The attitude assessment model presented here is intended to be an improvement over methods traditionally used to study attitudes. It takes into account findings by Astin (1969) and Berkowitz (1968), calling for a model expressing the covert behavior of a subject in terms equivalent to those used to anticipate overt behavior. This paper presents the development of an attitude scale from "real life" phenomena based on the actual behavior of the individual, both covert and overt. It assesses the intensity of a given attitude toward or about campus life and suggests an improved method of assessing the reasonable degree of predictability in the linking of attitudes and manifest behavior. The model allows the partitioning of stimuli involved in attitude intensities in such a way that they can be ranked in potency to yield a hierarchy of "volatile" areas deserving administrative attention. At the University of Tennessee, 5 areas of student life were chosen for investigation: classroom conditions, satisfaction with faculty, satisfaction with administrators, total University environment, and present level of morale. Black, white, and international students were selected according to race, sex, class, college and grade point average. The components for assessing the intensity of an attitude were determined by the students' responses during interviews to questions on their significant observations and experiences at the University. (JS)