Background: Cooperative learning(CL) has been well received in all levels of teaching and subjects all over the world.The purpose of this paper was to investigate whether the two
methods of cooperative learning(TGT & Jigsaw) was superior to traditional teaching(TT) in developing students’self-efficacy, Taichi performance and lowering social anxiety when they were studying
24-style Simplified Taichi
literature , experimental method, questionnaire survey and mathematical statistics were utilized in this paper, and the data acquired were calculated by SPSS 19.0.
Participants:3 classes（96 individuals in all） composed of male freshmen of different majors took part in this experiment. 33 students inTGT class, 31 in Jigsaw class and 32 in TT class.
Pretest : self-efficacy and social anxiety were tested before intervention at the beginning of the first session.
Group division：in TGT and Jigsaw class participants were randomly divided into 6 groups according to their academic score of P.E. of previous semester，and each group comprised one high-, three or
medium-, and one low-academic performance student.
Intervention and duration:in 3 classes TGT, Jigsaw and TT were applied , the intervention lasting for 14 sessions (14 weeks).
posttest: in the last session, self-efficacy, social anxiety and Taichi perfomance were examined.
measuring tools:Taiji performance was assessed according to the examination standard by another teacher who hadn't participate in the experiment;Social anxiety was surveyed with Interaction
Anxiousness Scale(IAS) made by Leary in 1983; Self-efficacy was investigated with General Self-efficacy Scale(GSES) by Schwarzer in 1981.
The post-test of Taichi performances were significantly different(F=34.234**, P= .000) , multiple comparisons showing both of the CL methods(TGT-P= .000;Jigsaw-P=.000) were superior toTT，and that
there was no difference between TGT and Jigsaw; In light of self-efficacy, there was no significant difference among the 3 classes in the pre-test of self-efficacy, whereas in the pro-test there was
a significant difference among the 3 classes(F=34.217**, P=.000), multiple comparisons indicating TGT teaching class(P=.000) and Jigsaw(P=.000) were significantly better than TT and that meanwhile
Jigsaw was better than TGT(P=.005). Paired sample T test found the difference between pretest and protest in three classes were all significant; as for social anxiety, there was no significant
difference among the 3 classes in the pre-test of social anxiety, while in the pro-test there existed a significant difference among the 3 classes(F=60.972**, P=.000), multiple comparisons suggesting
TGT (P=.000) and Jigsaw(P=.000) were significantly better than TT. Paired sample T test showed the differences between pretest and protest in two classes(TGT & Jigsaw) were significant (TGT-T=9.68**,
P=.000; Jigsaw-T=10.178**,P=.000) and that there was no significant difference in TT(T=.780; P=.441).
CL was an effective way to boost students' Taiji performance ,self efficacy and lowering social anxiety compared to TT, but there was difference among varied CL methods too.