The aim of using data mining in the education field is to enhance educational performance, by using the six school indicators, which are defined in this article to identify useful guidelines that can improve school performance. Knowledge discovery requires a clear methodology that can be successfully applied in the education sector. This can be obtained from the use of the CRoss-Industry Standard Process for Data Mining (CRISP–DM). The CRISP-DM was used in this article to implement data mining for knowledge discovery from the database of 104 schools. Three methods of data mining, Naïve Bayes, Nearest Neighbor and the C4.5 decision tree, are implemented on the school data. The results showed that the C4.5 decision tree is significantly more accurate compared with the other methods.