Irmawati, 2017. The Effectiveness of Clustering Technique to Improve Students’
Ability in Writing Analytical Exposition. Thesis English
Education Department, Faculty of Tarbiyah and Teachers
Training. Advisors :
(I) Dr. Hj Nida Mufidah, M,Pd, (II) Afifah Linda Sari, S.S, M. Pd
Keywords: Effectiveness, Clustering Technique, Writing, Analytical Exposition.
This research was conducted to find whether the use of clustering
technique as pre-writing stage is more effective than the traditional technique. The
problem statements of this research: (1) how is the effectiveness of clustering
technique to improve students’ ability in writing analytical expositions in
experiment class? (2) how is the effectiveness of teaching writing which taught
without using clustering technique to improve students’ ability in writing
analytical expositions in control class? (3) Is there significant difference in using
clustering technique between experiment class and control class.?
This research is quasi experiment research which the data analyzed
quantitative. The subject of this research are 31 students of the eleventh grade of
MA Thalabul Irsyad Sei Tatas Kapuas regency divided in two classes: XI A
(science program) as Experiment Class are 19 students and XI B (social program)
as Control Class are 12 students. The instruments of collecting data were test (pre
test and post test) to measure students’ ability, t test is used in order to find the
effectiveness of clustering technique in writing analytical exposition on
experiment and control class.
The result of this research shows that: (1) effectiveness of clustering
technique to improve students’ ability in writing analytical exposition in
experiment class was 83,06 that classified into very good category and had
increasing 40,28. While students’ ability in control class was 67,91 that classified
into fair category and had increasing 21,24. The difference of increasing
percentage from post test in experiment 89,9 was bigger than in control class with
33,3. (2) while pre test and post test the students’ ability without using clustering
technique in control class was 46,67 that classified into fair category, and post
test in control class was 67,91 that classified into good category. (3) there is
significant difference in the effectiveness of clustering technique between
experiment class and control class was T table (5% = 1,05) <T test (2,41) > t table
(1%=1,13) therefore Ha is accepted and Ho is rejected.