ENHANCING STUDENTS’ READING COMPETENCE THROUGH WEB-BASED COLLABORATIVE INQUIRY LEARNING: A STUDY ON MECHANICAL ENGINEERING STUDENTS AT THE STATE UNIVERSITY OF SEMARANG

Abstract

This study is to investigate: (1) the extent of reading competence when students were taught with and (2) without Web-Based Collaborative Inquiry Learning (WBCIL), also (3) whether there is a significant difference between the students taught with and without WBCIL. At the first stage, a reading test try out was administered two times to measure how proper the test to be good instrumentation. The value of good test was measured through item difficulty, item discrimination, the validity and reliability of the first and second try-out test. After the test was determined to be proper instrumentation, the samples of the study were selected. They were 91 students of Mechanical Engineering Department majoring Education of Automotive Engineering. The third stage was deciding experimental and control class to have pre-test. Yet both of the class did not receive the same model of English teaching. The experimental class experienced WBCIL while the control class received common teaching model. Both of them were taught for 8 effective meetings. Post Test was conducted at the last meeting (the 8th meeting) after review session. There were two indicators to measure the effectiveness of WBCIL model applied in this study. The indicators were the result of Independent Samples T-Test and Two-Way-Analysis of Variance. The result of Independent Samples T-Test indicated higher t_value compared to t_table which showed a significant difference between two groups. The t_value was 4.077 and the t_table was 1.662. Two-Way-ANOVA also yielded valid output interpretations such as Test of between Subjects-Effects, Multiple Comparisons, and Profile Plots of Estimated Marginal Mean Score between groups and test achievers. Both control and experimental groups achieved improvement at Post Test, but only experimental groups achieved better improvement with 8.37 points higher. By applying WBCIL model, slow and medium achievers got better Post Test scores in experimental group, but it was less than optimal for the high achievers. The high achievers in experimental group achieved lower score compare to control group. Obviously, collaborative group work used in WBCIL model brought benefit to slow and medium achievers, but weakened the achievement of high achievers.