Graduation rates are often used indicators of the performance of institutions of higher education. How institutions are represented can be profoundly affected by different approaches used to develop the data for the rates. The impact of variations in how these rates are constructed is illustrated by the example of the City University of New York (CUNY). The completion rate 3 years after entrance is less than 10 percent for the bachelor's degree but at 8 years after entry, the rate is 40 percent, a five-fold difference. It can be shown that many methods of graduation tracking are not adequate to accurately reflect student and college success. There is great distortion when conventional calculations are used to describe non-traditional students. Narrow time intervals and failure to include transfers tend to accentuate differences among institutions to the advantage of elite institutions and detriment of others. Open access institutions stand to gain the most when college completion is more broadly conceived and more flexible parameters are used to define graduation rates. (Contains 24 references.) (JLS)