I will protect your pensions. Nothing about your pension is going to change when I am governor. - Chris Christie, "An Open Letter to the Teachers of NJ" October, 2009

Tuesday, January 27, 2015

Why Does the Public Hate Standardized Tests?

The New Jersey Education Association, the state's largest teachers union, released a bombshell of a poll yesterday; click through to see the raw results for all voters and just for parents. My guess is that once we clear away the snow, we're going to be talking about these results for a long time.

nj.com has a breakdown:
In conjunction with D.C. based Mellman Group, NJEA administered two polls in December asking a range of questions related to standardized testing. One poll surveyed 800 likely voters, including 200 parents. A second poll added 200 more parents to the original 200, creating a 400-parent poll. 
Both polls reflected the same general opinions on testing, though by different percentages. The poll of 400 parents found: 
• 81 percent of parents are concerned that “teachers are forced to teach to the test.” 
• 80 percent are concerned that “too much of the school year is spent preparing for standardized tests.” 
• 78 percent want to limit the number of hours of testing. 
• 78 percent say testing “causes stress for students.” 
• 77 percent are concerned that testing “takes time and money from other educational priorities.” 
A majority of parents polled, 82 percent, said they want legislators to pass a testing "Bill of Rights," requiring transparency on high-stakes testing, including how much they cost taxpayers and how student data will be used. Parents also want the ability to opt their students out of tests — 66 percent said they support having a parental right of refusal.
I've noticed some folks already picking nits at the questions, trying to show they're biased (for the most part, I don't agree that they are). But even if you see a slant here, the resulting numbers are still astonishingly high. When four our of five parents are telling us that testing has become a pernicious force in their children's lives, something is clearly amiss.

Of course, this has been bubbling up for a long time and all over the country. Parents are increasingly worried that testing has taken over the lives of their children -- so much so that politicians who have traditionally relied on standardized tests to push their particular brand of "reform" are now forced to deal with the backlash against those tests.

So what's going on here? Why such a pronounced and forceful rejection of a testing regime that admittedly has been expanding, yet has actually been in place for a good numbers of years?

Part of what's going on, I think, is that the tests are a proxy for general discontentment with how the lives of our children are being lived these days.

Race To Nowhere remains, for me, one of the best indictments of the "achievement culture" that has consumed the lives of our children. I think there's a strong sense out there that kids are being run through a gauntlet, and if they don't survive, they aren't deemed worthy of entering the middle class and leading decent lives.

Standardized tests don't measure learning so much as they rank and order children according to a social construction that is based on the notion that we must have winners and losers in our society. 

Neo-liberals have warped our sense of social justice: we now believe in "level playing fields," rather than making sure everyone has a chance to contribute to our society and be guaranteed a life of basic human dignity. So long as poor people of color have an equal chance to work on Wall Street and pull down an obscene amount of money at the expense of everyone else, we can call our society "just."

As I've said now a thousand times: we have millions of people in this country doing hard, sometimes dirty, sometimes dangerous, often backbreaking and monotonous, but necessary work. These folks are working hard and playing by the rules, but they can't even afford to live in decent housing, let alone take a vacation every now and then.

Ranking and sorting students does nothing to address this core problem in today's America. All it does is ratchet up the pressure on children, whose parents understand that a life without a college degree increasingly means a life of misery. It's become clear to many (even if they don't articulate it in the same way I am doing here) that the standardized test is the lynchpin for this system.

And it's also become clear that the defenders of the real status quo refuse to acknowledge the truth about this state of affairs. Here in New Jersey, our Department of Education (and their willing mouthpieces) keeps telling us that standardized tests like the PARCC are really for the benefit of the students. Education Commissioner David Hespe, for example, swears the tests are really all about helping the wee ones:
Hespe said that state and federal regulations require at least 95 percent of students take the exams, or districts could potentially lose some undefined funding. Most of all, Hespe said, he wanted to point to the importance and benefits of the state assessments for individual students and their schools.
“The PARCC assessments will, for the first time, provide detailed diagnostic information about each individual student’s performance that educators, parents and students can utilize to enhance foundational knowledge and student achievement,” Hespe wrote. [emphasis mine]
 Too bad his own Assistant Commissioner, Bari Erlichson, disagrees:
ERLICHSON: In terms of testing the full breadth and depth of the standards in every grade level, yes, these are going to be tests that in fact are reliable and valid at multiple cluster scores, which is not true today in our NJASK. But there’s absolutely a… the word "diagnostic" here is also very important. As Jean sort of spoke to earlier: these are not intended to be the kind of through-course — what we’re talking about here, the PARCC end-of-year/end-of-course assessments — are not intended to be sort of the through-course diagnostic form of assessments, the benchmark assessments, that most of us are used to, that would diagnose and be able to inform instruction in the middle of the year.

These are in fact summative test scores that have a different purpose than the one that we’re talking about here in terms of diagnosis.

Kudos to Erlichson, because that is exactly right: standardized tests like the PARCC are accountability measures; they are not designed to help inform instruction for students.

One of the more ignorant arguments I've been hearing lately about standardized tests runs along the lines of: "But we've always had tests! These are just better tests, because they're better aligned to 'real-world' learning objectives! So they can help teachers create better instruction for their students!"

This is a poor argument for several reasons. First, a standardized test is of little use to a teacher who is instructing non-standardized students. 

Any teacher who went through a decent preparation program has seen some sort of variation of this graphic:

This is from Carnegie Melon's Eberly Center for Teacher Excellence and Learning Innovation, but it could have come from any number of sources, because it's such a basic idea. A teacher sets a learning objective, creates instructional activities, and designs assessments to see if the students actually learned what they are supposed to learn. It's a dynamic system: the assessment provides feedback on whether the objectives were well-designed and the instruction was well-executed. But the objectives and instruction also inform the assessment. Everything works together, and a good teacher is constantly adjusting and refining all parts of the system.

But a standardized test can't be changed; it gives very little useful information as to how a teacher should adjust instruction and objectives. As a practical matter, it's next to useless because by the time the test results get back to the teacher the kids have usually moved on to another grade. But even if the results came back instantaneously, the assessment wouldn't help inform instruction much because the test can't change as instruction and objectives change.

That's not to say that these tests don't have a role. I'm probably going to piss off some folks who normally agree with me, but I actually do think these tests are necessary as accountability measures. We should be able to judge at some level whether or not the system is working; certainly, I and others who attempt to judge the effectiveness of education policies benefit from having some sort of metric of student learning to help us make our determinations (so long as we use them appropriately).

But it's not really useful to anyone to have an accountability measure so overwhelm the system that it takes on unwarranted importance. Which brings us to the second reason why the pro-PARCC argument doesn't hold up: we don't know whether these tests are truly "better" or not.

Last year, we found that the New York State standardized exams were loosely aligned, in a convoluted way, to a vague learning outcome: getting a "C+" in a freshman course in math or a "B-" in language arts at a number of selective colleges. Getting good grades is certainly better than getting poor ones, but is anyone prepared to argue that it is any more of a "real world" outcome than the tests themselves?

One claim made about PARCC is that it measures "higher order thinking." From what I've seen, however, what's really being measured is whether a student can see past the tricks and distractors thrown up by the test to pick the correct answer. Test items like these are excellent for teasing out a normal distribution of scores:

But it's quite a stretch to claim this is "better" than the standardized tests we had before, which yielded the same distribution. Do we really believe the NJASK was giving a distorted view of how students were performing relative to each other

I don't have a problem with standards, and I don't have a problem with setting them high. I'm fine with testing, and I'm fine with the appropriate use of standardized tests. We certainly can use tests to inform decisions -- not compel, but inform -- about teachers and schools and policies. The great thing is that we can do this without an overly onerous or expensive testing regime simply by understanding a few basic statistical principles.

But even if we make our standards and tests truly "better," we're still going to have to solve this:

Nearly 70 percent of the variation in New Jersey school-level SAT scores can be explained by  levels of student economic disadvantage. Putting pressure on kids to perform better on "better" standardized tests isn't going to change this. Setting higher standards isn't going to change this.

And maybe that's why people are getting fed up with standardized tests: they are a distraction that keeps us from doing what really needs to be done.

Testing, testing, 1, 2, 3...

1 comment:

Christine said...

Here is a also a great short piece of Jia Lee testifying at NCLB hearings about harms of standardized testing
(done by Michael Elliot)


Here is the vimeo url