The purpose of this study is to correct the specific actions and improve sport skiils as well as use the effective educaional experiences during the teaching - learning, for which students sport actions were recorded by video, played individually or in groups, ans individual athletic skills were objectively observed. To achieve the aims above, the researcher selected at random two classes for girs, who are in the frist grade in J middle school in Kyo-yang city, Kyoung Ki Do. learning contents are gymnastic exercise for young people. After experimental group made up of 41 subjects was taken by video camera, the reseacher trained them to observe their own actions objectively, showed good sport conduct and encouraged them to correct thier specific actions. On the other hand, the control group of 37 subjects listend to teacher\'s verbal explanation to correct their acions without observing their actions by video To measure consistent scorer reliability, two colleagues as well as the researcher observed VTR together during the pre-test. The check the observation of individual sport action, the picture and interpret of gymnasric exercise for ydung people were use. To see if the teaching method through VTR influenced the action during the researcher observed the differences of sport actions during the pre-test, and the mid-test, and post-test according to the the inceasing times of the test in experimental group and control group for the each group. The collected data were analyzed and the results are as follows. 1. Differences in the level of action during the pre-test, the mid-test, and the post- test. 1-1 . The mean of action 1,2,3,4,5,9,10 was high in the experimental group during the pre-test but the rest of the actions except actional didn\'t show the meaningful differences withing P<.05 level. In control the men of action 3,6,7,8 was higher than in experimental group but only action 6,7 showed the meaningful differences within P<.05 level All of the actions from 1 to P<.05 level From this it came to be clear that the experiment was mad under thesame conditions. 1-2 During the mid- test the experimenta group correctxd action 1,8 very well and control group correct action 10 very well. The mean of all of the action froml to 10 was higher in the experimental group than in the control group . the experimental group showed the meningful differences within P<.05 level. Therefore it might be concloudedthat correction of actions the experimental group better than in the control group. 1-3 During the post -test the mean of action 1,2,3,4,5,7,8 was higher in the experimental group than the control group. the experimental group showed the meaningful differences within P<.05 level. According to the experiment, the conclusion was may be reached, that the experimental group whose specific actions were correted by VTR improved sport actions much more than the control group whoses specific action were corrected just by teacher\'s explanation whthout watching VTR. 2.Differences of actions according the increasing times of the test. 2-1 Every action of the experimental group showed the significant effect . effect during the pretest - the mid test- the post. Expecially action 1,5,6 showed the meaningful effect during the experiment between mid-test and the post-test. this means the effect the experiment was remarkable. And the whore mean of the actidons was increased meaningfully during the mid- test and the post- test. 2-2 the control group had the increased effect during the pre- test the mid- test the post- test but action 2 showed the meaningful effect ducing the pre- test and the post test without any meaning ful differences between the pre- test and the mid-test. the means of action 5.7.8. was high according to the incresing times but the mean didn\'t makeany meaningful differences within p<.05 level.