Most of the research on the effects of college on students that examines the influences of institutional characteristics--what Pascarella and Terenzini ("How college affects students: Findings and insights from twenty years of research". San Francisco: Jossey-Bass 1991) called "between-college" effects--indicate that the descriptors typically used (e.g., size, type of control, curricular mission, selectivity) are generally poor predictors of between-college differences in virtually "any" student outcome once students' precollege characteristics are controlled (economic and occupational attainment are the sole exceptions). Researchers have speculated that the conventional descriptors are too distal from students' experiences to have much effect on differences in outcomes. The between-college effects literature, moreover, concerns itself almost exclusively with the direct effects of institutional characteristics. Using data from a nationally representative study of engineering programs on 31 campuses, this study explores two propositions: (1) that the effects of institutional characteristics in the college effects process are indirect, shaping the kinds of experiences students have, and (2) that institutions' internal "organizational context" features (e.g., programs, policies, and faculty culture) have more influence on students' learning-related experiences than do institutions structural characteristics (e.g., type of control, size, wealth, or selectivity). Findings lend modest support to both propositions.