PENGEMBANGAN COMPUTERIZED ADAPTIVE TESTING (CAT) MENGGUNAKAN METODE POHON SEGITIGA KEPUTUSAN

Winarno Winarno

doi:10.21831/pep.v16i2.1132

PENGEMBANGAN COMPUTERIZED ADAPTIVE TESTING (CAT) MENGGUNAKAN METODE POHON SEGITIGA KEPUTUSAN

Winarno Winarno, Sekolah Tinggi Agama Islam Negeri (STAIN) Salatiga, Indonesia

10.21831/pep.v16i2.1132

Abstract

Penelitian ini bertujuan menghasilkan CAT menggunakan metode pohon segitiga keputusan dalam prosedur pemilihan item dan mengetahui kemampuan CAT dalam mengestimasi kemampuan peserta tes dengan tepat. Penelitian ini menggunakan Research and Development(R&D). Peng-ambilan data dengan observasi, dokumentasi, dan angket. Analisis data yang digunakan adalah teknik analisis deskriptif evaluatif dan teknik analisis deskriptif kuantitatif. Hasil penelitian adalah (1) CAT yang dikembangkan berdasar kebutuhan pemakai yaitu: berbasis internet, memiliki sistem ke-amanan, dan mudah diakses, (2) CAT dapat mengenali tiga pengguna, yaitu: administrator, guru, dan siswa, (3) CAT mampu memberikan butir-butir yang bersifat adaptif berdasarkan respon jawaban peserta tes. Secara keseluruhan kinerja CAT mampu melaksanakan tugas dengan baik untuk memilih butir tes dan mengukur kemampuan peserta tes dengan akurat dan tepat dilihat dari nilai korelasi antara hasil estimasi kemampuan (θ) dengan nilai ulangan murni (NUM) di sekolah siswa cukup tinggi yakni 0,67.

Kata kunci: metode pohon segitiga keputusan, metode maximum likelihood

______________________________________________________________

DEVELOPING COMPUTERIZED ADAPTIVE TESTING (CAT) BY USING THE TRIANGLE DECISION TREE METHOD

Abstract This research aims at producing a CAT software that uses the tri-angle decision tree method in the test item selection procedure and detecting the CAT ability in estimating the test-takers’ ability accurately and correctly. This research used the research and development approach (R&D). The data were collected through observation, documentation, inquiry, and the data were analyzed descriptively and quantitatively. The findings are as follows. (1) The CAT developed is:based on users’ need, web-based, user-friendly, interactive, highly secured, and easily accessible. (2) The CAT can recognize three different users: school administrators, teachers, and students. (3) The CAT software is able to provide adaptive items according to students’ answers. Overall, the CAT software can perform the tasks well to select test items and to measure test-takers’ability. It can be seen from correlation value between ability estimation result (θ) in CAT with nilai ulangan murni (NUM) students at school which is high enough t that is 0, 67.

Keywords: the triangle decision tree method, maximum likelihood method

Full Text:

FULL TEXT PDF

References

Borg, W.R. & Gall, M.D. (2003). Education research an introduction. New York: Von Hoffman Press, Inc.

Djemari Mardapi. (2008). Teknik penyusunan instrumen tes dan nontes. Yogyakarta: Mitra Cendikia Press.

Gronlund, N.E. (1976). Measurement and evaluation in teaching. New York: Macmillan Publishing Co.

Hambleton, R.K. & Swaminathan, H. & Rogers, H.J. (1991). Fundamentals of item response theory. Newbury Park, CA: Sage Publication Inc.

Lord, F.M. (1980). Applications of item response theory to practical testing problems. Hillsdale, NJ: Erlbaum.

Masters, N.G. & Keeves, P.J. (1999). Advances in measurement in educational research and assessment. New York: Pergamon Press.

Phankokkruad, M. & Woraratpanya, K. (2008). An automated decision system for computer adaptive testing using genetic algorithms. Ninth ACIS International Conference on Engineering, Artificial Intelligence, Networking, and Parallel, 655-660.

Wainer, H. (1990). Computerized adaptive testing: A primer. Hillsdale, NJ: Lawrence Erlbaum Associates, Publisher.

Weiss, D.J. (2004). Computerized adaptive testing for effective and efficient measurement in counseling and education. Measurement and Evaluation in Counseling and Development, 37, 70.

DOI: https://doi.org/10.21831/pep.v16i2.1132