The Flathead Valley’s Leading Independent Journal of Observation, Analysis, & Opinion. © James R. Conner.

 

2 May 2012

kg_mug

Is Kim Gillan really this popular with men?

Is Kim Gillan really ten percentage points more popular with men than women?

That’s what the latest PPP survey of Montana U.S. Senate and House races reports. But is that differential believable? Why would any woman running in a Democratic primary do so much better among men than women?

Here’s PPP’s cross tabulation on gender:

PPP crosstab table

kim_and_men

Here are the same data presented as a horizontal bar graph. Positive numbers indicate more support from men than from women. Negative numbers indicate more support from women.

Percent men supporting
minus percent women supporting

house_gender_gap_bar
PDF for printing.

And here’s a scatter plot showing how much of an outlier Gillan’s reported support among men seems to be:

Kim the outlier scatter plot

kim_scatter
PDF for printing.

I emailed PPP asking for clarification, but I’m not altogether certain I’ll even get a reply, let alone a prompt reply. Meanwhile, these seem to be the possibilities:

  1. The 27 percent figure is what the survey found, and the survey got it right.
  2. The 27 percent figure is what the survey found, but the survey got it wrong.
  3. The 27 percent figure is a typo; the true value should be 17 percent.
  4. The 27 percent figure is a typo; the true value is neither 27 nor 17 percent.

I lean toward the third possibility, but that’s a hunch based on knowing that more women than men are Democrats. It’s not a conclusion for which I have convincing evidence.

Still, one wonders. Diane Smith does much better among women than men, owing I suspect to admiration for her success in business. Franke Wilmer, who’s overtly courting an identity vote by women, actually does a point better among men, but that’s virtually a tie and one shouldn’t make too much of the difference.

Given the poll’s small sample, there’s a fair amount of uncertainty in the numbers reported. That’s not an argument for dismissing the numbers as worthless, but it is an argument for exercising caution when drawing conclusions.