White, KatherineBesner, DerekYoung, Torin2021-08-252021-08-252021-08-252021-08-09http://hdl.handle.net/10012/17254Lectures are an important part of the post-secondary experience. Optimizing various aspects of this experience for the benefit of students’ learning has been examined (Mayer, 2019). However, the linguistic features of lectures and how these features might affect student learning have been overlooked in the extant literature. Recent studies have utilised Coh-Metrix, an automated text analyzer, to examine discourse in both texts and lecture discourse (Graesser, McNamara, & Kulikowich 2011; McNamara, Graesser, McCarthy, & Cai, 2014; Medimorec, Palvik Jr, Oleny, Gaesser, & Risko, 2015; Morgan, Burkett, Bagley, Graesser, 2011). We extend this effort here by analyzing linguistic features of lectures and how they are associated with student performance. In particular, we were interested in determining whether (a) computationally generated measures of language are associated with student performance and (b) whether different associations are observed with different testing methods (multiple-choice vs. short answer). We demonstrate that a lecturer's narrativity, syntactic simplicity, and referential cohesion are associated with performance on multiple-choice tests. Preliminary results suggest a different pattern of association for short answer tests.enlinguistic featureslecturesstudent performanceMaster Thesis