls ask people which candidate they favor or lean toward. Those who say they don’t know or are undecided don’t get factored into calculations of candidate support.

The Daybreak poll, by contrast, asks voters, using a 0-to-100 scale, to rate their chances of voting for Clinton, for Trump or for some other candidate. As a result, everyone who responds to the survey has some impact on the results. Because that approach gathers information from everyone in the poll sample, it should give a better read on the many voters who remain ambivalent about their choices.

Using the 0-to-100 scale, however, almost certainly makes the Daybreak poll differ somewhat from other surveys. As with the bounce, any difference that results should shrink as election day gets closer and voters become more certain of their choices.

Finally, some analysts think the Daybreak poll is slightly tilted toward the Republican side because of how it accounts for the way people voted in the last election.

All pollsters weight their results somewhat to make sure their samples match known demographics — the right proportions of men and women, for example, or blacks, whites and Latinos.

The Daybreak poll goes a step further and weights the sample to account for how people say they voted in 2012: It’s set so that 25% of the sample are voters who say they cast a ballot for Mitt Romney and 27% for President Obama. The rest are either too young to have voted four years ago or say they didn’t vote.

The potential problem is that people tend to fib about how they voted. Polls have often found that the percentage of people who say after an election that they voted for the winner exceeds the winner’s actual vote.

If that’s the case this year, then weighting for the vote history would result in slightly too many Republican voters in the sample, which would probably boost Trump’s standing by a point or two.

[Keep reading. . .]

Original Article

Post Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.