This article reports on recent efforts to develop an intelligent tutoring system for proof construction in propositional logic. The report centers on data derived from an undergraduate, general education course in Deductive Logic taught at the University of North Carolina at Charlotte. Within this curriculum, students use instructional java applets to practice state-transition problem solving, truth functional analysis, proof construction, and other aspects of propositional logic. Two project goals are addressed here: 1) identifying at-risk students at an early stage in the semester, and 2) generating a visual representation of student proof efforts as a step toward understanding those efforts. Also discussed is the prospect for developing a Markov Decision Process approach to providing students with individualized help.