This paper investigates the problems of the impact of grouping on measuring academic performance. It focuses on inferences made from the group level to the individual level. Data for a large state college system with over 30 individual colleges is used. The problem of aggregation bias is studied using the analysis of covariance and is related to the clustering approach of generalized least squares. To illustrate the question, actual data on academic performance at the individual and group level is explored. The analysis is done by gender and minority group status. It is found that the individual level relationships and the college level relationships were generally quite different, with regression coefficients often having different signs. Based on the results of this research it is concluded that it would be generally inappropriate to use grouped data to investigate academic performance across colleges. (JAZ)