« Twitter roundup | Main | Twitter roundup »

May 09, 2011


What package did you use for those graphs?

I made them in Stata.

I think Pawlenty is a likely nominee, but for me that's more a judgment than a data-driven conclusion.

1. You excluded potential candidates related to a previous President, because these candidates have such a big structural advantage. In that case, your data alone couldn't eliminate Jeb Bush as the likely Republican nominee. I think Jeb's relation to George and George hurts him more than it helps. So, I agree that he's not a likely nominee, but that's a judgment.

2. You say, "At this point in the election cycle, the preferences that matter are those of the activists, elected officials, donors, and party elites who take part in the so-called 'invisible primary." I don't doubt that principle, but where you get your data on this group of people and how do you analyze it?

3. You don't mention money. An article in the New York Times a day or two ago indicated that Romney was way ahead in organizing his fund-raising. IMHO money is a big advantage in general. However, Romney is damaged goods. OTOH Pawlenty has looked good in his limited exposure. So, in this case, despite Romney's name-recognition and his financial edge, I think Pawlenty is a more likley candidate. Again, that's a specific judgment about this particular group of candidates.

On 1, the data only go through 2008, but the same issue would apply to Jeb. On 2, the invisible primary is (as its name suggests) much harder to cover and analyze, but there are ways to do it. For instance, see the book by my friend Hans Noel and his colleagues that focuses on elite endorsements: http://j.mp/mGMrn3 On 3, money is largely driven by candidate viability so it's very hard to analyze as an independent factor.

Another great post! Keep it up!

The comments to this entry are closed.