Evaluation of Effectiveness: Pennsylvania

2013 Identifying Effective Teachers Policy

Goal

The state should require instructional effectiveness to be the preponderant criterion of any teacher evaluation.

Meets
Suggested Citation:
National Council on Teacher Quality. (2013). Evaluation of Effectiveness: Pennsylvania results. State Teacher Policy Database. [Data set].
Retrieved from: https://www.nctq.org/yearbook/state/PA-Evaluation-of-Effectiveness-22

Analysis of Pennsylvania's policies

Commendably, Pennsylvania requires that objective evidence of student learning be the preponderant criterion of its teacher evaluations. The state's Teacher Effectiveness Tool will be the evaluation tool used for all teachers in the state. Full implementation is slated for school year 2013-2014. 
Student performance must count for 50 percent of a teacher's evaluation score. This half must be based on multiple measures of student achievement and be comprised of the following: building-level data (15 percent), must at least include student performance on assessments, value-added assessment system data, grad rates, promotion rates; teacher-specific data (15 percent), student achievement attributable to a specific teacher as measured by student performance on assessments, value-added assessment system data, progress in meeting student goals; and elective data (20 percent), including measures of student achievement that are locally developed.
Four rating categories must be used: distinguished, proficient, needs improvement and failing. Distinguished and proficient are considered satisfactory. Needs improvement is considered satisfactory, except if the teacher gets another needs improvement rating within 10 years, and then it is considered unsatisfactory. No teacher can be rated needs improvement or failing based solely on student test scores. 
Classroom observations are required. 

Citation

Recommendations for Pennsylvania


State response to our analysis

Pennsylvania reiterated that student performance must count for 50 percent of a teacher's evaluation score. It is comprised of 15 percent building-level data (available in 2013-2014), 15 percent teacher-specific data based on a three-year rolling average (available in 2015-2016 for teachers of tested subjects), and 20 percent elective data in the form of student learning objectives (required in 2014-2015). The system has been in development for three years.

How we graded

Research rationale

Teachers should be judged primarily by their impact on students.

While many factors should be considered in formally evaluating a teacher, nothing is more important than effectiveness in the classroom. Unfortunately, districts have used many evaluation instruments, including some mandated by states that are structured, so that teachers can earn a satisfactory rating without any evidence that they are sufficiently advancing student learning in the classroom. It is often enough that teachers appear to be trying, not that they are necessarily succeeding.

Many evaluation instruments give as much weight, or more, to factors that lack any direct correlation with student performance—for example, taking professional development courses, assuming extra duties such as sponsoring a club or mentoring and getting along well with colleagues. Some instruments hesitate to hold teachers accountable for student progress. Teacher evaluation instruments should include factors that combine both human judgment and objective measures of student learning.

Evaluation of Effectiveness: Supporting Research

Reports strongly suggest that most current teacher evaluations are largely a meaningless process, failing to identify the strongest and weakest teachers. The New Teacher Project's report, "Hiring, Assignment, and Transfer in Chicago Public Schools", July 2007 at: http://www.tntp.org/files/TNTPAnalysis-Chicago.pdf, found that the CPS teacher performance evaluation system at that time did not distinguish strong performers and was ineffective at identifying poor performers and dismissing them from Chicago schools. See also Lars Lefgren and Brian Jacobs, "When Principals Rate Teachers," Education Next, Volume 6, No. 2, Spring 2006, pp.59-69. Similar findings were reported for a larger sample in The New Teacher Project's The Widget Effect (2009) at: http://widgeteffect.org/.  See also MET Project (2010). Learning about teaching: Initial findings from the measures of effective teaching project. Seattle, WA: Bill & Melinda Gates Foundation.

A Pacific Research Institute study found that in California, between 1990 and 1999, only 227 teacher dismissal cases reached the final phase of termination hearings. The authors write: "If all these cases occurred in one year, it would represent one-tenth of 1 percent of tenured teachers in the state. Yet, this number was spread out over an entire decade." In Los Angeles alone, over the same time period, only one teacher went through the dismissal process from start to finish. See Pamela A. Riley, et al., "Contract for Failure," Pacific Research Institute (2002).

That the vast majority of districts have no teachers deserving of an unsatisfactory rating does not seem to correlate with our knowledge of most professions that routinely have individuals in them who are not well suited to the job. Nor do these teacher ratings seem to correlate with school performance, suggesting teacher evaluations are not a meaningful measure of teacher effectiveness. For more information on the reliability of many evaluation systems, particularly the binary systems used by the vast majority of school districts, see S. Glazerman, D. Goldhaber, S. Loeb, S. Raudenbush, D. Staiger, and G. Whitehurst, "Evaluating Teachers: The Important Role of Value-Added." The Brookings Brown Center Task Group on Teacher Quality, 2010. 

There is growing evidence suggesting that standards-based teacher evaluations that include multiple measures of teacher effectiveness—both objective and subjective measures—correlate with teacher improvement and student achievement. For example see T. Kane, E. Taylor, J. Tyler, and A. Wooten, "Evaluating Teacher Effectiveness." Education Next, Volume 11, No. 3, Summer 2011, pp.55-60; E. Taylor and J. Tyler, "The Effect of Evaluation on Performance: Evidence from Longitudinal Student Achievement Data of Mid-Career Teachers." NBER Working Paper No. 16877, March 2011; as well as H. Heneman III, A. Milanowski, S. Kimball, and A. Odden, "CPRE Policy Brief: Standards-based Teacher Evaluation as a Foundation for Knowledge- and Skill-based Pay," Consortium for Policy Research, March 2006.