Branches were produced by agglomerative clustering in R using cor() and hclust(), and the resulting tree cut to highlight major divisions.
Seven topics are excluded from this figure because they were especially non-content-bearing, e.g. topics comprised of non-words formed by unsuccessful character recognition during the scanning process or topics indicating merely a non-English language.
NB: These dissertations were written, as best I could determine, for PhD programs in the Consortium of Doctoral Programs in Rhetoric and Composition.
To be linked in this figure, two topics must each contribute at least 11 percent to the same dissertation, and such a pairing must occur at least 4 times.