Measuring teachers' beliefs in relation to teaching mathematics with mathematical practices in mind

Efforts to measure teacher self-efficacy beliefs have built upon the theories of Albert Bandura. Continuing in that vein, the purpose of this study was to develop a measure that assesses teachers' efficacy and expectancy beliefs regarding teaching mathematics with an emphasis on conceptual understanding and mathematical practices, and to use this instrument to examine growth in these variables and correlations between these and other measures of quality of instruction. The instrument, developed as part of an NSF-funded project, built upon prior measures of teacher self-efficacy and expectancy beliefs, by modifying or omitting existing items and developing additional items aligned with instruction that supports the development of mathematical practices and conceptual understanding of mathematics. A group of mathematicians, mathematics educators, and K-12 experts developed a collection of items that was reduced based on factor analysis and modified based on cognitive interviews. Both exploratory and confirmatory factor analyses provide robust support for the psychometric properties of the two measures, and suggest that the two subscales are distinct and measurable constructs. Evidence of concurrent and predictive validity was obtained using multi-year correlations calculated between these scores and other constructs measured for the teachers participating in the grant.