The feedback provided to the state Department of Education regarding Louisiana's No Child Left Behind waiver application outlines several deficiencies in the state's teacher evaluation system. Among the 60 pages of feedback provided by the peer review panel charged with reviewing Louisiana's No Child Left Behind waiver application are very pointed criticisms of the state's newly minted teacher accountability system.
As The Independent reported in its May 2 blog, "LDOE foot dragging tramples transparency," Louisiana is one of 26 states (plus Washington, D.C.) requesting a waiver from the cumbersome performance benchmarks and complex accountability systems tied to the federal No Child Left Behind education law.
The U.S. Department of Education sent feedback to all states April 17 in the form of peer review notes that outline both strengths and weaknesses in the alternate plans each state has come up with to improve the quality of public education without the red tape attached to the signature federal education law.
The six peers who evaluated Louisiana's application gave the state high marks on several components of the application, including its level of input from teachers and outside stakeholders and its Trailblazer initiative for districts to avoid state takeover of low-performing schools.
But the peer panel also noted numerous deficiencies in the state's alternative education plan, particularly when it comes to the controversial value-added teacher evaluation system known as Act 54 that's expected to roll out at the start of the 2012-2013 school year.
The peer reviewers are quick to point out that the state's plan to link student test scores to teacher performance could use data from only five students the teacher taught during the school year. All six peer reviewers maintain that the "n-size" of five students is too low, a critique that Lafayette Parish School System Federal Programs Specialist and former state Department of Education staffer Tom Spencer says many teachers would agree with.
Spencer, who worked for the state Education Department when NCLB was implemented, says the five categories of effectiveness teachers will fall into under the new system, which range from ineffective to highly effective, are based on a "fairly subjective" evaluation that labels the bottom 10 percent of teacher evaluation scores as "ineffective" and the top 10 percent of evaluations as "highly effective." Under Gov. Bobby Jindal's education reform package that sailed through the Legislature in February and March, a teacher will have to be labeled highly effective five out of six years in order to gain tenure.
"Let's say all the teachers happen to be mediocre to fantastic, so the ineffective will come out of mediocre because it's the bottom 10 percent," Spencer says. "Let's say five years later the teacher population has turned over completely. Let's say they're all from mediocre to terrible. The top 10 percent will be highly effective, and that'll come out of the mediocre pool. That's what happens when you're using 10 percent of the top and 10 percent of the bottom. It doesn't determine whether they're doing good or bad."
The feedback from the U.S. DOE also questions the "inter-rater reliability" of the evaluation system, or as Spencer explains, how uniform the evaluation results will be when implemented by various administrators across the state.
"If you are a principal of a school and you come to a classroom to evaluate a teacher, you should be trained to a degree that if she comes to my school the second half of the year and is still teaching the same way, when I'm evaluating her as principal I should come up with about the same evaluation as you did," Spencer explains. "The peers say there is nothing there to ensure it's being done the same way in St. Martin as it is in Calcasieu. They want to see that everyone's doing this fairly subjective evaluation in the same way, and peers are saying it's not obvious how that's going to happen.
"The peers also say that the way the state is proposing using value-added for teachers and principals would produce contradictory results," Spencer continues. "A school performance score may be wonderful, but when you release value-added measures on teachers and principals, it may say that a school isn't great at all."
Looking at the state's plan to label students according to their achievement levels, the peer panel notes that Louisiana's benchmarks of 80 percent graduation rates and a composite score of 18 on the ACT are not rigorous enough to launch students into career and college readiness status.
"The department explained how they picked 18 as the lowest ACT score that we give anybody any points. That was based on entry point into our state colleges, but ACT says you need about a 21 to be ready for college," Spencer says. "When NCLB passed, proficient was one of our official levels. We quickly changed the definition because our results would have been dismal. In our proposal we're saying we're going to use basic as proficient, and basic is lower than proficient. What the feds are saying is wait a minute, your score of basic being considered proficient is probably too low. Why aren't you using mastery? Basically what they're saying is that the bar is too low."
The peer notes paint a much different picture of Louisiana's NCLB waiver application status than the one outlined by state Superintendent John White in an interview he gave to The Times-Picayune days before LDOE released the federal feedback. Read more on what he told The Times-Pic here.
LDOE has included its response to the federal feedback on its website. Click here to read more from the state Education Department on its NCLB waiver application.