Many arguments on learning efficiencies for Roman scripts and Han characters were ideologically arose, rather than based on a scientific study. Although Chiung (2003) had done a scientific study in learning efficiencies based on experiments with a total of 803 subjects, the amount was limited. The purpose of this study is to feature out the leaning efficiencies of Han characters based on Chiung's previous study and methodology. A total of 105 junior nigh school students and 210 senior high school students were added in dictation tests. A total of 187 elementary school students and 30 collegians were added in oral reading tests. The results reveal that Roman scripts are superior to Han characters firstly in dictation tests and secondly in oral reading tests. In dictation tests, the subjects in Roman scripts group significantly increased their score each year until the fifth grade, by which time they had the same statistical score as college students. As for subjects in Han characters group, subjects did not statistically reach the same level as college students until they were a round at tenth grade. In oral reading tests, the second grade students in Roman scripts group and third grade students in Han characters significantly reached the statistical score as collegians.