Path analytical models of variables that influence science and chemistry teaching self-efficacy and outcome expectancy in middle school science teachers

Path analytical models based on a theoretical framework developed from the theories of Albert Bandura were constructed to determine factors that influence self-efficacy and outcome expectancy in both science and chemistry teaching for middle school science teachers. It was hypothesized that teaching chemistry in a general science curriculum at the middle school level was analogous to teaching science in the general elementary curriculum.;Instruments measuring science and chemistry teaching self-efficacy and outcome expectancy were modified from instruments developed by Riggs (1988). Two questionnaires, based on current literature, were developed. One probed teacher backgrounds; the other explored teacher belief systems for student learning.;The participants were 105 middle school science teachers in a large urban midwestern school. Data was analyzed via structural equation modeling and t-tests.;Findings indicate a deviation from Bandura's theory in that teachers' self-efficacy was not a determinant of outcome expectancy, therefore, separate path models were analyzed for self-efficacy and outcome expectancy in both science and chemistry teaching. Chemistry teaching self-efficacy is different from science teaching self-efficacy for the middle school science teachers in this study. However, both are influenced by teachers' past experiences. Coursework with laboratory experience had a greater effect on the development of self-efficacy in these areas than did teaching experience, but the external variables of district level and community support also play a role. Teachers' expectations for student learning are based to some extent on external variables connected with the school but also on variables associated with students' families and each student as an individual.;Factors initially hypothesized to influence science and chemistry teaching self-efficacies were coursework with laboratory experiences, science methods courses, teaching experience and other work experience not related to teaching. Outcome expectancies were hypothesized to be influenced by self-efficacy and a variety of variables that cluster into three groups, student variables, family variables and external variables.;Teachers with high chemistry teaching self-efficacy have characteristics important to effective chemistry teaching. Middle school science teachers perceive certain external variables to be more lacking for chemistry teaching than for science teaching.