Party or Progressivism? Determinants of Bill Activity in the U.S. Senate

Irregular Times reports today on the sponsorship (offering a bill for consideration) and cosponsorship (promoting a bill by signing one’s name) activity by members of the United States Senate in the years 2007-2008, offering figures for 98 Senators that range from a low of 22 bill (co)sponsorships to a high of 469 bill (co)sponsorships. See if you can guess who ranked lowest and who ranked highest in the Senate before you go see for yourself.

The article mentioned that “74.5% of Democratic Senators are located in the top half of the activity rankings, and 76.6% of Republican Senators are located in the bottom half of the activity rankings.” Looking at the figures, another way of expressing the relationship between party and sponsorship/cosponsorship bill activity is to say that being a Democrat is positively correlated (at a level of +0.46) with bill activity. Democrats on average tend to sponsor or cosponsor 89.7 more bills than Republicans. That’s a large substantive difference, but is the explanation of party telling the whole story? No: squaring the correlation tells us that variation in party accounts for only 21.6% of the variation in bill activity.

A 21.6% explanation isn’t awful when trying to predict the thoroughly chaotic behavior of people in groups, but we can do better. I’ve looked at ratings of Senators on an progressive political index at Progressive Patriots, and the correlation between Senators’ progressive action score and Senators’ level of bill activity is higher (+0.69). Just by itself, Senators’ progressivism accounts for 48.0% of the variation in bill activity.

When these two explanatory variables are combined into a single model so we can take a simultaneous look at the separate effects of party and progressivism, a record of progressive action trumps party in explaining the extent of a Senator’s bill activity. The scale of the progressive action score runs from 0 to 100, and the slope of the effect of that score on bill activity is 2.94. Substantively, this means that we can expect a Senator earning a progressive score of 100 to sponsor or cosponsor 294 more bills (2.94 * 100) than a Senator from the same party who earns a progressive score of 0 (2.94 * 0). What about the effect of party, controlling for progressivism? Such effect is very small (and would be considered statistically insignificant if we were making an estimation about a sample rather than the whole population of the Senate). Substantively, the effect of party on bill activity actually runs in the negative direction, meaning that if you compare two senators who are have the same progressive score, the Democrat will tend to have a slightly lower level of bill activity than the Republican (with 24 fewer bills to the Democratic Senator’s credit).

The bottom line is that if you’re wanting to understand why some members of the U.S. Senate are more active in bill sponsorship and promotion than other members, ditch party as an explanation. A much better explanation of that variation is the progressivism that a Senator maintains in his or her legislative conduct.