Aplikasi Jaringan Bayes pada Pembuatan Butir Soal Tes

  • Wahyu Hartono Universitas Swadaya Gunung Jati
  • Tonah Tonah``
Keywords: Fixed test; Adaptive test; Bayes network

Abstract

The course of differential calculus is essential because it is a prerequisite material in most classes at the next level. From experience, most of the students have not been able to master the prerequisite topic. These conditions will disrupt the teaching and learning process. Information about the students' initial knowledge will be useful for applying appropriate learning models. This research describes Bayes network application on the manufacture of items about the fixed and adaptive test related to differential calculus courses. The research method is an experiment. The sample used is the students of mathematics education program as many as 98 students who already finish differential calculus course. The results showed that the performance of adaptive test design in predicting student ability is better than fixed test design, especially after the fifth question. The performance of the fixed test items sorted from easy to difficult is better than other fixed test designs. This study is useful for making diagnostic test questions in mapping/predicting students' initial knowledge as well as evaluating their abilities. The suggestion for further research is to make the performance of fixed test design is equivalent to adaptive test in diagnostic capability.

Downloads

Download data is not yet available.

References

[1] Almond et al., Models for conditional probability tables in educational assessment, In Proc. of the 2001 Conference on AI and Statistics. Society for AI and Statistics.
[2] Conati, Cristina et al., On-line student modeling for coached problem solving using Bayesian networks. In Anthony Jameson, Cecile Paris, and Carlo Tasso, editors, Proc. of the Sixth Int. Conf. on User Modeling (UM97), Chia Laguna, Sardinia, Italy (1997) pages 231–242. Springer Verlag.
[3] Hugin Explorer. 2017. ver. 6.0. Computer Software. http://www.hugin.com.
[4] Lauritzen, SteffenL,Graphical Models. Clarendon Press, Oxford. (1996)
[5] Millan, Eva dan P´erez-de-la-Cruz, Jos´e Luis, A Bayesian diagnostic algorithm for student modeling and its evaluation. User modeling and User- Adapted Interaction,12 (2–3): (2002) 281–330.
[6] Mislevy, Robert J et al., Computerized Adaptive Testing: A Primer. Mahwah, N. J., Lawrence Erlbaum Associates, second edition. (2000)
[7] Vomlel, Jiri, Bayesian Networks in Educational Testing. International Journal of Uncertainty, Fuzziness and Knowledge Based Systems, Vol. 12, Supplementary Issue 12(004) pp. 83-100. A draft version.
Published
2018-05-15
Section
Articles