The purpose of this study is to explore whether the Korean students can be classified into some distinct clusters based on their computer use characteristics and to investigate whether theses clusters differ in terms of their demographic characteristics, characteristics as a learner, and computer use patterns. To do this, two-stage cluster analysis were conducted to PISA 2006 Korean data using four computer use characteristics variables; computer use for entertainment, computer use for knowledge acquisition/transfer, self-confidence in general ICT tasks, and self-confidence in high-level ICT tasks. The results indicated that three distinct patterns of ICT use exist for the Korean students; low-use/low-confidence group(cluster 1), low-use/ high-confidence group(cluster 2) and high-use/high- confidence group(cluster 3). The subsequent analyses on these clusters revealed that the male students attending rural-area vocational high schools have high probabilities of belonging to the low-use/low-confidence group(cluster 1). The students of this cluster were turned out to be the ones with the lowest achievement and the most negative attitudes towards science. Also, high proportion of these students reported that they never use computers at schools in spite of the fact that these students are more likely to attend the vocational high schools. These results suggest the needs for systematic intervention towards these disadvantaged group of students, including the more opportunities for computer use.