The study attempted to: (1) identify and examine college teachers' espoused and implicit theories of teaching, using Menges and Rando's analysis categories; and (2) explore differences between assumptions about traditional and non-traditional students when college teachers based their diagnoses of problematic events on untested speculations. Faculty (N=69) from a metropolitan community college in the midwestern United States were interviewed. Analysis indicated that, overall, a majority of the participants (57%) espoused content theories (CT), 23% subscribed to motivation theories (MT), and 20% indicated adherence to process theories (PT). The majority of the participants (65%) used assumptions to diagnose problematic events and responded by defaulting (68%). There was a significant relationship between espoused theories and diagnoses, but no significant relationship between teaching area (college-parallel or career) and espoused theories, diagnoses, or actions. The results of the study were consistent with some of the findings of Menges and Rando: the majority of college teachers relied on assumptions to diagnose situations that were not proceeding as anticipated. Many participants, especially the CT teachers, attributed more positive characteristics to non-traditional students than to traditional ones, and the extent to which participants relied on untested assumptions for diagnosis and action in problematic situations involving students is a cause for concern. Contains 10 references. (GLR)