Gender, grade-level and stream differences in learning environment and student attitudes in primary science classrooms in Singapore

Abstract

A major focus of this research was the validity and reliability of a learning environment and attitude questionnaire in primary school classrooms in Singapore. The learning environment scales were chosen from Constructivist Learning Environment Survey (CLES) and What Is Happening In this Class? (WIHIC). The scales to assess the attitudes to science were chosen from the Test of Science Related Attitudes (TOSRA). The study also investigated gender, grade-level and stream differences in learning environment and attitudes to science and also investigated the relationships between attitudes and the learning environment. Data were gathered from 1081 students from 55 different classes in 4 different primary schools in Singapore.This study is the first in the Singapore context that focused on Gifted Education (GE) pupils and High Ability (HA) pupils in the primary school setting. It is also the first study in Singapore which focused on investigating gender, grade-level and stream differences in learning environment and attitudes to science and associations between students’ attitudes to science and their perceptions of the classroom learning environment within the one study.Factor analysis was conducted for the 70 items in the learning environment and attitude scales based on the CLES, WIHIC and TOSRA. From the original 70 items, 61 items were kept in the same 10-factor structure. The 61 items each had a factor loading of at least 0.40 on its a priori scale and lower than 0.40 on all of the other scales. The 61-item version of the questionnaire containing learning environment scales based on the WIHIC (5 scales with 4−8 items in each) and the CLES (3 scales with 4−5 items in each) and 2 six-item attitude scales based on the TOSRA was accepted. Also the learning environment scales based on the CLES and WIHIC were capable of differentiating significantly between classes, and all learning environment attitude scales were reliable when used with this sample of elementary school students in Singapore.The use of MANOVA and ANOVAs identified the presence of any gender, grade-level, stream, stream-by-gender, grade-by-stream, grade-by-gender and stream-by-gender-by-grade interactions for each scale. The statistically significant findings were: • significant gender differences for Involvement, Teacher Support, Task Orientation and Cooperation • significant grade-level differences for Teacher Support, Task Orientation, Cooperation and Enjoyment • significant stream differences for Involvement, Cooperation and Personal Relevance • significant stream-by-gender interactions for Task Orientation and Enjoyment • significant grade-by-stream interactions for Investigation, Student Negotiation, Scientific Inquiry and Enjoyment • no significant grade-by-gender interaction for any dependent variable • no significant three-way stream-by-gender-by-grade interaction for any dependent variable.Simple correlation and multiple regression analyses replicated considerable prior research into associations between student attitudes and the learning environment: improved student attitudes were associated with more emphasis on the aspects of learning environment assessed in this study.