PERBANDINGAN METODE PENYETARAAN SKOR TES MENGGUNAKAN BUTIR BERSAMA DAN TANPA BUTIR BERSAMA

Heri Retnawati

doi:10.21831/jk.v46i2.10383

PERBANDINGAN METODE PENYETARAAN SKOR TES MENGGUNAKAN BUTIR BERSAMA DAN TANPA BUTIR BERSAMA

Heri Retnawati, (Scopus ID: 56896145400) Department of Mathematics Education, Universitas Negeri Yogyakarta, Indonesia

10.21831/jk.v46i2.10383

Abstract

Abstrak

Penelitian ini bertujuan untuk mengetahui kesetaraan skor 20 perangkat tes ujian akhir SMP dan membandingkan penyetaraan dengan butir bersama (equating) dan tanpa butir bersama (concordance). Metode penelitian yang digunakan adalah metode rerata dan rerata, metode rerata dan sigma, dan Haebara, Stocking dan Lord. Objek penelitian ini adalah perangkat tes ujian akhir SMP Mata Pelajaran Matematika Tahun 2014 dan 46.313 respons siswa. Estimasi parameter butir dilakukan dengan program QUEST dan penyetaraan dilakukan dengan program IRTEQ. Interpretasi hasil penyetaraan dilakukan dengan membandingkan kurva karakteristik tes dan mengestimasi kesalahan penyetaraan dengan root mean square of error (RMSE). Hasil penelitian menunjukkan bahwa pertama, dua puluh paket yang digunakan pada ujian nasional menunjukkan kecenderungan yang setara. Kedua, pada equating dengan metode grafis, rerata dan sigma menghasilkan skor paling setara. Ketiga, metode Haebara dan metode Stocking dan Lord yang menghasilkan skor-skor dengan RMSE yang paling kecil. Keempat, concordance menghasilkan RMSE yang lebih kecil dibandingkan equating.

Kata kunci: penyetaraan, concordance, equating, RMSE

THE COMPARISON OF TEST SCORES LINKING METHOD

USING EQUATING AND CONCORDANCE

This study was aimed at determining the linking score of 20 tests of the national examination and comparing test score linking methods using equating and concordance. This study used mean and mean, mean and sigma, Haebara, and Stocking & Lord methods. The objects of this study were mathematics national examination tests of junior high schools in 2014 and 46,313 students’ responses. The estimation of item parameters was done using the QUEST program while the equating used the IRTEQ program. The interpretation of the results was done by comparing the test characteristic curves and estimating the linking error of the Root Mean Square Error (RMSE). The results show that (1) 20 sets of tests in the national exams show equal tendencies; (2) in equating with graphical methods, the means and sigmas produce the most equal scores; (3) Haebara and Stocking & Lord methods generate the smallest RMSE scores; and (4) the concordance produces RMSE smaller than equating.

Keywords: linking, concordance, equating, RMSE

Full Text:

PDF

References

Anderson, B., Braunberg, K, Wiberg, M. (2013). Performing the Kernel Method of Test Equating with the Package Kequate. Journal of Statistical Software.55(6). 1-25.

Antara, A.A.P. &Bastari.(2015). Penyetaraan vertical dengan pendekatan klasik dan item response theory pada siswa sekolah dasar.Jurnal Penelitian dan Evaluasi Pendidikan. 19(1), 13-24.

Aşiret, S., & Sünbül, S.Ö.(2016). Investigating test equating methods in smallsamples through variousfactors.Educational Sciences: Theory & Practice, 16, 647-668.

Brennan, R.L.& Kolen, M.J. (2004). Concordance Between ACT and ITED Scores From Different Population. Jurnal Applied Psichological Measurement, Vol 28. No. 4, July 2004, 219-226.

Dorans, N.J. (2004). Equating, Concordance and Expectation. Jurnal Applied Psichological Measurement, Vol 28. No. 4, July 2004, 219-226.

Dorans, N.J., Moses, T.P, Eignor, D.R. (2010). Principles and Practices of Test Score Equating Research Report. http://www.ets.org/research/contact.html.

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

Hambleton, R.K. &Swaminathan, H. (1985). Item response theory. Boston, MA: Kluwer Inc.

Han, K. T. (2009). IRTEQ: Windows application that implements IRT scaling and equating [computer program]. Applied Psychological Measurement, 33(6), 491-493.

Kim S.H. & Cohen, A.S. (2002). A comparison of linking and concurrent caliberation under graded response model. Applied Psychological Measurement. 26(25-61).

Kolen, M.J. dan Brennan, R.L. (2004).Test Equating : Methods and Practices. New York : Springer.

Lumapow, H. (2012). Identifikasi Materi Sulit Ujian Nasional Bahasa Inggris Pada Siswa Jurusan Bahasa.Jurnal Kependidikan, 42(1), p. 61 - 75

Mardapi, D. (1998). Analisis Butir Dengan Teori Tes Klasik dan Teori Respons Butir. Jurnal Kependidikan, 1(28).

Masruri, M.S. & Nurhadi.(2007). Peningkatan kualtas pembelajaran mata kuliah penilaian dan pencapaian belajar geografi melalui penerapan model portofolio. Jurnal Kependidikan.37(2), 167-186.

Moses, T. & Liu, J. (2011). Smoothing and Equating Methods Applied to Different Types of Test Score Distributions and Evaluated With Respect to Multiple Equating Criteria. Research Report. http://www.ets.org/research/contact.html

Pang, X., Madera, E., Radwan, N., Zhang, S. (2010). A Comparison of Four Test Equating Methods Research Report.www.eqao.com.

Retnawati, H.& Hidayati, K.(2007).Perbandingan metode concordance berdasarkan teori tes klasik.Laporanpenelitian. Lembaga Penelitian UNY Yogyakarta.

Retnawati, H. (2014). Teori respons butir dan penerapannya.Yogyakarta: Parama.

Ryan, J. & Brockmann, F. (2011). A Practitioner’s Introduction to equating with Primers on Classical Theory and Item Respons Theory. Research Report.

Uysal, İ. & Kilmen, S. (2016). Comparison of Item Response Theory Test Equating Methods for Mixed Format Tests. International Online Journal of Educational Sciences, 2016, 8 (2), 1-11.

Yu, C.H. & Popp, S. E.O. (2005). Test Equating by Common Items and Common Subjects: Concepts and Applications. Practical Assessment, Research & Evaluation. 10(4), 1-19.

DOI: https://doi.org/10.21831/jk.v46i2.10383