This study examines the different processes that institutional researchers use to calculate the number of transfers-out for the Integrated Postsecondary Education Data System (IPEDS) Graduation Rate Survey (GRS), with a particular focus on how the process is influenced by institutional mission and data availability. Policy makers and the public have been very interested in what colleges are doing to improve degree attainment for over seventy years (e.g., McNeeley, 1937). In the 1970s, research on dropping out, stopping out, and transferring began to increase (e.g., Astin, 1975; Cope & Hannah, 1975; Tinto, 1975). Pantages & Creedon reported that only five out of ten students who enter college in the United States will eventually graduate from the same college and that, "Of the five students who dropped out of the college altogether, four will reenroll at a different college, and of those four reenrollees, only two will graduate" (1978, p. 49). In 1989, a study of graduation rates by the National Institute of Independent Colleges and Universities found that just 42.7% of students at 4-year public institutions and 54.2% of students at private institutions graduated within six years (Porter, 1989). Given the wide-spread interest in improving degree attainment, understanding what happens to a student after he or she leaves an institution is essential; unfortunately, very little is known about the different processes that institutional researchers use to track transfers-out of an institution. This study provides an important first look at what the data in IPEDS can tell us about transfers-out. Comparing the total number of transfers-out from IPEDS with the 1995-96 Beginning Postsecondary Students (BPS) Survey suggests that transfers-out are significantly underreported on the IPEDS Graduation Rate Survey. Key findings include: (1) Institutions in the lowest total cohort graduation rate quintile are more likely than institutions in the highest quintile to report the number of transfers-out; (2) While the percentage of institutions reporting the number of transfers-out declined between 2002 and 2007 (41% to 35%), the percentage of the adjusted cohort enrolled in institutions reporting the number of transfers-out has remained steady (~49%); (3) Institutions located in states with state and/or system student unit record systems are more likely to report the number of transfers-out; and (4) Larger institutions are more likely to report the number of transfers-out than smaller institutions. [This report is an AIR/NCES Data Policy Fellowship Report.]