It's something of an obsession with me, writing about AzMERIT scores. A new set of scores, a new use of the scores, a new news story about the scores, and there I am with another post or two or three. So here's yet another post, a rambling discussion on why the tests, the way they're reported on and the way they're used drive me nuts.
Let me start by getting something out of the way. The tests in and of themselves aren't bad. They give a reasonably accurate reflection of students' abilities in reading, writing and math. During my last few years teaching in a district outside of Portland, Oregon, I had to give the Oregon version of the high stakes standardized tests to my sophomore English classes. I did a pretty good job of predicting what my students' scores would be based on what I had learned about their reading and writing abilities during the eight months before the tests, which means the test scores generally reflected the students' skill levels. There were a significant number of exceptions, where students got higher or lower scores than I thought they would, which tells me the tests aren't always accurate on an individual level. But when you're looking at large numbers of students, and assuming everything is on the level—no "helpful encouragement" from teachers during the tests, no erase-and-replace of students' answers by staff after the students hand in their tests — their average scores tell you something about their skill levels relative to other groups of students.
Now, with that out of the way, the problems. The first is, the high stakes nature of the tests distorts the schools' curriculum and, sometimes, the test results. Since teachers, schools and school districts are judged by their students' scores, they're compelled to do everything they can to get the best results possible. That means teaching to the test, which means spending inordinate amounts of time and energy giving students the narrow skills needed to fill in the right bubbles. The give and take of loosely directed discussions is a luxury only to be indulged in when time allows. Creative pursuits, long term projects, even time on the playground are secondary to the central focus of the classroom: preparing students for test day. Teachers become mechanical skill-and-drill sergeants, which is not what they thought they signed up for when they decided to join the teaching profession. Students are encouraged to become robotic, learning how to be successful at performing variations of one repetitive task — answering short questions by picking the right answer from a short list of possibilities. The classroom is a different place — I would say a worse place — thanks to high stakes tests. And, sad to say, all that sweat, toil and tedium generally only adds a few points to students' scores and even less to students' actual skill levels, and since pretty much everyone is doing it, it's a wash. Every class, school and district's ranking in the state stays pretty much the same as it would have been if no one paid any attention to the test until test day.
And sometimes, the pressure to raise test scores leads individual teachers, or whole schools and districts, to cheat. Some schools and districts have been caught at it. Teachers and administrators in Atlanta went to jail for changing answers on student tests year after year. Others do it but haven't been caught. A series of articles in USA Today a few years back talked about a nationwide analysis of erasures on student tests and found that in many schools, including in Arizona, the number of wrong answers erased and replaced by right answers was as likely to be random as it was likely that the school be struck by lightning on test day. Though state departments of education rarely look deeply into suspicious scores, Arizona's ADE found
nine schools where the evidence is strong enough, it's highly probable students' test papers were altered. Most likely, those schools are the visible tip of a larger problem. And that's just the most easily detectable form of cheating. There are lots of undetectable ways to boost scores without increasing the students' skill levels.
Cheating can become addictive, and additive. If a teacher cheats one year, how does he/she go back to being honest the next year without having to explain the drop in scores? If third grade teachers cheat, fourth grade teachers look bad if their students score lower than they did back in the third grade—and so on, up the grades. Educators are basically an honest, moral, but not necessarily courageous lot. If you put their salaries and/or their jobs on the line, many of them are liable to do what it takes to push those scores up.