Kesahan dan Kebolehpercayaan Ujian Kemahiran Proses Sains untuk Murid Sekolah Rendah Berdasarkan Model Pengukuran Rasch (Validity and Reliability of Science Process Skills Test for Primary School Pupils based on Rasch Measurement Model)

Ellyza Karim, Kamisah Osman

Abstract


Kemahiran proses sains merangkumi kemahiran-kemahiran yang dilakukan saintis semasa mereka mengkaji dan menyiasat. Dalam tugasan di bilik sains, murid boleh mengaplikasi kemahiran berkenaan untuk menyerupai langkah-langkah prosedur yang sistematik sebagaimana saintis uruskan. Bagi mengkaji tahap kemahiran proses sains, sebuah instrumen berskala Likert lima mata yang mengandungi 68 item dibina. Kajian menggunakan instrumen ini dijalankan secara kuantitatif berdasarkan reka bentuk tinjauan. Tujuan kajian ialah untuk menentukan sejauh mana kesahan dan kebolehpercayaan instrumen ini dapat memenuhi ciri psikometrik sesebuah penyelidikan. Analisis item membolehkan item yang sah, adil dan berkualiti terpilih manakala item sebaliknya ditambah baik atau dibuang. Kesahan item dalam kajian ini dikaji dalam aspek keserasian item, polariti item, kebebasan setempat dan analisis unidimensi menggunakan analisis model pengukuran Rasch. Kebolehpercayaan yang terlibat dalam kajian ini turut dianalisis berdasarkan model pengukuran Rasch daripada aspek indeks pengasingan serta indeks kebolehpercayaan item dan individu. Data dikaji menggunakan perisian Statistik IBM SPSS dan Winsteps. Pembinaan instrumen ini menyumbang kepada pertambahan kepelbagaian bentuk ujian terutama jenis pentaksiran masa kini yang mula cenderung ke arah bentuk penilaian kendiri. Instrumen-instrumen sebelum ini lebih mengkaji kepada aspek pengetahuan namun instrumen ini menumpukan kepada aspek pengetahuan dan kebolehan. Implikasi dapatan kajian ini memberi kelebihan kepada guru sains menilai tahap kemahiran proses sains murid berkaitan.

Kata Kunci: Model pengukuran Rasch, kemahiran proses sains, kesahan, kebolehpercayaan, penilaian kendiri


Science process skills include the skills that scientists apply when they study and investigate. During scientific investigations, pupils may carry out those skills to resemble systematic procedural steps as the scientists organize. To study the level of science process skills, a five-point Likert scaled instrument containing 68 items was developed. The study was done quantitatively based on survey design. The purpose of the study was to determine the extent of the validity and reliability of this instrument to fulfill the psychometric characteristics of a research. Item analysis allows valid, fair and quality items to be selected while items otherwise enhanced or removed. The validity of the items in this study were analysed in terms of item fit, item polarity, local dependence and unidimensionality using the Rasch measurement model analysis. The reliability involved in this study was analysed based on Rasch measurement model from the aspect of item and individual separation index and reliability index. The data were processed using IBM SPSS Statistics and WINSTEPS. The construction of this instrument contributes to the increase in the variety of test forms, especially the types of current assessment that tend to be in the form of self-evaluation. The previous instruments studied the knowledge aspect but this instrument focused on the aspects of knowledge and ability. The implication of this study gives advantage to science teachers to evaluate the level of their student’s science process skills.

 

Keywords: Rasch measurement model, science process skills, validity, reliability, self evaluation


Full Text:

PDF


DOI: http://dx.doi.org/10.17576/JPEN-2018-43.03-01

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

....................................................................................................................................................................................................


Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License

© Malaysian Journal of Education | Jurnal Pendidikan Malaysia
ISSN 2180-0782 | eISSN: 2600-8823