Excessive course attrition is costly to both the student and the institution. While most institutions have systems to quantify and report the numbers, far less attention is typically paid to each student's reason(s) for withdrawal. In this case study, text analytics was used to analyze a large set of open-ended written comments in which students explained their reason(s) for course withdrawal in their own words. The text for all comments was extracted verbatim from the course withdrawals database of Florida State College at Jacksonville, a large, diverse, multicampus institution located in northeast Florida. An initial set of 616 comment records from the beginning of the fall 2010 term was used to develop a preliminary text analytics model. This model revealed 11 major category nodes and successfully classified 96.1% of all withdrawal records into one or more categories. The model was retained and further tested using a second set of 679 records from the spring 2011 term and found to successfully classify 98.7% of the spring records. At the broadest level, withdrawal explanations were found to include both academic and nonacademic student rationales. Leading academic rationales involve course scheduling adjustments, delivery method preference changes (e.g., classroom vs. online), and faculty related reasons. Leading nonacademic rationales include personal issues especially involving job/work, family, financial, and health matters. The limitations of the study along with implications for practice, administrative decision making, and future directions for the expanded use of text analytics in institutional research are discussed.