Test data becomes useful when it does a few things. I do not believe that any standardized test currently given in my state does these things. If we found a test that could do even half of these, I’d be ecstatic.
Test results rock when:
- measuring what is taught, not the other way around;
- teachers get them in time to inform the current year’s instruction;
- they show how well or poorly a teacher has taught a course;
- students know the message those results communicate about their performance;
- students take the test seriously;
- students believe those results matter;
- those results actually do matter;
- consequences occur for students with high and low performance;
- those results can show growth over the course of a year;
- the feedback is immediate;
- students can use them to “compete” for a better score;
- those results can be used to hold students accountable, almost every bit as much as teachers.
In the immortal words of Ed Rooney, “Wake up and smell the coffee, Mrs. Bueller. It’s a fool’s paradise; he is just leading you down the primrose path.” Don’t be fooled by the promise of standardized testing. It hasn’t arrived. But I keep hoping.
Get another side of all this at dy/dan’s blog and TMAO’s blog.
4 comments
2. mrc says:
I think all 3 of you guys are more-or-less on the same side of this… in the larger sense. You might disagree about the details or about the implementations. But it seems like we all want generally the same thing, and you’ve got a great list of criteria that the STAR really doesn’t live up to. (Mostly because of the timing and not because of the content.)
What I think Dan has done a good job of is making your list at least partially a reality in his classroom with his own assessments. Now if we could give STAR-like tests every month and get the results immediately, that would be a starting point for me. Personally I think that the STAR test is a fair test for math concepts, and so it passes on your first point. It doesn’t resort to preset question templates like the NY Regents. It doesn’t ask many off-topic questions or require weird random knowledge. I think it’s a pretty fair test, as far as they go. Most of the rest of your points are about timing and about what we as teachers can do with the results. And I generally agree.
One caveat: showing how well or poorly a teacher has taught a course is not just a function of the scores! It’s a function of the scores relative to the scores those same students had on previous tests of foundational knowledge and prerequisites for the course. It’s pretty hard to make Proficient in Geometry from Far Below Basic in Algebra I, even with outstanding teaching.
3. Todd says:
Dan, I think so. If we could take these tests on computer and have the test results spit out immediately upon completion of the test, that would be great. With a little staff development, we’d have ideas for standards we should focus on. With a little more staff development, we could start libraries of plans that effectively address certain standards and measure those against progress on the test later in the year. We used to give the NWEA test to our freshmen, once in September and again in April. That was nice and quick, telling us what we needed almost immediately. It was three days of testing twice a year, which is a fair chunk of time, but it could have been worth it. I walked around the room telling students what they earned and what that meant. I had 3 students (out of 20) that asked for a retake and that they didn’t take it very seriously because they didn’t realize the implications of their score. When I mentioned that the test in April would help determine their placement the following school year, students put out effort on the test. We now give the GRADE assessment (I’ll find the link later) and it’s horrible. I want NWEA back and I want it for all grade levels.
mrc, I think you’re right which is part of the reason I posted this. I am in favor of testing to some extent. I certainly want something in my classroom that can tell me if I’m doing a good job or not. As for the relativity you suggest, I don’t know if that always translates. It might for math, but I can see it failing to do so in language arts, where things are not as cut and dry, not as objective. I think a student can move to Proficient in Geometry if that student finally pays attention and doesn’t smoke a bowl the morning of STAR that year. That’s a large part of my point: students are blowing off STAR, not necessarily unprepared for it.
Giving a STAR-like test every month? Yuck! I’d hate that. It’s not often that knowledge will be tested this way once they leave academia. Are we preparing kids for life or for more school when we prepare them for standardized tests? I’m far more for students applying their knowledge to problems that exist in the real world as often as I can. At the very least, students should be able to demonstrate their knowledge in other ways than a standardized test. STAR-like tests should be given more than once a year, but every month sounds gross to me.
4. mrc says:
I definitely get the sense that students blow off the STAR. And why shouldn’t they? It has no impact on their lives. Now the CAHSEE, they take that thing seriously. Of course, it’s trivial and they still sometimes can’t pass it. So I don’t know where that really leaves us.
I guess I like the simplicity of assessing in a STAR-like fashion, which removes my own optimistic grading of short-response questions. (You know, when the student writes three squared equals six but deep down I want to give him some points anyway for at least multiplying correctly.) I guess I just respect how the STAR assesses math, and you’re right — language arts is a whole different game.
But moving a student from FBB to Proficient? Don’t get me wrong, I really want to be able to do that. I work hard at it. But I’m finding more and more that the majority down in FBB have deeper problems. Like, they don’t know how to multiply. Or how to add fractions. So they might actually be learning Geometry from me, but they’ll never be able to show it because they can’t make the numbers work. That’s a tough situation.
1. Dan Meyer says:
[1/7/2007 - 2:21 pm]
Is the solution local control over testing? A lot of these issues of immediate turn-around, consequence to student’s grade, accountability for teachers, just don’t seem feasible from a state level. Not without a lot of legal warring, that is. Thanks for keeping things fair and balanced.