This research project, from the University of Nebraska, analyzes patterns of diction in Austen's six major novels. The textual data from the project can be downloaded here.
Includes facsimile images and transcriptions over 1,000 pages of manuscript in Austen's hand, including portions of Persuasion, Lady Susan, Sanditon, and The Watsons. For ideas about pedagogical use of this archive, see this article.
The HathiTrust repository contains hundreds of digitized copies of works by Austen, which can be browsed, downloaded and searched. The HathiTrust Research Center provides text mining tools that can be used to analyze these digitized texts.