PENGEMBANGAN INSTRUMEN UNTUK MENGUKUR KOMPETENSI GURU SEKOLAH DASAR DALAM MENGEMBANGKAN KARAKTER SISWA

MAHATMA, . (2026) PENGEMBANGAN INSTRUMEN UNTUK MENGUKUR KOMPETENSI GURU SEKOLAH DASAR DALAM MENGEMBANGKAN KARAKTER SISWA. Doktor thesis, UNIVERSITAS NEGERI JAKARTA.

	Text 1_COVER- Revisi.pdf Download (808kB)
	Text 2_BAB 1.pdf Download (334kB)
	Text 3_BAB II.pdf Restricted to Registered users only Download (664kB) | Request a copy
	Text 4_BAB III.pdf Restricted to Registered users only Download (303kB) | Request a copy
	Text 5_BAB IV.pdf Restricted to Registered users only Download (696kB) | Request a copy
	Text 6_BAB V.pdf Restricted to Registered users only Download (218kB) | Request a copy
	Text 7_DAFTAR PUSTAKA.pdf Download (358kB)
	Text 8_LAMPIRAN_compressed.pdf Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Penelitian ini bertujuan mengembangkan dan menguji kualitas psikometrik instrumen kompetensi guru sekolah dasar dalam mengembangkan karakter siswa menggunakan pendekatan Multidimensional Item Response Theory (MIRT). Instrumen dikonstruksi berdasarkan empat dimensi laten yang saling berkorelasi, yaitu: (1) pedagogi, (2) sosial emosional , (3) pengetahuan, dan (4) kepribadian. Instrumen terdiri atas dua format butir, yaitu pilihan ganda (PG) dikotomus dan Situational Judgment Test (SJT) politomus, untuk mengakomodasi aspek kognitif dan pengambilan keputusan profesional guru. Penelitian menggunakan desain based research dengan dua tahap uji coba. Validitas isi ditentukan melalui penilaian delapan pakar dan dianalisis menggunakan indeks Aiken’s V, sedangkan reliabilitas antar-penilai dihitung dengan koefisien Hoyt. Uji coba tahap pertama melibatkan 330 guru untuk analisis validitas konstruk, dimensionalitas, kecocokan model, independensi local, invariansi, analisis Classical Test Theory (CTT) dan uji DIF. Uji coba tahap kedua melibatkan 600 guru untuk kalibrasi parameter butir dalam kerangka MIRT menggunakan program R. Butir PG dianalisis dengan model logistik dua-parameter multidimensi (2PL) untuk mengestimasi parameter diskriminasi (a) dan kesukaran (b), sedangkan butir SJT dianalisis menggunakan Multidimensional Graded Response Model (MGRM) untuk mengestimasi parameter diskriminasi multidimensi (MDISC) dan ambang kategori (MDIFF). Presisi pengukuran dianalisis melalui Fungsi Informasi Tes (TIF) dan reliabilitas marginal, serta diuji invariansi butir melalui Differential Item Functioning (DIF). Hasil penelitian menunjukkan bahwa seluruh butir memenuhi validitas isi (V > 0,64) dan model memiliki kecocokan yang memadai (GFI = 0,908; KMO = 0,697). Analisis dimensionalitas mengonfirmasi struktur multidimensi. Parameter butir PG menunjukkan daya beda positif (a = 0,57–1,88) dengan tingkat kesukaran yang bervariasi, serta memberikan informasi optimal pada kemampuan sedang. Instrumen SJT menunjukkan diskriminasi multidimensi sedang hingga sangat tinggi dengan reliabilitas marginal tinggi (≈0,87). Tidak ditemukan indikasi DIF yang substantif, dan analisis person-fit menunjukkan konsistensi pola respons dengan asumsi model. Setelah eliminasi lima butir bermasalah, instrumen final terdiri atas 40 butir (12 PG dan 28 SJT). Secara keseluruhan, instrumen yang dikembangkan terbukti valid, reliabel, bebas bias, dan memenuhi asumsi multidimensi, sehingga layak digunakan untuk pemetaan dan evaluasi kompetensi guru dalam mengembangkan karakter siswa. ***** This study aimed to develop and examine the psychometric properties of an instrument designed to measure primary school teachers’ competence in fostering student character using a Multidimensional Item Response Theory (MIRT) framework. The instrument was constructed based on four correlated latent dimensions: (1) pedagogi, (2) social emotional, (3) knowledge, and (4) personality. The instrument consisted of two item formats, namely dichotomous Multiple-Choice (MC) items and polytomous Situational Judgment Test (SJT) items, to capture both cognitive and professional judgment aspects of teacher competence. This study employed a design based research involving two field-testing phases. Content validity was established through expert judgment involving eight experts and analyzed using Aiken’s V index, while inter-rater reliability was estimated using Hoyt’s coefficient. The first field test involved 330 teachers and was used to examine construct validity, dimensionality, model fit, local independency, invariancy, Classical Test Theory (CTT) indices and DIF. The second field test involved 600 teachers and was used to calibrate item parameters within the MIRT framework using R software. MC items were analyzed using the multidimensional two-parameter logistic (2PL) model to estimate discrimination (a) and difficulty (b) parameters, while SJT items were analyzed using the Multidimensional Graded Response Model (MGRM) to estimate multidimensional discrimination (MDISC) and category threshold (MDIFF) parameters. Measurement precision was evaluated using the Test Information Function (TIF) and marginal reliability, and item invariance was examined using Differential Item Functioning (DIF) analysis. The results indicated that all items demonstrated acceptable content validity (Aiken’s V > 0.64), and the model showed adequate fit (GFI = 0.908; KMO = 0.697). Dimensionality analysis supported the hypothesized multidimensional structure. MC items showed positive discrimination parameters (a = 0.57–1.88) and a wide range of difficulty parameters, providing optimal measurement precision at moderate ability levels. The SJT instrument demonstrated moderate to very high multidimensional discrimination, with high marginal reliability (≈0.87). No substantial DIF was detected, and person-fit analysis indicated consistency between observed responses and model expectations. After removing five misfitting items, the final instrument consisted of 40 items (12 MC and 28 SJT). Overall, the developed instrument demonstrated satisfactory validity, reliability, measurement invariance, and multidimensional psychometric properties, and is suitable for assessing and profiling teachers’ competence in fostering student character.

Item Type:	Thesis (Doktor)
Additional Information:	1. DR. Riyadi, ST., MT 2. DR. Riyan Arthur, ST., MPd.
Subjects:	Pendidikan > Evaluasi Pendidikan Pendidikan > Pendidikan Dasar Pendidikan > Kualifikasi dan Profesionalitas Guru Pendidikan > Pendidikan Dasar > Pendidikan Anak
Divisions:	PASCASARJANA > S3 Penelitian Dan Evaluasi Pendidikan
Depositing User:	Mahatma .
Date Deposited:	11 Aug 2026 08:24
Last Modified:	11 Aug 2026 08:24
URI:	http://repository.unj.ac.id/id/eprint/69755

Actions (login required)

View Item