Assessing lower track students' learning in science inference skills in Singapore

Abstract

It is a generally held view amongst educators today that the development of students' inference skills is an important aspect in their education as 21st Century learners as it requires higher order cognitive competences. Oftentimes, students in the lower tracks are considered slower learners and may have difficulties with the development of such skills. There is, however, limited empirical evidence to support such claims. As a result, there is a lack of understanding how such skills are taught, and how lower track students acquire them. The purpose of this study is to investigate lower track students' science inference skills over one academic year, to better understand their learning and development. To determine this, three multiple-choice science inference skills tests were developed based on science syllabus and administered over a 9-month period. In total, 1397 Grade 7 lower track (i.e., Normal Academic) students from 38 Singapore secondary schools participated in the study. The students' performances were determined through three equated tests using Rasch common-item procedures. The results showed that students experienced greater difficulty with tests over time. They particularly had difficulties with questions pertinent to graphs, tables, diagrams, or charts, or required them to extend their thinking beyond the given information. They also had difficulty in deducing answers using the elimination technique, and items that involved experiments and variables. Items that involved pattern recognition, concluding using range, application of a given concept, and limited information were easier for them. The findings also have implications for science teacher education in terms of assessment literacy, and the science teaching of lower track students.