This one has been an outlier throughout mot of the electoral season, and now is alone among national polls in showing Trump ahead of Clinton by five points or more. It is clear the race has tightened, ever since the new campaign managers got Trump to shut up and read from a teleprompter more often. While some reputable national polls have shown Trump ahead of Clinton by one point, most still show Clinton ahead nationally by 2 points or more. So what's going on with the Los Angeles outlier? It's a very unusual poll, surveying the same 3000+ people throughout the electoral cycle. But how were they chosen? Here's where the trouble starts.
The poll "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 problem is that people frequently misreport who they voted for in a prior election--those who voted for the loser often report they voted for the winner, or reported not voting at all. That means, of course, that Republican voters are overrepresented in this sample (perhaps dramatically over-represented), hence the screwball results. For more details, see the NYT discussion.
Recent Comments