‘The Effectiveness of Peer Assessment to Teach Descriptive Writing for High and Low Proficiency Students: The Case at The English Department Students of Pancasakti University Tegal in The Academic Year of 2014/2015’

This experimental study attempts to investigate the effectiveness of peer assessment to teach writing descriptive essay in two levels of students (high and low writing proficiency level). 48 students out of 150 students at the third semester students of the English Department of Pancasakti University Tegal were taken as sample. They were grouped into control (n=24) and experimental class (n=24). Each group was categorized into two writing proficiency level, namely high and low levels. The methods of collecting the data were writing test, questionnaire, and classroom observation. The data were analyzed quantitatively and qualitatively. Two way ANNOVA was made use to analyze the quantitative data. The findings from quantitative data revealed that there was no significant difference between control and experimental class (p=0.856>0.05). Therefore, the Null hypothesis stating that peer assessment does not give more significant effect on students’ writing performance than teacher’s existing method is accepted. In addition, there was no interaction among peer assessment, students’ level, and writing competence (p= 0.968>0.05) and peer assessment worked in high (M=84.7) and low (M=69.4) level of students. On the other hand, the qualitative findings from classroom observation showed that peer assessment was difficult for student to carry on; however, from questionnaire, students showed positive attitude towards peer assessment. Finally, the qualitative analysis from students’ writing indicated that grammar was the most difficult aspect for students to master