This section addresses the second research questions: Are there any differences between the way of paper-pencil and software in the use of mind mapping? The homogeneity tests of software and paper and pencil show that these effect sizes varied significantly across studies since Q exceeded the critical value of the chi-square distribution, which means that the mean effect size was judged to be significantly heterogeneous. Hence we adopt the random effects models to calculate the effect sizes of software and paper and pencil (Lipsey and Wilson, 2001). On the contrary, the unknown usage adopts the fixed effects models to deal with the results. Thus, the final standardized weighted mean effect sizes of the fifty two (52) studies that measured students‘ achievement are presented in Table1.Generally speaking, students using software had the highest overall achievement according to the result of SE ESs.