The National Education Association’s Tim Walker writes: “The already diminished reputation of high stakes testing took another hit…” This time it’s in Florida, as news broke last week that only 27 percent of that state’s fourth graders passed the Florida Comprehensive Assessment Test (FCAT) in writing. The figure represents a drop from 81 percent the previous year. Apparently, test scores for eighth and tenth graders were not much better.
According to Walker’s article, “The news sent Florida’s board of education into a damage control frenzy, as the media, parent groups, and educators demanded an explanation. Are three out of four Florida students functionally illiterate?”
To understand this issue, it is important to understand how writing is scored on a standardized test. Writing can be scored by looking primarily at content, by primarily looking at mechanical errors like spelling, grammar, punctuation or capitalization, or by a combination of both. The prior scoring rubric on the FCAT did not contain criteria dealing with mechanical errors. The new rubric takes these errors into account.
When the Florida Board of Education officials voted to lower the passing mark from 4.0 to 3.0, the results were far easier to swallow: Now 81 percent of the sunshine state’s fourth graders are proficient in writing. Without these drastic measures, the low performance score could have led to a number of schools being downgraded and untold dollars being spent on remedial programs to correct the problem.
High stakes testing like the FCAT forces the “tail to wag the dog;” in this case, writing assessment drives writing instruction. Prior rubrics gave high marks to writing which created legitimate arguments with great ideas and coherent organization, but didn’t deduct points for mechanical errors. As a result, Florida writing instruction uses a highly scripted program: http://www.writemath.com/html/writing.html. In this program, students are taught to write a coherent, cohesive five paragraph essay, and to write to a prompt, but are not instructed in ways to write accurately in a high-pressure situation. This holistic model does not take into account mechanical errors like spelling, grammar, punctuation or capitalization.
Holistic models can be a great way to learn writing, but if the writing is being assessed using a rubric that counts these mechanical criteria, the mis-match between teaching and assessment results in low test scores. Undoubtedly, there will be changes in the way writing is taught in Florida next year and the 81 percent of students who passed the proficient criteria in the past will once again pass proficient criteria next February.
There is a bigger question, however, than which curriculum should be used to teach writing in Florida. As test and scoring rubrics are constantly in flux, and teachers and schools struggle to balance good teaching practice against the realities of high stakes testing, what real purpose do these standardized tests serve?
Marygrove MAT Academic Director Diane Brown addresses some of these issues and more on Blogtalk Radio. She discusses the military origins of standardized testing, and how the validity of this kind of assessment for children has been tangled in bureaucratic red tape. Dr. Brown points out that every test is really a reading test, even if you are assessing other skills like math or chemistry. So struggling readers are probably also going to struggle on a math test, which may or may not be an accurate reflection of a student’s skills.
“Teachers are the best ones to gauge their students’ learning,” she says. “Marygrove believes that teachers should be teaching to the standards, not so much teaching to the test.”
Take a listen! Then, let us know what you think.