In the pre-test, the grade mean of 40 participating treatment students is 80.96, while the mean of 46 participating control students is 80.95. The mean difference 0.01 is statistically very trivial (p = 0.997 > 0.05). Thus the treatment and control class were equal in vocabulary knowledge before the experiment. In the post-test, the grade mean of 39 participating treatment students is 92.27, while the mean of 41 participating control students is 87.58. The mean difference 4.69 is statistically significant (p = 0.04 < 0.05). Because the vocabulary test content in the post-test was the same as that in the pre-test, the longitudinal grade compare of either the treatment group or the control group is meaningful. The average grade of the treatment students in vocabulary test was increased (11.31) more significantly than that of the control students (6.63) throughout the experiment. In order to compare the effect size of both classes, the Cohen’s d based on sample size considering Hedge’s adjustment is calculated [13–15]. It is 0.261 for the treatment group, while 0.167 for the control group.