I find this research from Pew really interesting, as it tries to segment voters beyond a one-dimensional political spectrum. Segmentation can produce meaningful results, but it can also be misinterpreted. For instance, consider this chart:
To say that Democrats agree on the second point is entirely untrue. Based on the data, roughly three quarters of Democrats hold one view while a quarter of Democrats hold the opposite view. The fact that three particular segments have similar proportions in each camp is not evidence of agreement; it merely says that the segments Pew defined fail to explain the divergence of opinion.
Overall, though, the study looks great and I’d check it out in full.