The Effectiveness of the Use of Scientific and Scaffolding Procedures in Teaching Writing Skills for Grade VII Students at SMP N 1 Brebes.

This study aimed to reveal: (1) the effectiveness of the use of scientific procedure, scaffolding procedure and deductive procedure in teaching writing skills, (2) whether the use of scientific procedure was more effective than the use of scaffolding procedure in teaching writing skills, (3) whether the use of scientific procedure was more effective than the use of deductive procedure in teaching writing skills, (4) whether the use of scaffolding procedure was more effective than the use of deductive procedure in teaching writing skills, and (5) whether the use of scientific procedure was the most effective compared to the use of scaffolding procedure and deductive procedure in teaching writing skills. This study was quasi-experimental using pretest-posttest control group design. There were two variables involved: the use of scientific procedure, scaffolding procedure and deductive procedure in teaching writing skills as the independent variable and students’ writing skills as the dependent variable. The population was all students of grade VII consisting of seven classes in the even semester of the academic year of 2016/2017. The sample was students of VII D, E and G classes determined by using simple cluster random sampling technique. The students of VII E class was the first experimental group, VII G class was the second experimental group, and VII D class was the control group. The data were collected using a test. The instrument of this study was the writing test in the form of an essay. The validity of the instrument was content validity which was obtained through the consultation with experts (expert judgment). The inter-rater technique was used to test the reliability of the instrument with Pearson Product Moment Correlation. Normality test and homogeneity test were performed as the pre-requisite analysis of the test. Kolmogorov- Smirnov was used to test normality and Levene test was used to test homogeneity. Data analysis technique used was the covariant analysis (ANCOVA) test followed by Scheffe test supported by SPSS 16.0 for windows program. The results of this study indicated that: (1) the use of scientific procedure, scaffolding procedure and deductive procedure had significant difference in the effectiveness in teaching writing skills as indicated by the value of F was 17.527 and the value of (p) was lower than 0.05 (0.000 < 0.05), (2) the use of scientific procedure was more effective than the use of scaffolding procedure in teaching writing skills, indicated by the value of mean difference was 6.10000, (3) the use of scientific procedure was more effective than the use of deductive procedure in teaching writing skills, indicated by the value of mean difference was 12.00000, (4) the use of scaffolding procedure was more effective than the use of deductive procedure in teaching writing skills, indicated by the value of mean difference was 5.90000, and (5) the use of scientific procedure was the most effective compared to the use of scaffolding procedure and deductive procedure in teaching writing skills as indicated by the value of mean difference of the second experimental group (6.10000) and the value of mean difference of the control group (12.00000) which were positive