Gender- and Academic Level-Bias in MATS when Measuring Attitude Towards Science in Indonesia: A Rasch Analysis

Journal of Pedagogical Research. Duzce University, Faculty of Education, Konuralp Campus, 81620, Duzce, Turkey. e-mail: ijopr.editor@gmail.com; Web site: https://www.ijopr.com/

Y

16

Foreign Countries; Science Education; Achievement Tests; Secondary School Students; International Assessment; Attitude Measures; Student Attitudes; Validity; Test Bias; Instructional Program Divisions; Gender Bias; Evaluation Methods

Program for International Student Assessment; Attitude Scale

2602-3717

The waning interest of students in science became a global concern. The purpose of this research was to translate, adapt, and validate the My Attitude toward Science [MATS] questionnaire instrument, which was used to measure students' attitudes toward science in the Indonesian context. We also investigated the items that contributed to gender and academic level differences in responses. To analyze the psychometric properties of the instrument, 223 students participated in the four-dimensional Indonesian variant of the MATS instrument. The unidimensionality and model fit of the four dimensions of the questionnaire were evaluated using Rasch analysis. In general, In general, the MATS instrument has been confirmed as multidimensional, attaining a configuration of three dimensions through the amalgamation of two previously distinct dimensions. The reliability of the instruments was quite good, but the reliability of the instruments in the expectation and perception toward scientist dimensions is still relatively low. The questionnaire was incapable of distinguishing between students with positive and negative expectation and perception toward scientist. The range of MNSQ values for all three dimensions was approximately 0.73 to 1.55. In the gender based DIF analysis, two items were identified: MATS 14 (-0.68) and MATS 21 (-0.75). In the meantime, each class responded differently to one or more items based on their academic level. In conclusion, this study concluded by recommending the simplification of the scale used to assess the expectation and perception toward science dimensions. It is anticipated that the Indonesian version of the MATS instrument will aid educators, researchers, and policymakers in obtaining valid and dependable data regarding student attitudes toward science. The implications and future studies on masculinity, oldscientists, and lower ATS in early high school classes have been discussed.

As Provided

2024

EJ1417795