Thanksgiving has just passed when the talk of turkeys were on our minds. Only, again this year, the turkey on the minds of most teachers, school leaders and parents isn’t the one you roast and enjoy for a holiday family dinner.
No, it’s the one that roasts our Indiana students, teachers and schools, and it isn’t edible. It’s ISTEP, the legislatively mandated test given to all public school students. Simply put, again this year it is a turkey.
It is time that all voters, and especially parents and grandparents, begin to ask and to demand that the legislative mandate for such a test be reviewed. Nearly every year, and especially this year, the mechanics of taking the test were unreliable. The test was hurriedly put together.
The technology didn’t work in many cases, and schools had to find multiple ways and times for students to take the exam. The new company’s testing plan didn’t work much better than the fired one’s. All of this causes concern and frustration of students and teachers. Their futures and their careers, respectively, depend on the test results.
But, even more serious are the characteristics of the tests themselves and how the results are reported. State testing usually relies on one of two types of tests: norm-referenced or criterion-referenced scoring models.
Norm-referenced tests compare students to the norm. When the test is created the level of difficulty is targeted to the statistical mean ( average) student and then students who vary from the average are either scored higher or lower, depending upon the number of questions they answer correctly. Thus, a fourth-grade test would have different and less difficult questions than a 10th-grade exam. ISTEP, despite declarations to the contrary, appears to be a norm-referenced test. Results are reported as a grade-level equivalents or percentiles.
A criterion-referenced test is scored based on how a student performs to the standard. For example, a test that includes the multiplication tables for numbers up to 10 will report how many problems a student answered correctly and how many he/she answered incorrectly.
It also will report the results in a manner that helps the teacher and parent know what math facts need work. Scores are usually reported in levels thereby reflecting the achievement of the student in mastering the material. And, criterion-referenced tests measure and reflect student learning growth and progress.
Norm-referenced tests involve a projection of how students are expected to perform. Through the science that undergirds such tests, the test designers attempt to create tests that reflect the normal curve of student achievement. The average student will receive the middle score or be rated at grade-level with those below or above rated accordingly.
If the test is not valid, that is it doesn’t test what it purports to test, the results when students take the test will not be valid. And, likewise, if the test is not reliable, that is it doesn’t result in similar students receiving similar scores in repeated tests, it is not valid.
Such tests are turkeys. And, that is the case with ISTEP. If you have read of the process called “setting the cut score” you may recognize it as being similar to cutting the turkey in your own home.
State officials, once the actual test results are in, have to look at the scores and determine where they will “make the cut” as to what score is at grade-level, high achievement, remedial, etc. This is the point of the process where, without knowing the specifics of any individual student’s circumstance, the die is cast and students become compared to each other. Not to the criteria. To each other.
Notwithstanding the fact that a child’s parents were in divorce court, that Grandpa passed away last week, or the computer shut down for the third time the day the student took the reading section of the test, the results are the results. And, notwithstanding the fact that a new test was mandated in Indiana for 2015 without proper time to validate it or test it for reliability, the results are the results.
You as a student, you as a teacher and you as a school are rated; in the school’s case from A to F. For students and their parents, the results can be demoralizing and harmful. For teachers, the results can hold serious consequences for both your evaluation and your pay. For schools, the results can hold ridicule and criticism for the entire student body, teachers, administrators and community.
Do we need tests to hold schools accountable for student learning? Yes.
But, such tests need to be reliable and valid, and the results need to be such that the curriculum can be evaluated and adjusted as needed. Before No Child Left Behind came our way from the federal government, end-of-course assessments at the high school level were gaining in acceptance and use across Indiana as were criterion-referenced tests designed by NWEA and others for use in K-8 schools.
These tests provided an accurate measure of student learning growth along with data that permitted close evaluation and bases for curriculum change where merited. It is time to trade Indiana’s ISTEP turkey for such tests.
Bluffton resident Roger Thornton has had a 45-year career in education that includes 25 years as a public school superintendent and six years as the executive director of the Indiana Association of Public School Superintendents. Send comments to awoods@ tribtown.com.