Quantitative data on the use of imperium (empire) in early medieval advice to kings and queens

This dataset is a quantitative dataset of the use of the terms imperium, imperator, and connected words in a group of advice texts for Carolingian (c. 751-888, western Europe) rulers. I used Quanteda, a set of packages for the R programming language, to gather this data from the plaintext and perform the statistical operations needed, including collocation and keyness analyses. The dataset is then used in the associated article to argue for changing conceptual meanings of the term over the eighth and ninth centuries.