Proceedings of the Annual Meeting of the Cognitive Science Society

Psychological applications of human language technologycombined with multidisciplinary approaches to similarity cal-culations and data visualization offer avenues to broaden theuse of students’ own words in program assessment. Wecompared multiple analysis approaches on both simple to-ken counts (word roots and character trigrams) and top-downlanguage indicators from 85 student essays about service-learning events. Bioinformatic distance calculations on wordroot counts provided useable assessment information on at-titude change, showing patterns of word use that match theholistic goals of the assignment. Although these patterns werenot found in a subsequent batch of 81 essays, the tools we areproviding may facilitate other efforts to detect attitude changein student writing about service-learning events.