Too much is riding on testing data that cannot be trusted

From the Orlando sentinel, by Andrew Sparr

With the recent release of the FCAT writing scores, we begin what looks to be one of the most turbulent times for our public schools and for the education profession.

Last week, the Florida Department of Education sent notice to districts that our students' performance on a new FCAT writing test was significantly poorer than last year. In fact, last year 81 percent of students passed the FCAT writing and fewer than 30 percent passed this year.

To address this potentially huge problem, the State Board of Education reduced the score needed to pass the test, in order to have about the same number of students pass this year as did last year. Is this the correct course of action?

School-district administrators and teachers are left wondering what caused the incredible drop in the number of students passing.

Could it be that the new standards were not properly conveyed to teachers?

Could there be an issue with the test design?

Could there be problems with how the test was graded?

Was there a mismatch in the rubrics used to teach the students versus the ones used to grade them?

These are just a few of the important questions that need to be answered, and answered quickly. FCAT is an extremely high-stakes test. Students' promotion and graduation hinges on it. Teachers' and administrators' evaluations are based on it. School grades and funding are attached to it.

We cannot just have a quick fix. We must have answers, and we must ensure that the test is valid and reliable.

When the Florida Legislature passed Senate Bill 736 last year, it mandated that 50 percent of every teachers' and administrators' evaluation be based on standardized tests.

Also, put aside the issues over the validity of the model Florida is using to measure teacher effectiveness. With the recent results on writing, many are now concerned about what the FCAT reading and math scores might look like, since those, too, are new tests. Will the scores be accurate, or will we see a repeat of what we are seeing with the writing test?

Sure, the state board can change the passing score for students, but teachers and administrators are going to be judged using a complicated formula that is based on how much their students grow, not on how many pass the test.

If the test is flawed, there is no way to compare students' scores from one year to the next to determine how much they have learned.

Unless we are able to know exactly what is happening with the test, there is no way we can or should trust the test data.

With test data we cannot trust, a system that does not evaluate teachers based on what each teacher teaches and a complicated model that is not based on sound research, we are not only jeopardizing our teachers and administrators; we are also jeopardizing our whole public-school system, including our students.

We must stop this madness. This is not about accountability; this is about our children and our public schools. For the sake of our children, our teachers and our administrators, we must change this law and use common sense.

Of course, whether or not students learn should be part of a teacher's and administrator's evaluation, but when you have high stakes for students, teachers and administrators and little or no accountability for the $254 million contract lawmakers have given to the testing company, something is wrong.

Let us hold lawmakers and the testing company accountable, and let teachers teach and students learn. We must end high-stakes testing, and instead develop a system based on multiple forms of measurement that is fair for all.

Countries that we are often compared to, such as Finland and Singapore, do not use high-stakes testing to judge students and teachers. We should not, either.

Andrew Spar is president of the Volusia Teachers Organization.

http://articles.orlandosentinel.com/2012-05-20/opinion/os-ed-fcat-test-score-andrew-spar-052012-20120518_1_fcat-reading-and-math-scores-from-one-year-teachers