The purpose of this study was to analyze the effectiveness of kindergarten experience through a follow-up on the 4 different models studied earlier by Rhee and Lee(1983). The subjects of the present study were 117 elementary school children in grade 4. Among the 117, 86 children were from the original study and 31 children without kindergarten experience were added as the control group. Data for the study were collected on intelligence, creativity, reading and math achievement, socio-emotional development, self-concept, behavioral characteristics and family background. ANOVA as used to analyze group differences and stepwise multiple regression was applied to determine the relative effectiveness of independent variables contributing to children's growth. The major finding of the study were as follows: 1. There were no significant differences in IQ among the 4 program models and between the experimental and contrl groups. The regression analysis indicated that intelligence and school readiness at kindergarten explained most of the variance for the children's present intelligence level. 2. There were no significant differences in creativity test scores among the 4 program models and between the experimental and control groups. School readiness and IQ at kindergarten explained most of the variance for the children's present creativity. 3. There were no significant differences in school achievement test scores among the 4 program models, but significant differences were found in math between the experimental and control groups(p＜.05). School readiness at kindergarten explained most of the variance for the children's reading and math ability. 4. There were singificant differences in socio-emotional development rating among the 4 program models (p<. 01), but no significant differences were found between the experimental and control groups. Father's occupational level and ratings on socio-emotional devleopmnet at kindergarten explained most of the variance for the children's present socio-emotional development. 5. There were no significant differences in children's self-concept test scores among the 4 program models and between the experimental and control groups. The HOME score and father's educational level explained most of the variance for children's self-concept. 6. There were singificant differences in recognition for special talent among the 4 program models(p<.01) and between the experimental and control groups(p<.01). Significant differences were also found in leadership among the 4 program models(p<.01).