The present study was designed to develop and standardize the Korean version of Children’s Eating Behavior Inventory (CEBI). The parents of preschool and elementary school children ranging in age from 3 to 10(N=713) and those of children showing significant growth failure as well as eating problem(N=96) administered CEBI, and some of them also completed K-CBCL(N=300) for validationg CEBI. After correlation analysis, 27 items out of 40 were selected. Based on the exploratory and confirmatory factor analysis along with theoretical assumption, 5-factor structure was selected: prevasive problematic eating attitude, parent stress, problem behavior during mealtime, eating prohibited food, and helping prepare mealtime. internal consistency and test-retest reliability demonstrated that CEBI was a reliable instrument. The significant correlation between K-CBCL and CEBI and differences between clinic and normal group showed that CEBI was also a valid instrument to examine children’s problematic eating behavior. The result of two-way ANOVA showed that there were significant differences between preschooler and school-aged children in problematic eating behavior. Clinical and research implications and limitaions are discussed.