It's raining colors here too
It's tiny

Divided nation

Professor Gelman generally believes the red state, blue state paradigm is too simplistic to describe the American electorate.  He has been sharing some of his work on his blog, and has just published a book about this topic.  Recently he produced the following chart, which is gimmick-looking but crystal clear in its message.


Here, economic and social ideology are plotted on a scatter chart, with positive values indicating conservatism and negative values liberalism.  Further, each state is represented twice on the chart, the red point for the Republicans and the blue for Democrats within the state.

This is a cluster analyst's dream data set.  The absolute separation of the Republican cluster and the Democrat cluster is astounding: imagine a diagonal line perfectly classifying all points.

We should not miss a host of details:

  • as Andrew pointed out, "the big thing we see from the graph ... is that Democrats are much more liberal than Republicans on the economic dimension: Democrats in the most conservative states are still much more liberal than Republicans in even the most liberal states."  This is clear from the wide gap on the horizontal axis.
  • there is a small degree of overlap on the social ideology axis so the nation is closer together on that front.
  • but wait a minute, the scale on the social axis is not the same as that on the economic axis.  This means that the extremes are more extreme on the social axis: the difference between MS and VT is roughly 0.8 on the social scale while the largest difference on the economic scale is roughly 0.5.  (here, I am assuming that the scales are comparable to each other)
  • there is high correlation between social and economic ideologies: the points are well-aligned along the 45-degree line
  • especially on social issues, the Democrats are divided within (the elongated shape of the blue cluster).

Reference: Gelman, "Ranking states by conservatism/liberalism of their voters", June 30 2008.


Feed You can follow this conversation by subscribing to the comment feed for this post.


Notable also is that, within Democrats, there seems to be a moderate correlation between social scores and economic scores.

A few more comments:

- There is a division between Dem and Reps; I don't see an internal division within either Dems or Reps. Although the range of social scores is ~25% larger within Dems than within Reps, the range of economic scores within Reps is ~33% larger than within Dems.

- Although the range for the social scale is larger, it has been re-scaled (see
). Assuming that the re-scaling is related to the variability (which it usually is), then we might infer that there is more relative variability in the social scale (which, in fact, is obvious from the plot), but it says nothing about the range. FWIW, the social score is calculated from 13 questions, and the economic score from 10.

- The observation that the points appear to be 45 degree line overlaid onto the plot is incidental to any correlation; correlation is scale (and location) invariant, that is, the correlation between 1,2,...,10 and 1010,1020,...,1100 is exactly the same as the correlation between 1,2,...,10 and 1,2,...,10.


Pretty graph, but the process used to collect the data is completely unclear and not reproducible.
Would be great to get the details here, otherwise it looks a bit like the self fulfilling prophecies - which is still good academic practice ...


DavidS: thanks for the additional comments. You're right about the 45-deg line; I even pointed out that the two scales were not the same so it's any positive slope line

Martin: if you follow the link in DavidS's post, you will see a fuller explanation of how the ratings were derived (from survey data)



thanks for pointing to David's link - that explains a bit more, but does not really make me more comfortable with the data.
The way the survey was constructed and conducted may influence the results extremely - at least it has a good leverage on how strongly you may separate the two clusters.


You're using averages though, and that has a clustering effect which gives the impression that there are NO voters that fall in the middle (of which there are many). It would be interesting to somehow rate the people that voted democrats in one elections and then republican in the next (or vice versa). THAT would tell you something.


I recently did some analysis of the General Social Survey. There is a set of the GSS that splits into two very well-defined clusters as well. See my demonstration here:

The comments to this entry are closed.