Rick Perry is slightly to the right of Mitt Romney, but only slightly.
Political scientists have worked for years to find ways of measuring politicians’ ideology. By far, the best method we presently have is known as NOMINATE (and its variants, W-NOMINATE and DW-NOMINATE). I have written about NOMINATE scores before; back in May, I worked with a student to calculate these scores for each member of the Utah legislature. I won’t explain here how NOMINATE scores work here, since I explained it in that post.
The trouble with NOMINATE scores is that you can only calculate them for legislators or members of Congress–that is, for people who vote on a LOT of bills. You use their votes on all those bills to calculate the scores. That means you can’t calculate NOMINATE scores for presidential candidates unless those candidates have previously served in Congress (such as Michele Bachman, Ron Paul, John McCain, Barack Obama, Joe Biden, John Kerry, etc).
Or can you? I just came across an interesting little analysis by Keith Poole, a political scientist at the University of Georgia and one of the folks who developed NOMINATE scores back in the 1990s. Keith Poole has found a way to estimate these scores for Mitt Romney and Rick Perry. He looks at all the members of Congress who have endorsed either Romney or Perry, and he averages the NOMINATE scores of those endorsers to estimate the ideological position of Romney and Perry.
And what does he find? Rick Perry is slightly to the right of Mitt Romney, but only slightly. The gap between Jon Huntsman and Mitt Romney is WAY bigger than the gap between Mitt Romney and Rick Perry.
One thing to be careful with, though: It appears that there are only a couple dozen endorsements by members of Congress, so the margin of error might be wide on these estimates. Jon Huntsman, for example, only has one endorsement by a member of Congress, so we should be particularly careful with his estimated ideology.
A nice way that Keith Poole could double check the accuracy of his method would be to use this roundabout method to estimate Obama’s ideology based on 2008 primary-season endorsements, then see whether that score lines up with Obama’s actual NOMINATE score based on his Senate voting record.
For details, check out Keith Poole’s blog.
Possibly related posts:
Possibly related posts (automatically generated)
BYU vs Utah: Not Just a Sporting Rivalry (December 9, 2011)
Interesting, but I think, as you say, a wide margin of error.
Another potential problem may be in that the ideology measure is based on the ideology of the endorser, not of any issue specific legislation. Endorsers choose their candidate for all sorts of reasons, not all of them ideological. Who is more likely to win, likability, momentum, common sense, religion, etc…not necessarily just ideology.
So, an interesting way to peg the ideology of the presidential candidates, but with potential for wide inaccuracy.