PENDETEKSIAN KEBERFUNGSIAN BUTIR PEMBEDA DENGAN INDEKS VOLUME SEDERHANA BERDASARKAN TEORI RESPONS BUTIR MULTIDIMENSI

Heri Retnawati

doi:10.21831/pep.v17i2.1700

PENDETEKSIAN KEBERFUNGSIAN BUTIR PEMBEDA DENGAN INDEKS VOLUME SEDERHANA BERDASARKAN TEORI RESPONS BUTIR MULTIDIMENSI

Heri Retnawati, Fakultas Matematika dan Ilmu Pengetahuan Alam UNY

10.21831/pep.v17i2.1700

Abstract

Studi ini bertujuan untuk mengidentifikasi keberfungsian butir pembeda (DIF) berdasarkan teori respons butir dengan model dua dimensi. Penelitian ini merupakan penelitian eksploratif dengan pendekatan kuantitatif. Data yang digunakan pada penelitian ini adalah data ujian nasional mata pelajaran metematika 2003 yang sudah terbukti memuat 2 dimensi (Badrun Kartowagiran & Heri Retnawati, 2007). Untuk mendeteksi DIF, digunakan indeks volume sederhana yang merupakan pengembangan dari indeks daerah bertanda. Pada penelitian ini peserta tes dikelompokkan berdasarkan jenis kelaminnya, kelompok siswa laki-laki dan siswa perempuan. Estimasi parameter butir dilakukan dengan bantuan BILOG dan TESTFACT, kemudian estimasi volume bertanda dan menggambar permukaan karakteristik butir dengan program MAPLE. Hasil analisis menunjukkan bahwa dari 28 butir yang dianalisis, sebanyak 2 butir berpotensi memuat DIF seragam dan 26 butir berpotensi memuat DIF yang tidak seragam.

Kata kunci: teori respons butir unidimensi, teori respons butir multidimensi DIF, UPD, SPD

______________________________________________________________

INDENTIFYNG DIFFERENTIAL ITEM FUNCTIONING USING SIMPLE VOLUME INDICES BASED ON MULTIDIMENTONAL ITEM RESPONSE THEORY

Abstract The aim of the studi is to identify differential item fungtioning (DIF) based on item response theory using bidimensional model. This study uses explorative quantitative approach. Data used in this study is a mathematics test of national exam 2003 data which has been proved to contain 2-dimensions (Badrun Kartowagiran & Heri Retnawati, 2007). The simple volume indices are used to detect DIF which is the development of the simple area indices. The testee are grouped based on their gender, groups of male students and female students. The item parameter estimation was performed with BILOG and TESTFACT, and then to estimatevolume and draw the surface characteristics curve was conduct with MAPLE. The results showed that of the 28 items analyzed, 2 items potentially contained uniform DIF and 26 items potentially contained a non-uniform DIF.

Keywords: unidimensional IRT, multidimensional IRT, DIF, UPD, SPD

Full Text:

FULL TEXT PDF

References

Ackerman, T.A., et. al. (2003). Using multi- dimensional item response theory to evaluate educational and psycholo- gical tests. Educational Measurement, 22, 37-53.

Angoff, W.H. 1993. Perspectives on diffe- rential item functioning methodology. Dalam P.W. Holland dan H. Wainer (Eds.), Differential item functioning. Hills- date, NJ: Erlbaum, Pp. 3 – 23.

Bolt, D.M. & Lall, V.M. (2003). Estimation of compensatory and noncompensa- tory multidimensional item response models using Marcov chain Monte- Carlo. Applied Psychological Measurement, 27, 395-414.

Camilli, G. & Shepard, L.A. 1994. Methods for identifying biased test items, Vol.4. London: Sage Publications, Inc.

Hambleton, R.K. & Rogers, H.J. 1995. De- veloping an item bias review form. From http://www.ericcae.net/ft/ta mu/biaspub2.htm March 10, 2007.

Hambleton, R.K., Swaminathan, H., & Ro- gers, H.J. 1991. Fundamentals of item response theory. London: Sage Publi- cations, Inc.

Hambleton, R.K. & Swaminathan. 1985. Item response theory. Boston, MA: Klu- wer Nijjhoff, Publisher.

Kartowagiran, B & Retnawati, H. (2007). Validasi Dimensionalitas Perangkat Tes UN 2003-2006. Laporan Penelitian. Lembaga Penelitian Universitas Ne- geri Yogyakarta.

Kartowagiran, B & Retnawati, H. (2008). The Probability Difference Indices and Empirical Sampling Distribution for DIF Indices for Identifying Item Bias In Multidimensional Item Res- ponse Theory. Paper on International Conference on Matemathics 3th on July 18th 2008.

Kartowagiran, B & Retnawati, H. (2009). Pengembangan Model Identifikasi Differential Item Functioning dengan Teori Respons Butir Multidi-mensi. Laporan Penelitian. Lembaga Penelitian Universitas Negeri Yogya-karta.

Retnawati, H. (2008). Identifying Item Bias Using the Simple Volume Indi-ces and Multidimensional Item Res- ponse Theory Likelihood Ratio (IRT- LR) Test. Paper on International Confe- rence on Matemathics 3th on July 18th 2008.

Mislevy, R.J. & Bock, R.D. (1990). BILOG

: Item analysis & test scoring with binary logistic models. Moorseville: Scientific Sofware Inc.

Reckase, M.D. (1997). A linear logistic multidimensional model for dichoto- mous item response data. In W.J. Linden & R.K. Hambleton (Eds), Handbook of modern item response theory (pp. 271-286). New York: Springer.

Spray, J.A., et al. (1990). Comparison of two logistic multidimensional item res- ponse theory models. ACT Research Report Series. United States Gover- ment.

Thissen, D. et al. (1993). Detection of dif- ferential item functioning using the parameters of item respons model. Dalam P.W. Holland & H. Wainer (Ed). Differential Item Functioning. Englewood Clifp, NJ : Lawrence Erlbaum. p.67-86.

Wilson, D., Wood, R. & Gibbons, R. (1984). TESTFACT: Test scoring and fullinformation item factor analysis. [Com- puter program]. Mooresville, IL: Ssi.

DOI: https://doi.org/10.21831/pep.v17i2.1700