Modeling Of Context-Based Adaptive Assessment: Case Study On Mathematics In Elementary School

Yuhana, Umi Laili (2019) Modeling Of Context-Based Adaptive Assessment: Case Study On Mathematics In Elementary School. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

Text 07111660010016-Doctoral.pdf Restricted to Repository staff only Download (3MB)

Abstract

Saat ini, kebanyakan institusi pendidikan menggunakan kurikulum berbasis kompetensi. Di Indonesia, sejumlah 148.244 sekolah dasar menggunakan kurikulum berbasis kompetensi. Ada 25.486.506 siswa di sekolah tersebut pada tahun ajaran 2017/2018. Besarnya jumlah siswa menuntut sekolah untuk menyiapkan alat yang cocok untuk menilai capaian kompetensi. Siswa di sekolah dasar berada pada enam tingkatan kelas. Kemampuan siswa pada kelas yang sama dapat berbeda-beda. Informasi tentang capaian dan ketidakcapaian kompetensi siswa merupakan hal yang penting untuk membantu guru merancang target dan proses pembelajaran yang tepat. Sistem penilaian adaptif dapat digunakan untuk mengukur kemampuan siswa. Sistem adaptif dapat memilih pertanyaan yang sesuai untuk setiap siswa. Namun, tingkat kesulitan soal pada sistem penilaian adaptif saat ini secara umum sama dan tidak spesifik menilai kompetensi. Pada faktanya, dengan kemampuan siswa yang berbeda, persepsi terhadap kesulitan soal yang sama dapat berbeda. Karenanya, tingkat kesulitan soal, tingkat kemampuan awal dari siswa, dan pemilihan soal seharusnya didefinisikan secara adaptif sesuai dengan konteks siswa. Konteks dalam disertasi ini mencirikan situasi dari siswa dan pertanyaan, seperti tingkat kemampuan siswa, kompetensi yang diujikan, dan tingkat kesulitan soal. Sistem penilaian adaptif berbasis konteks bertujuan untuk menilai dengan tepat capaian kompetensi siswa. Tujuan dari disertasi ini adalah memodelkan penilaian adaptif berbasis konteks yang dapat mengukur kompetensi siswa. Ada empat kontribusi utama dalam disertasi ini, yaitu model untuk menentukan tingkat kesulitan soal secara adaptif, model untuk mengestimasi tingkat kemampuan awal siswa, model untuk pemilihan soal berbasis konteks siswa, dan model untuk mengidentifikasi kompetensi siswa. Disertasi ini menggunakan mata pelajaran matematika di sekolah dasar sebagai studi kasusnya. Penelitian ini sangat berguna sebagai pengetahuan awal dan desain perangkat lunak dalam mengembangkan sistem penilaian berbasis konteks.
==================================================================================================================================
Recently, most education systems utilized a competency-based curriculum. Around 148,244 elementary schools in Indonesia implement the competency-based curriculum. There are 25,486,506 students in the schools for the academic year 2017/2018. This vast number of students requires the school to prepare a suitable tool for assessing the achievement of competency. The students are distributed to the six grade levels. The ability of the students at the same level could be different. The information about the achieved and the lack of student's competencies is essential to help the teacher design the learning target and process. An adaptive assessment system often used to measure the ability of student. The adaptive assessment system can select suitable questions for each student. Unfortunately, the difficulty level of questions in the current adaptive assessment system is defined generally and unspecify to assess the competencies. In fact, with a different ability, perception of the difficulty level of the same question between students could be different. Therefore, the difficulty level of the question, the initial ability level of the student, and the selected item should be defined adaptively based on the student’s context. The context in this dissertation characterizes the situation of student and question, such as the ability level of the student, the tested competency, and the difficulty level of question. The context-based adaptive assessment system aims to assess how precisely the achievement of the student’s competency. The objective of this dissertation is to model the adaptive assessment based on the context that can measure the competencies of the student. There are four main contributions of this dissertation, i.e., the model for determining the adaptive difficulty level of question, the model for estimating the initial ability level of the student, the model for selecting the item based on the student's context, and the model for identifying the student competency. This dissertation used the mathematics subject in elementary school as the case. This research would be useful as the preliminary erudition and software design in developing context-based adaptive assessment system.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	Adaptive assessment system, context-based modeling, difficulty level of question, initial ability level of the student, item selection, software design.
Subjects:	T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming
Depositing User:	Yuhana Umi Laili
Date Deposited:	23 Jul 2026 08:20
Last Modified:	23 Jul 2026 08:23
URI:	http://repository.its.ac.id/id/eprint/70325

Actions (login required)

View Item