The civil rights issue space in Chapter 3 was estimated using Bayesian Markov Chain Monte Carlo (MCMC) methods (see Clinton, Jackman, and Rivers 2004; Martin and Quinn 2002). I assume there is a unidimensional civil rights policy space over the time period from 1969 to 2004 (the 91st to 108th Congresses). I then estimate House members' ideal points on this civil rights issue space.

The estimation is based on the civil rights votes of House members during these Congresses. These civil rights ideal point estimates – similar to ideological positions – are comparable across time periods and vary over time for legislators who serve over multiple Congresses. I estimate all legislators' positions on a scale on which negative indicates more liberal positions on civil rights and positive numbers indicate more conservative positions on civil rights. Enelow and Hinich (1984) posit that a roll-call vote is a choice between a position along a (unidimensional, in this instance) policy space consistent with a “yea” vote (ϕj) in favor of the proposal and that with a “nay” vote (γj) against it. The unidimensional civil rights policy space, X, a subset of the real line, is a continuum of policies ranging from extremely pro-civil rights on the left (negative values) to extremely anti-civil rights on the right (positive values). Ideological predilections along the civil rights space may be induced by preferences and other various factors, but that is beyond the scope of this research.