Perceptions of the learning environment, attitudes towards science, and understandings of the nature of science among prospective elementary teachers in an innovative science course

Abstract

The major purpose of this study was to evaluate the impact of a science course for prospective elementary teachers on their perceptions of the learning environment, attitudes towards science, and understandings of the nature of science. The sample consisted of 525 female students enrolled in 27 classes of A Process Approach to Science (SCED 401) at a large urban university in Southern California. Also comparisons were made between SCED 401 and the students' previous laboratory course with regard to the learning environment and attitudes. Perceptions of the learning environment were measured using scales from the Science Laboratory Environment Inventory (Open-Endedness and Material Environment) and the What Is Happening In this Class? (Student Cohesiveness, Instructor Support, Cooperation, Investigation). Attitudes towards science were assessed using the Enjoyment of Science Lessons scale from the Test of Science-Related Attitudes (TOSRA). Students completed the Nature of Scientific Knowledge Survey (NSKS) based on their entire science education experience-not just the one laboratory class which they had taken previously. Comparisons were then made with their understandings after having completed SCED 401. Finally, associations between the learning environment and the student outcomes of attitudes and understandings of the nature of science were explored. This study embraced the current trend in classroom learning environments research of combining quantitative and qualitative methods. Qualitative components included items from the open-ended questionnaire, Views of Nature of Science, interviews with students, and an analysis of concept maps. The qualitative findings expanded and complemented the quantitative results and, in several cases, supported the construct validity of scales assessing the learning environment and attitudes.Another purpose of this study was to investigate the effects of using real research data for growth rates of four species of Antarctic seabirds (i.e., implementing an 'intervention') in six classes of SCED 401. The objective of the intervention was to increase the authenticity and quality of an experimental design project. In addition, the wildlife biologist who collected the data guided the students during the project. Although the intervention did not lead to an appreciable improvement in students' perceptions of the learning environment, differences between intervention and nonintervention classes were statistically significant for Enjoyment of Science Lessons from the TOSRA and for Creative from the NSKS (effect sues were 2.64 and 2.06 standard deviations, respectively). Results of this study indicated that during a factor analysis, the large majority of learning environment items belonged to their a priori scale (43 out of 46 items had factor loadings above 0.40) A valid instrument for use with prospective elementary teachers was produced by combining relevant scales from the Science . Laboratory Environment Inventory and the What Is Happening In this Class? A weaker factor structure was found for the Nature of Scientific Knowledge Survey. However, by moving close to half of the -faulty' items from the NSKS, the internal consistency reliability of scales improved considerably. This study also found large and statistically significant differences between students' previous laboratory class and SCED 401 for all six leaning environment scales. The largest difference was found for the level of Open-Endedness (effect size was 6.74 standard deviations).A statistically significant difference also was found for Enjoyment of Science Lessons (effect size was 2.98 standard deviations). Differences were not as dramatic with regard to understandings of the nature of science, although differences for two scales (Creative and Unified) from the NSKS were positive and statistically significant. This study replicated past research by finding statistically significant positive correlations between all six learning environment scales and Enjoyment of Science Lessons. However, by far, Instructor Support had the largest independent association with enjoyment, using both the individual and class mean as the units of analysis. A positive link between a favorable leaning environment and the student outcome of understanding nature of science also was found. This research makes a distinctive contribution to the learning environments field because it is the first study to investigate laboratory classroom environments at the university level with prospective elementary teachers. The study is also the first to build a bridge between the classroom learning environment and the student outcome of understanding the nature of science. The study has implications for undergraduate laboratory course instructors, for science teacher educators who develop and instruct in elementary teacher preparation programs, and for future elementary teachers and the science learning of their future students.