What Makes a Good Teacher?

Photo courtesy of the Princeton Regional Schools

A Task Force report unveiled last month by New Jersey Governor Chris Christie recommends making student test performance an important criterion in teacher evaluation. While that would reflect well on Princeton teachers, it runs counter to the approach adopted by Princeton Regional Schools. Princeton Charter School, on the other hand, has expressed measured support for Christie’s reforms.

The task force proposes that equal weight be given to student achievement (primarily through state assessments) and teacher practices (largely through classroom observations).  The report also encourages greater emphasis on measuring student achievement over time. “I don’t agree with it at all,” said Aaron Thayer, an English teacher at Princeton High School, adding that it was important to look at other  facets of individual student progress, although it would be a time-consuming process.

Standardized test scores carry relatively little weight in the evaluation of teachers at Princeton Regional Schools, according to Lewis Goldstein, the PRS Assistant Superintendent for Human Resources.  “What the student is doing during the year, you don’t see in test scores but you see it in the classroom,” he said.  While standardized testing is listed in the Staff Evaluation Packet as one of several indicators of pupil progress, Goldstein emphasized the use of other performance markers.

“There is an assessment of the teacher’s work in the classroom, teacher portfolios, teaching abilities, the teacher’s interaction with students,” said Goldstein, citing the ability to motivate student participation as an example.  “We see if the teacher is able to get a student out of his or her shell.”  Evaluation criteria in the current PRS Teacher Evaluation System also include planning and organization, staff and parent relations, and professional growth, among others.

Click here to read a complete description of Princeton’s teacher evaluation system. The results of the evaluation of PRS teachers are published yearly on the district’s website.

Goldstein felt that basing 50% of a teacher’s evaluation on student achievement would not adequately reflect teaching abilities.  “There’s so much more that goes on in the classroom.  Suppose the student is not feeling well on that day [of the test] or is having problems as home,” he explained, adding that test scores should constitute at most 25% of a teacher’s evaluation.

Agreeing that test scores are not the only measure of student learning, Steve Baker, Associate Director for Public Relations at the New Jersey Education Association (NJEA), said that the governor’s approach would lead to “massive expansion” of standardized testing.  “It is not a good way of measuring teacher effectiveness, and it is not beneficial to students,” he explained.  “If we have standardized testing in every grade level, students are going to do nothing but prepare for tests.”

Baker pointed to an Economic Policy Institute study, released in September 2010, that deemed it “unwise” to allocate as much as 50% of teacher evaluation to test scores, saying that scores alone were “not sufficiently reliable or valid indicators of teacher effectiveness.”  Education researchers had found that under this system, of the teachers judged to be in the top 20%, less than a third were in that top group the following year.

The Charter School approach

At the Princeton Charter School (PCS), performance review is also based on multi-dimensional criteria, such as the teacher’s classroom management and impact on student learning, level of professionalism, teamwork, special contributions to the school’s program and commitment to the PCS mission and goals.  However, test scores do play a role in teacher evaluations.  Important assessment tools include not just written evaluations from classroom observations, but also comparisons of students’ performance on major assessments – namely the Educational Records Bureau standardized tests and New Jersey-mandated Assessment of Skills and Knowledge tests.

Saying that Governor Christie’s reforms were “on the right track,” Maureen Quirk, a PCS founder, said that fear of incorporating student performance measures into teacher evaluations stemmed mostly from the emphasis on final test scores rather than how teachers change student performance.

“Like most people in New Jersey, I am in favor of making student performance one of the factors in teacher evaluation.  The hard part will be creating a fair system that emphasizes student progress throughout the year rather than the final level,” said Quirk, who is also the current Treasurer on the school’s Board of Trustees.

To create such a system, PCS separates its students into three groups: those performing at the level of the lowest quarter of the students tested in suburban schools nationwide, those in the highest quartile, and those in the middle.

“For each group, for each of the areas tested, we track the progress from year to year.  What we are looking for is students moving out of the lowest level into the middle, and out of the middle into the highest level,” Quirk explained, adding that this would be a good model for other schools to adopt as well.

PCS students are tested in late October to get a baseline measure of performance, with subsequent testing each year.  Scheduling tests in the fall, with the results out in December, gives PCS teachers more stake and interest in improving student performance since they would still be teaching the same pupils for the rest of the academic year.  Results from spring testing, as done in most schools, are only known in the summer, limiting its impact.  “The teachers are not around, and the students are in different classes [the following year],” Quirk said.

Steve Baker of the NJEA criticized this “value-added model” based on student test score improvement, as it is known, calling it a “bumper-sticker type model.”  “The data is just not reliable,” he said.  “Test scores are not designed to measure teacher effectiveness, yet we are going to put incredibly high stakes in bad data.”

Princeton Charter School considers other means of evaluation.  Beyond test scores, student achievement is measured in terms of unique “Milestones,” which are significant learning tasks accomplished in each grade, such as comparing two historical civilizations or writing a quantitative science laboratory report.

            PCS also takes into account parent feedback.  In the regular Princeton public schools, parents are not part of the teacher evaluation process.  However, Quirk said that PCS takes input from yearly parent surveys “very seriously.”

“Although it is not a tool for individual teacher evaluation, it does give the administration insight into how well things are going in various subjects and grade levels, and can help focus attention on areas of concern,” she explained.

Because PCS is a small school, there is no need for a specific formula of various components in a teacher’s evaluation, although Quirk said that a percentage formula would be useful in ensuring uniform standards under the district system, which covers multiple schools and supervisors having limited contact with teachers.

“Our staff is small enough that everyone knows each other.  And they are all measured by the same standard since a small cooperative group of supervisors does all the evaluations,” Quirk explained.

Rethinking teacher tenure

While New Jersey law states that non-tenured teachers must be evaluated three times a year and tenured teachers once annually – as Princeton Regional Schools does – PCSconducts more than the legally required number of teacher evaluations each year, with up to two for tenured teachers.  It also carries out informal evaluations in which teachers are not notified beforehand.

“[The Head of School] simply drops into the classroom to see how a random lesson is being done,” said Quirk.  “Charter schools don’t have unions, so the principal can go into the classroom when he or she feels like it.”  In addition, new teachers are mentored for three years instead of the required one.

In the public schools, tenured teachers who are found to be ineffective are put on improvement plans.   Goldstein explained that salary increments are withheld from those who still fail to meet desired standards, with the final step being termination of employment.

“Contrary to popular belief, there is a process [to handle tenured teachers], though it is lengthy and involves a lot of paperwork,” said Goldstein.

Both Goldstein and Quirk agreed that the state’s policy to award teacher tenure after three years was a poor system.

“It’s not fair to the district to have to make a lifetime commitment [to employ a teacher] after just a little over two years,” said Goldstein.  “It’s hard to judge.”

“Tenure in New Jersey comes too soon for inexperienced teachers.  It’s a system that supports the status quo,” said Quirk.  “It’s hard to tell [within three years] how good they are, whether they will become a fantastic teacher.”  She suggested progressively extending teacher contracts, from one to three to five years, based on evaluations of teacher performance.

PRS’s own evaluation reform

Even before Gov. Christie fired off his recent salvo of reforms, Princeton Regional Schools was in the process of updating the district’s current evaluation model, which has been in place for thirty-five years.

“Last year we put together an administrative evaluation team to assess different models in the nation,” said Goldstein.  “We are getting to the point of finalizing [the changes], taking into consideration what the governor and others have said.”

Reforms include:

·       new observation and Annual Performance Report forms,

·       making the evaluation process “more reflection-oriented” for both teachers and administrators, such as what strategies they have been using for students who are struggling, according to Goldstein. 

·       a four-tiered rating system (“Exceeds Performance Expectations,” “Meets Performance Expectations,” “Performance Needs Improvement,” and “Does Not Meet Performance Expectations”) instead of the current three categories of “Satisfactory,” “In Need of Improvement” and “Unsatisfactory.”

In the 2009-2010 school year, 96.8% of the 341 teachers in the Princeton Regional Schools district received a “Satisfactory” rating.  1.1% was found to be “In Need of Improvement,” while 2.1% of teachers were rated unsatisfactory.

If not Christie’s formula, then what?

Ideas abound on how to reform the teacher evaluation system in New Jersey.  Many agree that a framework with clear criteria is needed, although there is no consensus yet on what factors – including student achievement – should carry more weight.

“What he [Gov. Christie] doesn’t have is a rubric, a set of guidelines upon which one can judge a teacher,” said Quirk. “We hope that as a charter school in New Jersey we can help him.  It is the right thing to do to have some measure of teacher quality based on student performance.”

Goldstein said that schools should adopt a more “holistic view of how teachers teach, based on specific criteria,” with less emphasis on “one-day tests.”  He proposed adapting the higher education model of teacher evaluation, which assesses not just teacher effectiveness in the classroom but also his or her contributions to the school and community.  Even students should have some input, Goldstein added.  For example, students could evaluate teachers on an informal basis, making up 5-10% of the total grading.

“There are so many factors that go into student education and student learning,” said Baker.  “We [NJEA] are not opposed.  Test scores have a place [in teacher evaluation], but in a much more limited role than what the governor [is doing].”

Rhonda Maguire, a parent looking to enroll her child in the Princeton school district, saw a “political agenda” behind Gov. Christie’s reforms in using the value-added model and linking teacher pay to evaluation based primarily on student test scores.  “It’s a tricky catchphrase, a way of cutting back on teachers’ salaries and pensions,” she said.

She was skeptical of spending millions of dollars to implement any statewide system.  “We should let the superintendents decide how to evaluate their teachers,” she said, explaining that it was “impossible for a centralized state government” to fully understand the widely varying circumstances and needs of different school districts. 

About the Author

Audrye Wong

Princeton University, Class of '13