Using the Korean Labor and Income Panel Study(KLIPS) data of 1,767 households in 140
gu/gun/small cities in Korea, this study investigated how household and regional factors influence the
trends of private tutoring expenditure. To accomodate the nested structure of the data and research
questions, the study used the Hierarchical Linear Modeling(HLM) methods to estimate regional effects
on private tutoring expenditure. We found that households and regions varied in terms of initial
average expenditure in 2000 and annual expenditure rate of private tutoring. Regions and households
both contributed significantly to the expenditure of private tutoring of students. At each of household
and region levels, annual household income, total number of children, social status and the location
of each household had the significant strong effects on the private tutoring expenditure. Suggestions
and implications of conducting a future study on private tutoring expenditure of each household are
also discussed.