SILVER; US House

Last week I posted the complete SILVER listings for the US Senate, now for the US House. First a graph of the House Republicanss, just like when I did this for the Senate, the DW-Nominate scores are plotted on the x-axis and PVI on the y-axis.

SILVER; US Senate

As promised here are the SILVER rankings for the entire U.S. Senate, but before we get to that, a few graphs.



That's a scatter plot of Senate Republicans, the X axis represents the DW-Nominate score and the Y axis is the PVI score. You'll notice that the scatter plot is basically random and that is reflected in the r-squared number, there isn't much of any correlation between Senate Republicans voting tendencies and the partisan nature of the states they represent.

Here's the Senate Democrats scatterplot:



What you see is a somewhat tighter, somewhat more linear distribution, with an r-squared value approaching 50. The Senate Democrats DW-Nominate and PVI scores are more correlated than the Senate Republicans.

When you put the two graphs together and you can actually see the partisan split.



What you can also see by looking at the SILVER rankings for the entire Senate GOP below, and as Nate pointed out last week, the Tea Party has actually been pretty smart in the incumbents they've taken out, as Bennett and Murkowski are among the least valuable GOP Senators according to this metric.

SILVER 2.0

Back in June I introduced SILVER as a metric to measure how partisan a member of congress is compared to the voting tendencies of their district or state. I did this by combining PVI and DW-Nominate and I scaled it (or attempted to) to the same -1 to 1 scale that DW-Nominate uses. Since that time I’ve been forced to give the issue some more thought because of a baseball stat called NERD.

Believe it or not when I first developed the SILVER stat I was not aware that Crisitunity from the excellent Swing State Project blog had already developed a similar thing. Crisitunity computed his version in a different way, ranking the PVI and DW-Nominate scores of all members of the house on a 1 to 435 scale and then using the difference between the two rankings as his score. This might be a better method than the one I employed in my first version, but in the end we’re trying to measure two different things.

The system that Crisitunity developed measures house members against each other without regard to party affiliation, the point is to measure the overall liberalness of a legislator in the context of their district against the house as a whole. SILVER is meant to measure a legislator against his or her own party, not the entire legislative body.

The Senators

Over the past few weeks I've been analyzing Minnesota's congressional delegation, focusing first on the state's house delegation, and then on some specific house members. Now it's time to look at the state's Senators.

You can go here for a more detailed explanation of what the different numbers represent; briefly, the "DW-Nom" number is the legislators DW-Nominate score and the "PVI" number is Minnesota's Partisan Voting Index score. The "PVI v ave" and the "DW-N v ave" represent a comparison between the legislators score and the average score of their party caucus. The "SILVER" score is a composite of these two comparisons and the rank number is the legislators SILVER rank among their party caucus.



This shouldn't be that surprising; Al Franken is one of the most valuable liberals in the Senate, behind only Russ Feingold (-0.757) and Sherrod Brown (-0.375) in SILVER and his DW-Nominate score is to the left of even Keith Ellison, making him the state's most liberal legislator. Amy Klobuchar meanwhile is the only member of Minnesota's congressional delegation who has a SILVER score on the wrong side of 0 for her party.

In a strange coincidence Wisconsin has a similar situation with Russ Feingold being the most liberal member of the Senate, according to SILVER, and Herb Kohl (0.112) ranking 44^th, tied with Amy Klobuchar.

Let's take a look at another chart.



Here you can see the DW-Nominate score's of Minnesota's last eight Senators, as well as a graph of the total DW-Nominate scores for the state's two active Senators combined. As you can see, besides the 107^th congress when Minnesota had an all liberal Senate delegation, the total DW-Nominate score of the current Klobuchar/Franken alignment is as liberal as Minnesota's had since the Humphrey/Mondale/McCarthy days. And that probably won't change anytime soon, given the current dynamics there's a pretty good chance that this will be our Senate configuration until at least 2014 when Al Franken will be up for re-election.

The Swingers and 2010

Tim Walz and Eric Paulsen represent the two swing districts in Minnesota according to PVI, the 1^st and the 3^rd. They are both young for congress persons, Walz is 46 and Paulsen is 45, they are both relatively new to the house and while Paulsen has yet to face reelection, Walz breezed to victory in '08 with over 62% of the vote. What Walz has done in his district is what Paulsen appears to be trying to do in his, moderating himself from the parties activist base.

name	PVI	PVI v Ave	DW-N	DW-N v Ave	SILVER
WALZ	1	10	-0.278	0.07	-0.15
PAULSEN	0	-11	0.543	-0.09	0.14

In case you missed the post last week, DW-Nominate is a partisan scoring system ranging from (-1) to 1, with (-1) being very liberal and 1 being very conservative. PVI, partisan voting index is a measure of the partisan nature of a congressional district that I've formatted the same way as the DW-Nominate score, positive for a conservative leaning district, negative for liberal.

What you see when comparing Walz to Paulsen is just how similar they are among their respective party caucuses. They are both from districts that have almost no partisan tilt, far from the average of their party. Their DW-Nominate scores however are very close to the average member of their party, suggesting that they are more valuable to their party than the average legislator, which you can see by looking at their SILVER scores.

While taking out Eric Paulsen would be nice, if Tim Walz's success in the 1^st is any indication he will be hard to unseat, especially in a cycle that looks to be favorable to GOP candidates. In addition Paulsen got elected in a three way race with almost 50% in his first go and hasn't really done anything to change voters' perception of him since that time. Based on PVI however, the 3^rd is the most favorable district for democratic takeover of the three GOP held districts.

This brings up the question of who the easier target is for Democrats this November, Eric Paulsen or Michelle Bachmann. While Bachmann's district has a much more daunting +7 PVI, Bachmann herself only won reelection by 3 points against a candidate who was underfunded most of the race and only started gaining ground after Bachmann imploded on Hardball. Since that time Bachmann has further defined herself as the nuttiest of wing nuts, capable of saying just about anything.

So, we have a favorable district with a savvy candidate versus an unfavorable district with a loon, which race would you put resources into if you could only choose one?

The Chairmen

Among Minnesota's Democratic congressional delegation there are two in top leadership positions; Jim Oberstar is Chairman of the House Transportation and Infrastructure Committee and Colin Peterson is Chairman of the House Agriculture Committee. Both have represented the state for a long time; Oberstar for over 25 years, Peterson for close to 20, and both represent large, mostly rural districts. But as we saw with their respective SILVER scores, they are quite different in how they vote, at least in the 111^th congress. While Colin Peterson had a slightly liberal for his district score of -.06, Jim Oberstar had a superstar level -.55. Let's take a look at their DW-Nominate scores over the entire course of their time in office to see if that changes our perception any.

What you see is that over the years Oberstar has become marginally more conservative while Peterson has made moves in the liberal direction. This becomes even more evident when you look at their current scores for the 111^th congress versus their mean and first year scores.

	111	mean	1st	+/- mean	+/- 1st	stddv
PETERSON	-0.175	-0.139	-0.103	-0.036	-0.072	0.0242
OBERSTAR	-0.539	-0.544	-0.550	0.005	0.011	0.0033

As we saw earlier, in the 111^th congress Peterson is voting more liberally then he has in the past and Oberstar is voting slightly more conservatively. What you can also see by looking at the standard deviation of their scores, as well as the +/- versus the mean and first year, is that Peterson's voting patterns have shifted more substantially over the years then Oberstar's, who has essentially the same score now as when he started.

What does this mean? Well, Jim Oberstar has been voting the same way for 25 years and Colin Peterson has ever so slightly become more liberal than when he first joined the congress. If we assume the PVI of the 7^th CD is the same now as it was in 1991 (a dubious assumption) then Peterson would've had a SILVER score of .08 in his first congress, compared to the -.06 he has so far this congress, a -.14 swing in the liberal direction. It's probably not incidental that his vote share has also risen steadily over the years to 72% in 2008. While Colin Peterson is no liberal, he is liberal for his district and getting more liberal every year.

Minnesota’s Congressional Delegation

In an effort to get a better understanding of the voting tendencies of our congressional delegation I compiled the DW-Nominate scores for all of Minnesota's politicians from the 111^th congress. DW-Nominate is a scoring system for politicians, running from around -1 to 1, with -1 being very liberal and 1 being very conservative. There's a few ways to look at these numbers; first we'll just look at the raw DW-Nominate scores.

Representative	DW-N Score
ELLISON	-0.58
OBERSTAR	-0.54
MCCOLLUM	-0.44
Average Dem	-0.35
WALZ	-0.28
PETERSON	-0.18
PAULSEN	0.54
Average GOP	0.63
KLINE	0.69
BACHMANN	0.71

This is probably not too shocking of a chart, for the most part reflecting the conventional wisdom about the liberal/conservative spread of our state's delegation, except for maybe Jim Oberstar who ranks as more liberal then I would have suspected. Now let's adjust those scores to the party average, so for instance, the Democrats average DW-Nominate score was -.35, so Keith Ellison would then score a -.23 and Colin Peterson would score a .17, meaning Ellison is more liberal than the average Democrat and Peterson is more conservative.

Representative	DW vs average
ELLISON	-0.23
OBERSTAR	-0.19
MCCOLLUM	-0.09
WALZ	0.07
PETERSON	0.17
PAULSEN	-0.09
KLINE	0.05
BACHMANN	0.08

Again, nothing to ground breaking; Paulson is more liberal than the average GOPer, Walz and Peterson are more conservative than the average Democrat. The next thing I want to do is take a look at the PVI, partisan voting index, for the districts. PVI measures how Republican or Democratic a district is by averaging the districts results of the past two presidential elections and comparing them to the overall results. The difference between the two numbers is the districts PVI score, which for the purposes of this exercise we'll format the same way as the DW nominate scores, positive for GOP leaning districts, negative for Democratic leaning districts.

Representative	PVI	PVI vs Average
ELLISON	-23	-14
OBERSTAR	-3	6
MCCOLLUM	-13	-4
WALZ	1	10
PETERSON	5	14
Average Dem	-9
PAULSEN	0	-11
KLINE	4	-7
BACHMANN	7	-4
Average GOP	11

Tim Walz and Eric Paulson reside in the two swing districts in the state, according to PVI, with Oberstar's 8^th and Kline's 2^nd on the verge of being swingy. Obviously Colin Peterson's 7^th is in a bit of a special category, being solidly Republican yet sending Peterson back to congress year after year.

Now we can take these two measures and combine them into one number that will tell us how partisan a member is for their district on a rough -1 to 1 scale. We'll call this number the Selectively Integrated Legislator Variance Evaluation Resource, or SILVER for short.

Representative	SILVER
ELLISON	-0.03
OBERSTAR	-0.55
MCCOLLUM	-0.04
WALZ	-0.15
PETERSON	-0.06
PAULSEN	0.14
KLINE	0.31
BACHMANN	0.26

What stands out to me on this chart is when put into the context of their districts Oberstar is quite a bit more liberal than his colleagues; in fact he's 7^th among all Democrats in SILVER. My representative, Keith Ellison, whom the conventional wisdom has as an uber-liberal, is right around average. I also think this is vindication of a sort for Colin Peterson who has taken a lot of flak for some of his recent votes but is, by this measure, more liberal then Keith Ellison or Betty McCollum.