-
oa Development and large-scale validation of an instrument to assess precollege Arabic speaking students' attitudes toward science
- الناشر: Hamad bin Khalifa University Press (HBKU Press)
- المصدر: Qatar Foundation Annual Research Forum Proceedings, Qatar Foundation Annual Research Forum Volume 2012 Issue 1, أكتوبر ٢٠١٢, المجلد 2012, AHO4
ملخص
This study reports on the development and large-scale validation of the "Assessing Arabic Speaking Students' Attitudes toward Science Survey" (ASSASS). The study is part of a larger project funded by the Qatar Foundation and aimed at identifying factors that impact precollege (grades 3 through 12) Qatari students' interest in, and attitudes toward, science. The development was primarily motivated by the fact that no instruments have been specifically designed and systematically validated for use with Arabic speaking students. Additionally, most extant instruments were developed for use with a specific grade or school level, and many lacked grounding in a robust theoretical framework. ASSASS was grounded in the most recent revision of the theories of reasoned action and planned behavior (TRABP). A 10-member international panel of science educators and education researchers reviewed an initial pool of 74, 5-point Likert scale items for alignment with TRABP. A revised pool of 60 items was piloted with a purposively selected sample of 396 grade 3 through 12 students in Qatar. Survey administration was followed by individual interviews with a 10% random sample of the students to ensure that ASSASS items were comprehensible to, and meaningful from the perspective of, students. Exploratory factor analysis (EFA) of pilot data resulted in the deletion of 14 items, which had poor loadings on an initial six-factor model for the instrument. Next, the resulting 46-item version of ASSASS was administered to a nationally representative sample of 2,778 grade 3 through 12 students in Qatar. Again, survey administration was followed by individual interviews with a 3% randomly selected sample of respondents to help triangulate assertions derived from quantitative data. EFA followed by confirmatory factor analysis (CFA) resulted in a robust model showing a very good fit, with a Root Mean Square Error of Approximation (RMSEA) at .042 (among many other robust statistical indicators). The model included a single global factor ("attitudes toward science") plus four orthogonal residual factors (a sub-set of the six factors identified in the pilot study). These factors were: intention to pursue or engage in science, negative outlook toward science, perceptions of school science, and perceived utility of science.