Flawed FSA shouldn’t be used for grading anyone

COMPUTER TESTING ROBERT RENDOIllustration by Robert Rendo

No one is surprised that the Florida Standards Assessment (FSA) was deemedvalid by the EdCount/Alpine study. What’s odd is that the equally important factor of reliability was not considered. Finding the FSA psychometrically valid does not address the terrible conditions under which the test was administered, nor does it resolve significant concerns about psychometric reliability. Ignoring this, Education Commissioner Pam Stewart assumed that “all Floridians will share my confidence in the assessment.”

There isn’t a Florida student, parent, teacher, superintendent, board member or administrator who doesn’t see through this charade. Superintendents fromLeon to Miami-Dade have expressed their deep concerns. The study’s own numbers point out that just 65 percent of the test items match the Florida Standards. It concludes that it would be wrong to retain students or deny diplomas based on the 2015 FSA, yet Commissioner Stewart plans to use these same flawed test results to set pass/fail cut scores, grade schools and evaluate teachers. It’s fundamentally unfair to punish teachers and grade schools based on scores where 35 percent of the test items were never taught to Florida students.

Put another way, if a student answered every question based on the Florida standards correctly, he would receive a 65 or a D letter grade. It’s hard to reconcile this poor finding with Commissioner Stewart’s glowing reaction to the study. Who or what is she trying to protect?

Stewart and the Board of Education have a political problem. Admitting failure is not an option. Years of data manipulation, policy meddling, political favors and cruel stakes have resulted in the sort of public mistrust that has parents refusing to let their children take the FSA.

They know that 2016 presidential candidate Jeb Bush and his foundation lobbyists will not tolerate the unraveling of his Florida school reform miracle. Legislators also share responsibility for following Bush’s policy lead and wasting billions of tax dollars blindly passing one high stakes testing reform after another. Lawmakers such as Sen. David Simmons pushed for an FSA study that was not based solely on psychometric validity. This is his chance to abandon the “reform” agenda and do something right for Florida public schools.

Sorry, Commissioner Stewart, people aren’t buying the crazy wizardry that says a “partially-valid” FSA should be used for grading schools, establishing cut scores and evaluating teacher performance. We know that using a single test to fail a child or deny a diploma is akin to malpractice. We know that using a flawed test for docking pay or firing teachers based on one test, smacks of big government

Stop pretending that computer-based testing is more efficient when it eats up weeks of class time and is unavailable to many of Florida’s 2.74 million public school students. Stop denying the unmitigated disaster that was the 2015 AIR test rollout. Start demanding that we recover some of that $220 million dollars we paid to AIR. If we can’t afford art and music or enough guidance counselors, we sure can’t afford to pay for the failures of wealthy vendors.

It’s time to take fear out of our classrooms. Standardized tests should be limited and used for diagnostic purposes only. Let’s operate under complete transparency. Teachers, parents and students deserve to see the tests, examine the questions and understand deficits to ensure successful learning. It’s imperative that we shift to what’s best for kids and move away from the cynical, political premise of “education reform.”

Opinion by Kathleen Oropeza, Gainesville Sun

Addendum:

Predictably, the EdCount/Alpine verdict comes with strings. The department made a great show of finding an independent entity, but the two firms share close ties to FSA test creator American Institutes of Research (AIR).  AIR is a client of EdCount, EdCount founder, Ellen Forte, once worked for AIR and AIR researcher Dr. Abdullah Ferdous and Alpine employee, Dr. Chad W. Buckendahl, have collaborated on research papers with titles such as, “Recommending cut scores with a subset of items: An empirical illustration.”  Is it any wonder that the EdCount/Alpine team was the sole applicant for Florida’s $600K “validity” study?

Reprinted from The EdVocate Blog

Support our work.

Join us. Stay informed. 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s