Principal component analysis and correspondence analysis are used to classify the 96 British universities into three categories. With different input information, the two methods provide similar results. For the input of correspondence analysis, we categorize each 14 criteria values into two categories and construct a binary table. We also separate each of the criteria values into three and four categories and the results are robust to the number of categories. We find that the results are not due to the high degrees of correlation among the criteria values. Surprisingly, there seems to be no loss of information in categorizing the continuous data. This shows that correspondence analysis is useful in the multi-criteria decision making problem for the case of categorical criteria values. In addition, the technique provides a simultaneous graphical representation of alternatives and criteria. This can be used as an aid to the decision maker in understanding the structure of the problem.
