Learning continually is the trends of the world. The number of night school students has a certain portion in technology colleges and universities. The paper is applied Kano model to understand demand of night school students. According to different students' attributes, we tend to understand their variances in education quality demand. Students are clustered by their grades of satisfaction indicators. The research findings are important information to education improvement and to increase student learning quality. There are significant difference among different levels' students are also caused different classification of quality attributes.