The study of this research is quantitative and the main objective of this study is to find out whether there is or not a significant difference in vocabulary score of student taught using comic stories and those taught using non-comic stories. The method of the research is experimental study. The data were obtained by giving test to the experimental class and control class after giving a different learning to both classes. The researcher gave two times teaching to both classes. The number of the students is thirty-five in experimental and control class and thirty-nine in try-out class. They are VIII E is as experimental class (the students who are taught using comic stories), VIII D is as control class (the students who are taught using non-comic stories) and VIII C is as try-out class. The instrument to collect the data were; observation and test. Observation was only used to support the data about students’ imagination on reflected on their engagement in learning processes. Test was used to know students’ competence before and after the experimental run. There are two kinds of test. They are pre-test and post-test. Before items of the test were given to the students, the researcher gave try-out test to analyze validity, reliability, difficulty level and the discriminating power of each item.

After the data collected, the researcher analyzed it. The first analysis is from beginning of experimental class and control class that is taken from pre-test value. It is the homogeneity test and t-test. But, before calculating the homogeneity test and t-test, the researcher calculated mean, variance and standard deviation. It is used to know that two groups have same variant or not. Another analysis data is from ending of experimental class and control class. It is used to prove the hypothesis that has been planned. The result of the research: the mean of the vocabulary score of experimental class (the students who are taught using comic stories) are 79.57 and the mean of control class (the students who are taught using non-comic stories) are 74.57. From that result, there is a difference in vocabulary score between students taught using comic stories and students taught using non-comic stories. It is showed of the mean of experimental score is higher than the mean of control class (79.57>74.57). On the other hand, the test of hypothesis using t-test formula showed the value of the t-test is higher than the value of t-table. The value of the t-test is 2,328, while the value of t-table on α = 5% is 1,66 (2.328>1.66). The hypothesis is accepted. Based on the result of this study, it is expected in teaching vocabulary. So, that can produce students who master English well.