A recent article in The Atlantic highlighted the efforts of a California charter school to evaluate students on yearly portfolio presentations instead of the ubiquitous tests we all have become so used to:
On a recent Wednesday morning, 11th-grader Sophia Wellington took to the undersized stage at the front of her high school gym and, with seamless poise, demonstrated what a smarter method of assessing students could look like...Wellington is 15, has a desire to travel, and is "into bands." She wants to go to a good University of California system school, preferably Los Angeles or Berkeley...Then Wellington showed, without dropping so much as an "um" or a "like," how she will prove she has the intellectual skills and the grit to get there.
I have a few thoughts on this. The first one is: well, good. There are 14,000 school districts in the U.S., and some of them are bound to include schools that are willing to take a chance on the simple idea that you can find out more about what a student does and does not know from a portfolio and presentation than you can from a standardized test. As I often tell my students, the problem with standardized tests, if they contain multiple choice questions, is that you can always be right without knowing anything: you just have to get lucky and pick the right answer. Yeah, yeah; it’s a statistical improbability for a test-taker to be right often enough to pass a test without knowing anything, but when we use “cutoff scores” to determine everything from who graduates to which classes are avaiklable to students, getting another question right (or wrong) can make a pretty big difference. When students have to actually present work they have created and answer questions about it, they may actually have to know what they’re talking about, especially if the teachers evaluating them do.
As a matter of fact, buried in this article is a quote from Linda Darling-Hammond and Frank Adamson that illustrates why portfolio assessments are so much more valuable than standardized ones. Here’s what they said:
Researchers argue that, by tapping into students' advanced thinking skills and abilities to explain their thinking, performance assessments yield a more complete picture of students' strengths and weaknesses...By giving teachers a role in scoring essays and other performance measures, the way the Advanced Placement and International Baccalaureate programs do today, performance-oriented assessments encourage teachers to teach the skills measured by the assessments and help teachers learn how to do so.
This is a funny quote, because it starts by suggesting that researchers are simply making an “argument"—as if this knowledge being presented to us is contested, open to interpretation and debate. And, okay, that’s probably true; researchers are only arguing that you can get a more complete picture of student learning by actually asking students to explain their thinking, and it’s theoretically feasible in the universe of possiblities to get a more complete picture of what students have learned if you don’t ask them to explain their thinking—I mean, who knows, really?—but this is kind of like saying that researchers argue that gravity is what keeps us all from flying off into space, even if it’s theoretically possible that we all stick to the earth for reasons we have never even thought of. It would be nice to see fewer academics make their points using such conditional language, as it opens the door to unhelpful (and unnecessary) “differences of opinion” about things that ought to be pretty non-controversial. But I digress.
The real point is that when teachers actually evaluate the work that students do—teachers, as opposed to, say, computer scoring machines or underprepared and underqualified human scorers who are paid to churn out as many scores as possible as quickly as they can—they are more likely to teach in ways that will prepare students for the assessments. The author of the article takes this idea and runs with it in a different direction, more or less making the simplistic point that if we just let the teachers grade the tests (or, at least, benchmark tests specifically indexed to The Tests) then when they teach to the test the kids will do a better job on them. That’s probably true, but it misses the point.
A better approach would be for more schools—and not just charter schools, but traditional public schools too—to turn to some form of performance assessment that requires (and allows) students to show what they know instead of focusing so much on what they don’t know. One of the biggest problems I have with standardized testing (aside from the way we use the tests to set invisible and ever-changing benchmarks of “effectiveness"; and aside from the fact that we give tests “in the dark"—meaning teachers never know what’s going to be on them, and students don’t either; and aside from the fact that they are “sold” to us by for-profit companies instead of being made as part of the educational process...) is that they are divorced from the curriculum that actually gets taught in schools. One solution to this problem is hinted at in the article: if we could get Common Core up and running, and get teachers to teach it, and then get tests that tested it—and then, apparently, get teachers to score those tests—then we’d be on to something. Maybe.
I have a different vision myself. In this vision, teachers would work together, with the helpful guidance of a standard to strive for, to create curricular experiences that are engaging and interesting to students. They would start this process by identifying what they want students to known and be able to do, then circle back to those instructional goals when it is time to assess them. Their instructional and curricular decisions would be guided by a shared set of larger educational goals but tailored to the specific contexts in which they teach. Most importantly, their work would encourage students to demonstrate what they have learned and would keep decisions about assessment “in house.” It’s not that hard to imagine a “system” like this actually working.
Would it mean we could no longer have a standard of quality control? Well, in the first place, we don’t really have one now. We can give standardized tests in an attempt to compare schools until we’re blue in the face, but without alignment between them and what teachers teach the tests are pretty meaningless—not to mention really expensive. (So much for efficiency.) Sure, we should be worried about subjectivity, especially since we know that teachers have a tendency to grade students based on what they think the students’ individual abilities are, and we also know that schools have a tendency to place students on tracks based on perceived abilities as well, which only applies more pressure to teachers who want to help students get outside of the boxes they have been put in. But we could work around that. For one thing, if teachers were able to find consensus around a set of shared goals—and if they also had the respect that comes with professional autonomy, allowing them to teach in ways that suit their knowledge and personalities—then assessing what students have learned wouldn’t be so difficult after all. It might even provide us with really helpful information about how different schools and teachers approach the task of educating their students, which, in turn, could help ensure that good ideas spread an bad ideas don’t. And isn’t that the point of this whole accountability thing in the first place?