Teacher Assessment in California: Is It a Reliable Measure of Success?

• In Event: Context Matters: Unpacking National Teacher Assessments and the edTPA in Four States

Abstract

Objectives
This paper aims to contribute new knowledge on teacher preparation and assessment by taking a closer look at the subset of beginning teachers in Alternative California’s[1] credentialing program who have failed the required performance assessment.

Perspectives/Theoretical Framework
In 200, twelve California colleges and universities developed a new gatekeeper test to grant K-12 teaching credentials based on teacher “performance.” The assessment process they devised is known as PACT and is used today by 30 college and university teacher training programs, including the non-traditional pathway to teacher credentialing under discussion in this paper. In 2009, U.S. Secretary of Education Arne Duncan earmarked the PACT assessment as exemplars for educational change in the United States. Following Duncan’s endorsement the PACT assessment program launched a national roll-out and is currently being replicated in at least 17 additional states (Sawchuk, 2010). While PACT has been hailed as a marked improvement over existing modes of assessment, patterns in the failure records of Alternative California’s teachers raise concerns about this assessment tool and about new teacher education more broadly.

Data Sources & Methods
We conducted a mixed-methods study of 10 teachers who submitted the PACT between 2009-2012 and failed on the first scoring. Quantitative data included teacher test scores (CBEST, PACT), undergraduate and graduate GPA, years teaching, and an analysis of written portions of the PACT. Qualitative data included interviews with teachers, administrators, and program director about candidate performance and professional judgments about the lesson plan structure and written reflection of PACT. We used a “Situational Analysis” approach (Clarke, 2005) to try to understand why these teachers were failing and to propose possible interventions in teacher training and in the assessment itself. Rather than entering into our research with a specific hypothesis, we begin by focusing on a phenomenon and asking: “what is happening here?” We looked for patterns, correlates, anomalies, and absences in our data to generate new research questions.

Results
Three main findings emerged: 1) higher degrees do not correlate with success on the PACT assessment, 2) the data point that relates most directly with PACT failure is scoring low on the state’s standardized writing assessment: CBEST, and 3) higher failure among candidates of color and EL candidates in contrast to Pecheone & Chung’s analysis of scores separated by race, gender and EL candidates (2006).

Significance
Our research enables us to begin the process of untangling the variables that are coming into play around fail rates. At the program level, we can address and act on the changes in the pathway to better meet the needs of new teachers so that they pass the PACT on the first try, and that men, EL candidates, and candidates of color pass at the same rates as white, female candidates. Our work further highlights more wide-reaching issues such as the validity and reliability of the PACT assessment and set of scoring rubrics. Public Education deserves a diverse set of students who deserve a diverse teaching staff.

Authors

• Amy Reising, High Tech High

• Zoltan G. Sarda, High Tech High