Students’ Perceptions of Their Learning Environments and Outcomes in Mathematics and Statistics Classrooms at Rajabhat Universities in Thailand

Abstract

The purposes of this study were first, to assess students’ perceptions of their classroom environment in mathematics and statistics classrooms at Rajabhat Universities in Thailand using modified and validated versions of the Questionnaire on Teacher Interaction (QTI) and the College and University Classroom Environment Inventory (CUCEI); second, to gauge students’ attitude toward mathematics and statistics using an abbreviated version of the Test of Science-Related Attitude (TOSRA); third, to investigate associations between students’ learning environments, their attitudes toward mathematics and statistics as a subject, and their cognitive achievement scores, and last to investigate any differences between male and female students on their perceptions of their classroom in mathematics and statistics classes using the QTI and CUCEI.

A sample of 1860 students in mathematics and statistics classrooms was divided into two subgroups, the first subgroup of 860 students in 29 classes completed the QTI, and the second group of 1000 students in 32 classes completed the CUCEI. The first subgroup completed the QTI in two forms, the Actual and Preferred Forms. The procedure began with the Actual Form and the Preferred Form simultaneously. The students in the sample responded to the Actual Form first, and then the Preferred Form. When all students had finished, the TOSRA scale of “Attitude Towards Subject” was delivered.

Following the same procedure, the second subgroup completed the CUCEI in Actual Form first and then Preferred Form in the same way as the QTI. The TOSRA scale of “Attitude Towards Subject” was administered to students after they completed both the QTI and CUCEI in Preferred Forms.

Whilst students were completing the questionnaires, the teachers responded to the Questionnaire on Teacher Interaction: Teacher Self Questionnaire, teacher version. After all students and teachers had completed the questionnaires, they were collected for data analyzing.
The Actual Form and the Preferred Form of the QTI and the CUCEI were examined to find the Cronbach alpha reliability coefficients using individual and class mean as the unit of analyses. The Cronbach alpha reliability coefficients of TOSRA in the scale of “Attitude Towards Subject” were also investigated both in the QTI and the CUCEI subgroups for individual student scores and class mean scores as the units of analyses. Simple and multiple regression analyses were used to examine associations between the classroom learning environment scales and students’ attitude toward mathematics and statistics and cognitive achievement scores.

The Cronbach Alpha reliability coefficients of the QTI’s scales ranged from 0.48 to 0.85, for individual student scores as the unit of analysis and from values ranging from 0.71 to 0.85 using class mean scores as the unit of analysis. For the CUCEI, the Cronbach alpha reliability coefficients range from 0.68 to 0.85 and from 0.93 to 0.97 respectively, using the individual student scores and class mean scores as the units of analyses in Actual Form scores. For the Attitude Towards Subject, the Cronbach alpha reliability coefficients were 0.88 and 0.97 for individual student scores and class mean scores in the QTI subgroup and 0.87 and 0.95 for the CUCEI subgroup. These reliability coefficients of any instruments are high values for uses.

Simple, multiple correlation and regression analysis revealed reasonably strong and positive associations between each of the classroom learning environment scales and students’ attitude toward mathematics and statistics, and cognitive achievement scores.

The QTI scales in Leadership, Helpful/Friendly, Understanding, and Student Responsibility were the strongest independent predictors of associations with the students’ attitude toward mathematics and statistics, the subjects and 14% of the variances in students’ attitude toward subjects could be attributed to their perceptions of their classroom learning environments.

In the same way, the Leadership, Helpful/Friendly, Understanding, Student Responsibility/Freedom, Uncertain, Dissatisfied, Admonishing, and Strict scales were significantly (p<0.001) correlated and the strongest independent predictors of associations with the students’ cognitive achievement scores. 57% of variance in students’ cognitive achievement scores can be attributed to their teachers’ behaviours.

For the CUCEI scales, the most positive and strongest independent predictors of association with students’ attitude toward mathematics and statistics were the Personalisation, Involvement, Student Cohesiveness, Satisfaction, Task Orientation, Innovation, and Individualisation scales and 13% of variances in students’ attitudes toward subjects could be attributed to their perceptions of the classroom learning environments.

However the strongest positively independent predictors of association with the students’ cognitive achievement scores were the Personalization, Involvement, Satisfaction, and Individualisation scales and 8% and 67% of variance in students’ cognitive achievement scores accounted for, respectively, by the CUCEI scales using individual student and class mean as the units of analyses.

This study is significant because it is one of the first to use learning environment instruments to gauge students’ perceptions of their learning environment in mathematics and statistics classrooms at Rajabhat Universities in Thailand. As a result, these instruments are now available and can be used by teachers to monitor what is occurring in their classrooms and thus guide improvements in their teaching, thereby leading to improve learning at Rajabhat Universities in Thailand.