Open AccessAnalysis on Achievement Test in Intensive English Program of IAIN Samarinda
Author(s) -
Sari Agung Sucahyo,
Widya Noviaoor
Publication year - 2018
Publication title -
fenomena
Language(s) - English
Resource type - Journals
eISSN - 2460-3902
pISSN - 2086-3632
DOI - 10.21093/fj.v10i2.1320
Subject(s) - test (biology) , reliability (semiconductor) , psychology , quality (philosophy) , mathematics education , achievement test , construct validity , psychometrics , standardized test , developmental psychology , paleontology , power (physics) , philosophy , physics , epistemology , quantum mechanics , biology
As one of the tests, achievement test has to be qualified. A qualified test will be able to give the information about teaching correctly. If the achievement test is less qualified, the information related to students’ sucesss to achieve the instructional objective will also be less qualified. It means the test has to meet the characteristics of a good test. In fact, there has not been any effort yet to identify the quality of the achievement test which is used in Intensive English program. It means the information of the test quality cannot be found yet. Therefore, researchers are interested in analyzing the quality of achievement test for students in Intensive English program of IAIN Samarinda. Design of this research belongs to Content Analysis. Subject of this research is English achievement tests and 28 to 30 students were involved in the process of try out. Data were collected through three steps. Data were analyzed based on validity, reliability, and item quality. Finding of the research reveals 60 % of the tests have a good construct validity justified by related theories. It was found 55% of the tests have a good content validity. Reliability coefficient of the first tests format is 0, 65 and the second tests format shows 0, 52. Calculation of item difficulty shows 68% of the test items were between 0,20 – 0,80. The estimation of item discrimination shows 73% of the test items were between 0,20 – 0,50. While calculation of distracter efficiency shows 65% of the distracters were effective to distract the test takers.