Tuesday, May 27, 2014

Who is the “most” conservative House candidate? The data--not so much.

Sometimes data illuminates and makes patterns easier to see. Other times, data do not illuminate and instead confuse. When it comes to understanding the ideologies of the Republicans running for the House here in Montana, it seems to obfuscate more than clarify.

Mike Brown over at the Western Word opines thusly about the upcoming Republican House primary:
“Republicans across Montana can be heard saying, “Candidate, candidate on the ballot, who is the purest conservative of them all.”

One of the problems with primary elections is they are, by their very nature, low information environments. This makes it hard for voters to distinguish clearly the ideologies of the candidates. I said in a recent interview with Chuck Johnson that endorsements matter a lot in this type of environment because they provide easy cognitive shortcuts for voters to make decisions.

One possible way to help sort this out is by looking at empirical data. I propose using endorsements and the state NOMINATE values generated by Boris Schorr and Nolan McCarty to put three of the House Republican candidates (Else Arntzen, Matt Rosendale, and Corey Stapleton) on a left-right spectrum to decide who is the “purest” conservative of them all. In an earlier blog, I used these same data to conclude that John Bohlinger’s voting record in the legislature suggests he would be quite a conservative Democrat if elected to the United States Senate.

Corey Stapleton and Elsie Arntzen served in the state legislature before 2011, so they actually have NOMINATE scores based upon their votes. Stapleton had a score of .934 as a representative and 1.032 as a senator—averaging the two scores gives us a .983 for Stapleton. Arntzen had a score of .795. 

Rosendale is not included in the Schorr and McCarty data. I have, however, generated a score from him based upon the 11 legislators who have endorsed him and have NOMINATE scores in these data. Note that Rosendale has more than 50 current and former state legislators who have endorsed him, so this is only a slice of his total endorsement haul. The endorsements were culled from Rosendale’s campaign website. Averaging the scores of these 11 gives us a value of .872.

Again, the higher the value, the more conservative. The mean value of Republican state legislators for the entire period is .976. According to this analysis, that would place Elsie farthest to the left, followed by Rosendale, and then Stapleton—with only Stapleton to the right of the average state legislator.

Finally, there is the case of Ryan Zinke. As far as I can determine, he has no public endorsements from state legislators, current or former. Chuck Johnson of Lee Newspaper reported none, I found none listed on Zinke’s website, and an e-mail to Zinke’s campaign manager requesting a list has received no response. 

So, using this method, I can’t evaluate Zinke’s conservatism relative to Arntzen, Rosendale, and Stapleton.

But I CAN use Common Space scores calculated by Poole andRosenthal to help us pin Zinke down. Zinke received the endorsement of former U.S. Senator Conrad Burns. Rosendale received former Congressman Ron Marlenee’s blessing, and Artnzen landed former Congressman Rick Hill’s support. Stapleton, unfortunately, has not received the support of a former federal officeholder, so again, I’m left without complete information.

Using this metric, Arntzen comes out the most conservative as Rick Hill’s score of .448 is the most conservative, followed by Marlenee’s .397, and Burns’ .364. This suggests a ranking from least to most conservative of Zinke, Rosendale, and Arntzen with Stapleton as indeterminate.

One method finds Arntzen the most conservative, and another Stapleton. What do we make of this?
I’d put a bit more stock in the method based upon legislative voting records if only because there are more data points available for analysis compared to the voting records of only three federal officeholders. It is clear to me that Zinke is the hardest to pin down ideologically—and while that may provide him with a clear advantage in the general election, it may serve to give some Republican voters pause in the Republican primary.

Unfortunately, the mirror on the wall doesn’t give us a clear answer as to who is the “purest” conservative of them all. Instead, it leaves us with a blurry image of the voting records and beliefs of all three candidates.