ABSTRACT
Program surveys of over 2,000 novice teachers and 1,000 of their reflective coaches are analyzed to explore the impact of induction, generally, and coaching, specifically, on two program outcomes. Using Exploratory Structural Equation Modeling, we find that certain induction structures, aspects of coaching activities, and teacher-coach match characteristics can have an impact on professional learning and teacher pedagogical success. Findings have implications for induction program design and the impact teacher coaching can have in professional development.
Acknowledgments
We would like to thank Barbara Howard, director of CTI, in allowing us to share these findings. We would also like to thank Linda Sanada for her technical assistance, the CTI governance team for their feedback, and all Candidates and Coaches for their tireless work in the classrooms. We appreciate Michelle Kwok, Kurt Kowalski, and Iris Riggs for their manuscript feedback and Matthew Ronfeldt and Kiel McQueen for assistance in navigating the induction literature. The 2015–2016 Candidate-Coach Match Satisfaction Follow-up Survey is available upon request; please email the lead author. This material is based on work supported by Riverside County Superintendent of Schools under C-1006110 and The Regents of the University of California. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the Riverside County Superintendent of Schools or The Regents of the University of California.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Missing data were handled in two steps. First, since we had a very large sample, cases with a substantial number of missing data elements were eliminated from the analysis. The impact of these deletions was assessed by demographically comparing the deleted cases with the final sample. We finding no significant demographic differences between the final and the incomplete cases we did not pursue data imputation procedures for these cases. Cases with less than 8 missing data points were retained and means-substitution was used to allow these cases to be analyzed using SPSS© and AMOS© statistical packages.
2 The possibility that skewed data distribution would undermine the reliability of our SEM models was addressed by building the models using the AMOS Maximum Likelihood (ML) estimation routine and then retesting the model utilizing the AMOS Asymptotically Distribution-Free (ADF) estimation routine to check on whether non-normality in the data would invalidate the SEM model generated. The ML estimation procedure is superior for model building in two respects. First, it allows for the estimation of variable means and standard deviations as well as regression coefficients. And second, ML estimation allows for the calculation of ‘Modification Indices’ which are particularly helpful in parsimoniously trimming unimportant variables from initial exploratory models. The ADF estimation process recalculates all of the coefficients in our final model to adjust them for non-normality in the data. As can be seen in Appendix A, the substantive interpretation of the ADF estimated model parameters is essentially identical to that indicated in the ML estimated model. Hence it is safe to conclude that our model findings are not, to a substantial degree, influenced by the non-normality of the survey data.
Additional information
Notes on contributors
Andrew Kwok
Andrew Kwok is an assistant professor in the Department of Teaching, Learning, and Culture at Texas A&M University. He focuses on the preparation, development, and support of classroom management for beginning teachers, particularly those who work in high needs areas.
Douglas Mitchell
Douglas Mitchell is a professor at the University of California, Riverside. His research and published writings have been on state legislative decision-making, labor relations, teacher incentive systems, public support for public schools, desegregation, class size and school board elections.
Debbee Huston
Debbee Huston is a statistician at the University of California, Riverside. Her work focuses on large-scale data analyses on teacher induction and coaching.