Student test scores could account for as much as 50 percent of teachers’ performance evaluation ratings under a proposal the Indiana State Board of Education is expected to consider next week.
That would be a huge about-face for Indiana, which made a point of allowing local schools to decide how much test scores should count in 2011 when other states mandated that student test scores factor in at a high percentage. The proposal could require legislative action before it is put into practice.
Under a 2011 state law that overhauled teacher evaluation, test scores were required to “significantly inform” a teacher’s evaluation score, which was interpreted very differently by different school districts. The different approaches to counting test scores made it difficult to compare teacher results across school districts. But critics of evaluation systems that rely heavily on student test scores say it is an unreliable method of quantifying a teacher’s impact that can vary wildly from year to year.
As part of an effort to craft more specific guidance for how test scores should count, the state board brought in as a consultant The New Teacher Project, which met with the board’s strategic planning committee today. The New York-based company, started by former Washington, D.C., school chancellor Michelle Rhee, suggested scores could be factored into teacher ratings at different percentages depending on what subjects they teach.
For example, teachers who teach subjects tested on state ISTEP exams could have student scores count for 33 to 50 percent of their ratings, the company suggested, while teachers who don’t teach tested subjects could have test scores count less, perhaps a range of 25 to 40 percent. Districts could determine locally what percentage to use based on the evaluation models they’ve chosen.
Indiana districts are allowed to create their own evaluation models, which has led to very different results in different districts. Ratings this year were exceedingly favorable — more than 97 percent of the state’s teachers who were rated were deemed “highly effective” or “effective,” the top two of four categories. Hardly any were rated “ineffective.”
Jessica Conlon, presenting for the company, said test scores shouldn’t be the only objective measure used to determine how teachers are performing. Others, such as portfolios of student work and classroom observation, are also important.
“A teacher’s job is far too complex to be able to look at one metric only and decide that tells us that this person is a good teacher,” she said.
Conlon also suggested the state board wait until 2016-17 to implement these new ranges so the state has one year of data from its new standardized tests, which kids will take for the first time in 2015-16. The extra time would make the evaluations more reliable, she said, and give districts more time to change their evaluation systems.
“With that baseline and plenty of time to improve, I think that can help address many of the concerns that have been raised by teachers and others who are going to be evaluated under this metric,” board member Gordon Hendry said.
The company’s recommendations also included changing the way evaluations affect teacher pay. Some teachers and administrators see evaluations as a tool for keeping salaries low, Conlon said, so the system could be improved if pay increases aren’t withheld when teachers receive low ratings but are not deemed ineffective.
Board member Brad Oliver, who is also an education professor, agreed that teacher pay needs to be part of the evaluation discussion, especially for cash-strapped districts that might give only performance bonuses for good evaluations but not regular pay raises. He said if administrators are making evaluation decisions based purely on raising pay for teachers, it undermines the evaluation process.
“The focus seems to be on increasing the performance grant, which I’m all for,” Oliver said. “But if it’s being done with the belief that that will fix the underlying problem, I think there’s a disconnect between that and what’s really happening.”
He said the legislature should participate in these discussions to make sure performance bonuses don’t replace yearly cost-of-living increases.
The committee voted to send the recommendations to the state board for discussion and a vote at its Feb. 4 meeting. Hendry said he thinks the recommendations make good steps toward improving teacher evaluation in the state.
“We know that the teacher evaluation system in Indiana is not perfect and that there is room for improvement,” Hendry said. “But that’s OK. And I think we’re going to get there eventually, and I think it’s going to be better for everyone who is part of Indiana’s education system, especially, I think, teachers.”