The purpose of this thesis was to find out the more desirable and effective teaching method of sight singing through the experimental research fir the improvement of high school stufents` sight singing ability. The main content of this research was to determine which group is better in sight singing ability between a group taught on the basis of the meterials provided systematically and a group taught in other way-not by systematical materials but by only a text book. In order to conduct this experiment, 228 students of the 1st grade of C High School were chosen as a research group and were pretested right after being entered the school in March. The content of the pretest consisted of rhythm sight siging, melody sight singing, melody ear training, and singing test. In the test method, rhythm sight singing and melody sight singing were tested individually in a separate room for the reliability of the test but melody ear training and singing test were done collectively in a classroom. The research group was unconditionally divided into two parts : one was the experimental group and the other was the control group. These groups have been taught respectively with a difference in learning materials and teaching methods through regular classes twice a week. Next the research group was post-tested in the same content and method as those of pretest. T.Test method and Pearson Correlation Test were applied in order to examine whether or not the average gap regarding the development rate of each variable in each group was meaningful and what the relationship between each variable was like. As, the result, conclusions were drawn as follows ; First, the sight singing ability of the experimental group attaching importance to the logical thinking process of music on the basis of essential pulse and tonic in sight singing was higher than that of the control group depending on only a text-boor. Second, the ear training ability of the experimental group attaching importance to the logical thinking process of music on the basis of essential pulse and tonic in sight singing was higher than that of the control group depending on only a textbook. Third, as the sight singing ability of the experimental group improved, the singing ability of the experimental group was much higher than that of the control group. Fourth, in interrelationship of sight singing ability, ear training ability, and singing ability of each group, the coefficient of correlation in each variable of the experimental group was higher than that of the control group. And the same in both groups, sight singing ability and ear training ability showed the hightest relationship, next was the relationship of ear training ability and singing ability showed the lowest. In conclusion, in order to improve sight singing ability, students should be taught systematically and sequentially by Solffeggio learning method from elementary school and should be made to approach music gradually through the logical thinking process of music. For the purpose of this, systematic and sequential musical notations for sight singing should be included in the text-book form elementary school to high school. In teaching method, it is necessary that students should be taught pitch recall practice and ear training practice according to the degree of interval on the basis of rhythm learning, putting great emphsis on the concept of pulse.