(Dis-)proportional representation


An interesting point raised by Chuck over at To Be Determined (via firedoglake):
The 44 Democrat senators represent about 3.5 million more people than the 55 Republican senators do.

By my calculations, this means each Democrat represents around 3.4 million people and each Republican represents around 2.7 million. On a souls per seat basis, it's a significant difference.

In a way, this is unsurprising: The houses of Congress were defined to be different in a couple of important ways, one of which was the way that representation is mapped onto population. In the House, it's one vote per district (population unit), whereas in the Senate it's two votes per state (arbitrary geographical unit).

I.e. —not that you need me to clarify this elementary point, but I shall anyway because everyone loves a pedant — the Senate isn't population-proportional and wasn't designed to be. So it comes as no surprise that there is a difference, even a large one, in the number of people represented on either side of some partition drawn through the Senate (in this case, along party lines).

This difference is systematic enough to make a difference: More populous states do tend to vote Democratic, but we sorta already knew that.

The reason I'm bringing this up at all is this: I wonder to what extent the assignment of Representatives is population-proportional.

It's supposed to be, but it can't possibly be precise: There are 435 seats in the House, a number that has remained unchanged since 1912. After each Census, seats are sometimes swapped between states on the basis of a mathematical formula that depends on the population of each state (see "Congressional Apportionment"), but since one can't re-apportion a fractional Congressperson, some error will creep in at that point.

After apportionment, districting decisions are left up to the states themselves, where the "fractional Representative" issue can generate more discrepancies. Also note that a systematic migration of individuals from country to city will result in smaller changes in voter number in rural areas than urban areas (because many rural areas feed a smaller number of urban areas).

So I wonder: Even though each district is supposed to have a (roughly) equal number of citizens in it, is that actually the case?

And then, given that each district almost certainly doesn't have the same number of citizens: Is there a systematic disproportional representation in the House of Representatives.

And finally: If there is a systematic disproportionality in the House, does it favor Republicans or Democrats?

Figuring this out will take a little bit of research: For each district, I need the Representative's party affiliation and the district's population. The first one is easy but the second one is harder. I have some ideas about where to start looking, but if anyone stumbles over a table with that information, please post to comments or send me a Feedback email.

Posted: Fri - April 1, 2005 at 02:47 PM   | Category:     |   |   | |


©
eXTReMe Tracker