Saturday, June 15, 2013

Are Thompson And Liu Being Undercounted In The Polling?

Errol Louis thinks so:

Rasmussen conducts its surveys by landline phones, which almost guarantees an undercount of black and Latino voters: nationally, according to federal government figures, 51% of people living in poverty are in cell phone-only households, as are 43% of Latino adults and 37% of black adults.
By contrast, only 29% of white adults don’t have a landline phone. So Rasmussen polls may well miss trends among low-income, black and Latino voters.

Something similar is going on with Asian voters, by the way.

Strategists in city Controller John Liu’s campaign raise the question: What do telephone pollsters do when someone answers the phone in Mandarin? (Answer: They probably just hang up, which might distort and undercount Liu’s support.)

Louis argues that black support of Thompson is also being undercounted because many black voters make up their minds very late in a campaign.

He says there's no way Quinn or Weiner will garner more black support than Thompson.

If anything, Louis writes, Liu, a "constant presence" in black communities, has a better chance to take black votes away from Thompson than either Weiner or Quinn do.

Interesting thesis and something to think about before the UFT announces its endorsement this coming week.

They're looking to ride a winner (having failed to pick one since Dinkins.)

Unless something big happens, it's hard to see de Blasio winning the race.

But if Thompson is being undercounted by pollsters, it is conceivable that a late shift in black voters to Thompson, along with erosion of support for Weiner as more and more voters get sick of his act, could give Thompson a shot for the runoff with Quinn.

The UFT is rumored to be debating between  endorsing de Blasio or endorsing Thompson.

I wrote a while back that they would most likely endorse Thompson.

I still think that's the case.

1 comment:

  1. Why not endorse Liu? Take a chance on someone who might actually support us if elected.