Abstrak

The main objectives of this research are to investigate whether: (1) Clustering Technique is more effective than direct instruction to teach writing; (2) the students having high creativity have better writing skill than those who having low creativity; and (3) there is an interaction between teaching techniques used to teach writing and students’ creativity.
   A quasi experimental research was applied in this study. The population of this research was the tenth grade of SMAN 4 Madiun in the academic years of 2016/2017. The sampling used in this research was cluster random sampling. The sample of this research was 56 students including 28 students of experimental class and 28 students of control class. Each class was divided into high and low levels of creativity. To collect the data, there were two instruments used namely writing test and verbal creativity test. After treatment was given to both classes in eight meetings, the researcher conducted a post-test of writing to get the data. The data were analyzed in terms of their frequency distribution, normality, and homogeneity. Then 2x2 Multifactorial ANAlysis of Variance (ANOVA) and Tukey test were used to test the research hypothesis.
The result of this study reveals that: (1) clustering technique is more effective than direct instruction to teach writing; (2) students having high creativity have better writing skill than those having low creativity; and (3) there is an interaction between teaching techniques and students’ creativity. It shows that the use of clustering technique is effective to teach writing for the tenth grade of SMAN 4 Madiun. Therefore, it is recommended for English teacher to apply clustering technique in writing class, to promote an effective teaching writing.

Keywords : Clustering Technique, Direct Instruction, Writing, Creativity