Below we use Optimal Classification (OC) in R to plot three votes by the Senate today concerning a Continuing Resolution to fund government operations past September 30. The first is a 79-19 vote to invoke cloture and end debate, the second is a party-line 54-44 vote to remove the amendment inserted by the House stripping funding for the Affordable Care Act (ACA), and the third is an identical 54-44 party-line vote to pass the Continuing Resolution that includes ACA funding.
OC perfectly classified both party-line votes and does a good job modeling the 79-19 vote on cloture. The 19 Nay votes came primarily from the most conservative wing of the Senate Republican caucus and OC picks this up with a cutting line dividing the Republicans along the first (liberal-conservative) dimension with only 3 classification errors.
This vote illustrates the usefulness of the spatial model as opposed to other measures like party support scores because it picks up ideological divides within the parties, both from the center and the extremes.