We’re about to enter a Virginia election season and there’s one thing that’s certain: We’ll hear Republicans accuse Democrats of being liberals and Democrats accuse Republicans of being conservative.
Of course, sometimes not even those labels suffice as pejoratives. Sometimes we’ll hear candidates tack on the adjective “most” as in “so-and-so is the most liberal legislator” or the “most conservative” — depending, of course, on what part of the state you’re in.
It’s amazing how much power we give to certain words, as if those words actually mean something. Sometimes they do, of course, but sometimes they don’t. Calling someone something does not make it so. Wouldn’t it be better if we had some objective measure of just who is liberal and who is conservative and how liberal and conservative they actually are?
Behold! There is! In the 1980s, two political scientists — Keith Poole of the University of Georgia and Howard Rosenthal of New York University — developed a computer model to crunch roll call votes and spit out a score. This algorithm is known today as DW-Nominate, which stands for Dynamic, Weighted Nominal Three-Step Estimation.
The Society for Political Methodology has said that “One can say perfectly correctly, and without any hyperbole: the modern study of the U.S. Congress would be simply unthinkable without NOMINATE legislative roll call voting scores.”
And now, not just Congress. A statistics major at Yale University recently poured the voting records of Virginia state legislators from the 2020-2021 sessions into this model and, after some computer gyrations (our layman’s term), Armin Thomas produced two fascinating maps showing where the House of Delegates and state Senate fall on the scale. You can find them here: https://thorongilforecast.shinyapps.io/dwnominateleg/.
In the House, you might think that the chamber’s self-avowed socialist — Lee Carter, D-Manassas — would be the most liberal. You’d be wrong. On the left-to-right “economic redistributive” axis, Carter actually ranks as the 53rd most liberal legislator. A veritable moderate! If 0 counts as the midpoint, or the political center, and -1.0 is the ultimate liberal score on the left and +1.0 is the ultimate conservative score on the right, Carter is -0.69 on the leftward side of the scale.
Here’s something Republicans might like: That means they can run against 52 of the 54 other Democrats in the House and claim, with factual support, that they are “more liberal than a socialist.”
The most liberal legislator on this scale is Lashrecse Aird, D-Petersburg. Her score is -0.96, just shy of liberal perfection. Aird’s name might be familiar with some of our readers because a few years ago she joined with a conservative Republican — Will Morefield, R-Tazewell — to support a bill he was pushing to provide tax breaks to companies locating in certain economically-distressed localities.
That just goes to show two things — the economic interests of rural areas often overlaps with those of central cities, and political ideology isn’t always the best way to look at politics even though it’s the one we use most.
The most conservative legislator in the House is Dave LaRock, R-Loudoun, who scores 0.98, a shade closer to conservative nirvana than Aird is to liberal nirvana. LaRock, you may recall, was one of those pushing to have Congress reject Virginia’s electors for Joe Biden.
Here’s something that might surprise you: This chart shows many Republicans falling closer to the political center than the Democrats, who are pretty uniformly on the left. The delegate closest to the center is Carrie Coyner, R-Chesterfield, who scores +.13. There are no Democrats who come close to the center.
In the Senate, the most liberal legislator is Mamie Locke, D-Hampton, who scores -0.99, almost perfectly liberal. The most conservative is — no surprise — Amanda Chase, R-Chesterfield County, who at 0.9984 is almost perfectly conservative.
Unlike the House, the chart does show some relative moderates in the Senate — Lynwood Lewis of Norfolk (-.0.6977) and Chap Petersen of Fairfax City (-0.6844) on the Democratic side, Jill Vogel of Fauquier (0.4144) and Siobhan Dunnavant of Henrico (0.5034) on the Republican side.
Now for the really interesting part: We’re all familiar with the left-right scale in politics. This ranking adds a north-south scale that measures their deviation from the party line or, as the chart shows it, the “social/crosscutting/maverick axis.”
In the House, the most maverick Democrats are Carter (whose maverickness, if that’s a word, often makes him more liberal than his colleagues) and Roslyn Tyler of Sussex and Steve Heretick of Portsmouth (whose maverick tendencies sometimes pull them toward the center). That maverick voting record is what actually makes Carter score less liberal than he might seem; sometimes those on the far left and the far right might vote against the same bill, just for very different reasons.
The most maverick Republicans are Lee Ware of Powhatan (tending to be more socially liberal) and John Avioli of Staunton and Terry Kilgore of Scott (tending to be more conservative).
In the Senate, the most maverick Democrats are Lionell Spruill of Portsmouth (tending to be more liberal) and Petersen and Creigh Deeds of Bath (tending to be more conservative). The most maverick Republicans are Emmett Hanger of Augusta (tending to be more liberal) and Vogel and Dunnavant (tending to be more conservative).
What’s the opposite of being a maverick? The chart also allows us to measure who is the most conventional. In the House, two legislators almost zero maverick votes — Democrat Jeffrey Bourne of Richmond and Republican Kirk Cox of Colonial Heights sit squarely on that left-right scale (just in very different places) and rarely deviate up or down.
In the Senate, the least maverick legislator is Todd Pillion, R-Washington, followed by John Bell, D-Loudoun.
What does all this tell us? Keep in mind the old saying that all models are wrong; some are just more useful than others.
This one gives us a statistical basis to make judgments from, so that any labels we apply (far left, centrist, far right) actually have some basis in fact.
Sometimes the numbers here confirm what we already knew, other times they challenge them.
The results also help us see some broad patterns that may or may not be surprising (there’s more diversity of thought among both House and Senate Republicans than there is among House and Senate Democrats). And there are no real moderates left.
Who knew? Now you do, and you have some numbers to back it up with.