Examining a large sample of teacher candidate responses (n=1711), we found a sufficient degree of internal structure validity evidence to support the intended, continued use of the Performance Assessment for California Teachers (2008) instrument to measure teacher candidates’ skills and proficiencies with professional standards in teaching (CCTC, 2012). Our examination of the Elementary Literacy Teaching Event found that the unidimensional IRT model fits the data very well and results in high reliability (.914), which is reassuring given the dependence on raters, the varied nature of the portfolio submissions and types of tasks, and the differences in California teacher preparation programs and their candidates. Further multi-dimensional analyses revealed two distinct constructs related to the Teaching Event with implications for pre-service preparation.
Brent Duckor, Katherine Castellano, Kip Tellez, and Mark Wilson. "Validating the internal structure of the Performance Assessment of California Teachers (PACT): A multi-dimensional item response model study" American Educational Research Association (AERA) Annual Meeting (2013).