This project analyzed BSc student project dissertations to investigate whether the quality of writing was related to the students grade. The results were indeterminate likely due due to marker variation and the relation of the grade to a verbal presentation.
There is some value in the manual processing of larger texts, so making this public.
Readers should note that there is a preprocessed version on the British National Corpus saved as a pickle file. This is a lightweight method to determine if whether words are legitimite English (up to 1998).