College students seem especially adept at procrastination. Large percents report that it causes lower quality papers and projects, lower exam scores and, to a lesser extent, late or missing assignments. Using survey data collected from a broad cross-section of students and faculty we argue that procrastination can be modeled as a logical, albeit sometimes inefficient, behavior within a rational choice framework. By looking at differences in perceived costs and benefits across individuals and across circumstances we can predict who procrastinates, under what conditions they procrastinate, and how it might effectively be countered. (Contains 2 tables.)