An item analysis of english summative test at the third year students (a descriptive analysis at the third year students of ?SMPN 15? Bekasi in the Academic Year 2012/2013)
Main Author: | Siti Fiqhi Rasnawati |
---|---|
Format: | Bachelors |
Terbitan: |
Fakultas Ilmu Tarbiyah dan Keguruan
|
Subjects: |
Daftar Isi:
- This study examines the test items that have been tested toward the third students of ?SMPN 15? Bekasi at odd semester in the academic year 2012/2013. The objective of this study is to analyze the quality of test items of English summative test and to find out the items which have qualified as good test items in terms of difficulty level, discriminating power, and distractors effectiveness. The method used in this study is descriptive analysis using quantitative approach. It describes the existing information concerning the test items quality of English summative test by analyzing the three calculations of item analysis. The data was obtained from school?s documentation, namely the students? answer sheets and the final odd English summative test of third class in the academic year 2012/2013. The English test is divided into two parts; part 1 consists of 40 multiple choice items and part 2 consists of 5 essay items. The data were analyzed manually by using some formula of each point of item analysis. Therefore, it also uses some data tabulation to interpret the data explicitly. The findings of this study showed that the English summative test of the third year of ?SMPN 15? Bekasi has low quality. Most of the test items haven?t fulfilled the criteria of good difficulty level and discriminating power, and distractors effectiveness. Almost the entire test items are easy (76%), poor (58%), and ineffective distractors (51%); thereby those items should be revised. To conclude that the difficulty level of each test item is poor, the discrimination level of each test item is poor, and the distractors in the multiple-choice items are ineffective. Through this study, it is expected that the teachers or test makers generally know the quality of good test items through item analysis so that they can compose a test which has good quality