Orientation and Getting Started
Early Development of the Legislative Branch and the Problem of Representation
Enumerated Powers of Congress
Implied, Expansive, and Limited Powers
The Two Congresses: Representation and Lawmaking
Separation of Powers and Interaction between the Branches
Institutional Development and Change
Congress and the American People

Bicameralism Today

Increasing Partisanship and Geographic Bases

Increasing levels of partisanship in both the House and Senate, beginning in the 1980s, partly reflect the regional and geographic bases of the two parties. As a result of gradual realignment, Republicans represent more Southern states and congressional districts within those states as well as areas with large rural populations, whereas Democrats now tend to represent Northeastern and Western coastal states and districts within them, as well as areas with large urban populations. The trends are worth noting. In 1961, there were no Republican Senators representing the eleven Confederate states (AL, AR, FL, GA, LA, MS, NC, SC, TN, TX, and VA), by 1980, there were 10 and by 2000 there were 14 Republicans, more than half of the total. Today, 19 of 22 Senators from Confederate states are Republican. In the House, the margin of Democrat to Republican seats in the Confederate states was 110-10 in 1955; 83-47 in 1985; and 64-73 by 1995, after the GOP won the majority in the 1994 elections. Today, Republicans hold 101 of those seats and the Democrats 37. A greater mixture of Democrats and Republicans represent Midwestern and Mountain states, but on the whole, Senators and House members from the same party tend to represent similar voting constituencies. (Abramowitz 2010)

Tracking Partisanship

The degree of partisanship in the House and Senate can be measured by the average party unity scores on roll call votes taken on the floor. A “party unity” roll call vote is one in which the two parties are divided; more than 50% of Democrats vote on one side (either yea or nay) and more than 50% of Republicans vote the opposite. A party unity score for an individual member is the percentage of times the member votes with the majority of his/her party on party unity votes. A member with a 95% party unity score is deemed highly partisan because he/she votes with the party on nearly every vote that divides the two parties, whereas one with a 55% score is not very partisan because he/she only votes with the party just over half the time. The average party unity score for all Democrats and all Republicans in a given year serves as a measure of overall partisanship for each party. By tracking average party unity voting over time, we can detect trends in partisanship. 

The two figures below track average party unity for both parties in the House and Senate from 1970 to 2010 (the averages after 2010 are very similar. The figures show that the chambers have become more partisan. The figure for the House shows that the average party unity voting increased for Democrats from 75 in 1970 to 93 in 2010, and for Republicans from 78 to 93 over the same period. The scores for both parties dipped in 2013, though they are still high. 

A similar pattern holds for the Senate. Democratc party unity rose from 71 in 1970 to 94 in 2010 and remained at that level in 2013. Party unity for Republicans rose from 71 to 93 during from 1970 to 2010 and dipped in 2013, but it remained very high for this period. Thus, in terms of trends in partisanship, the Senate looks very much like the House.

Re-election of House Members Compared to Senators

The chambers differ in several other respects, however. The differences in term lengths and the size of constituencies (with a few exceptions) affect the electoral and representational demands of senators and House members.  Outside of the seven states that have one at-large House member—Alaska, Delaware, Montana, North Dakota, South Dakota, Vermont, and Wyoming—senators have larger, more diverse constituencies than House members. Thus, a larger portion of senators than representatives are likely to face a challenger, either from within their own party or by a candidate from the other party, and a larger portion of Senate incumbents are defeated for re-election. In elections from 1990 to 2012, an average of 93% of House members compared with 88% of Senators were re-elected from among those running for office. (Vital Statistics on Congress) A larger percentage of Senators than House members face competitive elections. In 2012, for instance, 56% of Senators and only 39% of House members won their election by 60% or less, and 32% of Senators compared with 18% of House members won by 55% or less. (Vital Statistics on Congress) Senate incumbents spend more money on re-election than the House incumbents, even when controlling for the length of terms. In 2012, the average cost of winning a House seat was $1,596,953 whereas the average Senate seat cost $10,351,556 (Vital Statistics on Congress), largely as a result of the need for senators to run statewide across multiple different media markets.