Hutchinson (’06) Performance Strong Predictor of Barkley Countywide Support in Minnesota Senate Race

As documented by Smart Politics last week, Independence Party candidate Dean Barkley took a very different path to reach 15 percent in the 2008 Minnesota U.S. Senate race than did 2002 Independence Party gubernatorial nominee, Tim Penny (who reached 16 percent in that race).

Penny’s support was heavily skewed towards his old 1st Congressional District stomping ground, but generally performed fairly weak elsewhere. Barkley, unlike Penny, failed to finish second place in even a single county (Penny won 7 and placed second in 8 more), but his support was more evenly distributed throughout the state.

The Penny campaign was thus not a good template to project how Barkley’s support was to be allocated from county to county across Minnesota in the 2008 Senate election. However, the campaign of 2006 Independence Party gubernatorial nominee Peter Hutchinson, did provide such a blueprint.

Hutchinson, who finished a distant third in the 2006 race, ended up with just 6.4 percent of the vote, without reaching double-digits in any county. However, a Smart Politics analysis found countywide votes for Hutchinson and Barkley to be positively correlated (.573, significant a the .01 level). In sum, higher percentages of support for Hutchinson in 2006 are associated with higher percentages of support for Barkley in 2008. (And lower percentages with lower percentages).

But we can glean more from Hutchinson’s performance in 2006. Smart Politics conducted a linear regression analysis with 2006 gubernatorial countywide vote percentages for Hutchinson as the independent variable and 2008 senatorial countywide vote percentages for Barkley as the dependent variable. The model finds that 30 percent of the variation (R Square = .295) in the vote for Barkley is explained by the 2006 Hutchinson vote, with every 1 percent earned in a county by Hutchinson begetting a 1.6 percent vote in 2008 for Barkley (the model is very significant, at the .001 level).

In other words, putting aside all other reasons for vote choice (candidate travel to counties, candidate resource availability, candidate campaign advertising, candidate personality traits, candidate name recognition etc.), and just looking at the Independence Party gubernatorial vote in 2006, we have a very strong predictor to estimate the county-to-county distribution of Barkley’s countywide vote in 2008.

For example, consider the performances of Hutchinson and Barkley in Pipestone (3 and 9 percent respectively), Rouseau (3 and 10 percent), and Houston (4 and 9 percent) counties. All three counties were among the four worst for each candidate across the state.

On the flip side, three of top four counties with Hutchinson’s best performances (Waseca, 9 percent; Renville, 8 percent; Meeker, 8 percent) were three of the top four counties for Barkley as well (Waseca, 23 percent; Renville, 23 percent; Meeker 22 percent).

Barkley’s much stronger performance than Hutchinson was no doubt in some part attributed to the greater name recognition and favorability rating he enjoyed. For example, the (final) October 28th Rasmussen poll of 500 likely voters found only 15 percent of Minnesotans had no opinion of Barkley, with 53 percent having a favorable view of the IP nominee. By contrast, a November 1, 2006 Rasmussen poll found 25 percent of Minnesotans had no opinion of Peter Hutchinson, with just 41 percent viewing him favorably.

Barkley also likely benefited by some measure due to the fact that Norm Coleman and Al Franken collectively had a higher unfavorability rating (48 and 51 percent respectively, according to the late October 2008 Rasmussen poll) than compared to Tim Pawlenty and Mike Hatch (47 and 45 percent respectively, according to the November 2006 Rasmussen survey). Barkley became an attractive third choice to a slightly larger pool of voters who didn’t like either the DFL or GOP nominee.

Countywide Vote for Penny (’02), Hutchinson (’06), and Barkley (’08)