Study of the Impact of the California Formative Assessment and Support System for Teachers. Report 3, Relationship of BTSA/CFASST Engagement and Student Achievement

Author(s):

Thompson, Marnie
Paek, Pamela
Goe, Laura
Ponte, Eva

Publication Year:

2004

Report Number:

RR-04-32

Supplemental Report Number(s):

CFASST-3

Abstract:

The California Formative Assessment and Support System for Teachers (CFASST) is a structured, 2-year induction program for beginning teachers that is used as part of California's Beginning Teacher Support and Assessment (BTSA) program. This study employed a quasiexperimental design to assess the impact of BTSA/CFASST on teacher practices and student achievement, by comparing teachers who had a high level of exposure to BTSA/CFASST to teachers who had little or no exposure. This report, the third of four stemming from the study, focuses on BTSA/CFASST's impact on student achievement. We collected data from California's Standardized Testing and Reporting (STAR) program for the students of 144 elementary teachers who had different levels of exposure to the program. We hypothesized that the higher a teacher's BTSA/CFASST engagement level, the better their students would do on the six STAR tests. To account for effects contributed at the school level, we used the Academic Performance Index (API) of the teacher's school as a covariate. Across all six STAR tests, we found that the students of teachers who had a high level engagement with BTSA/CFASST outscored the students of teachers with a low level of engagement with the program, after controlling for API. Though none of these score differences were statistically significant, the consistency of the results is suggestive that BTSA/CFASST has a positive impact on student achievement. Effect sizes ranged from .03 to 0.40 standard deviations, with an average effect size of 0.25 standard deviations, roughly equivalent to half a year's growth. Using hierarchical linear modeling, we found that the students of teachers who engaged strongly with the program scored 3 to 4 percentile points higher than the students of teachers who did not, after controlling for API. However, this gain only partially offset the lowered scores associated ith a student being poor (6 to 10 points), an English language learner (5 to 11 points), or having a disability (17 to 19 points). By itself, BTSA/CFASST is not able to ameliorate these correlates of lower achievement, but it does counteract them.