It's a poor craftsman who blames his tools
April 01, 2010
Conventional wisdom in many education circles (see here) tells us that multiple choice tests are the enemy of critical thinking and deep content mastery. Such tests, we're told, can't really assess student learning. What's worse, they ?encourage? teachers to teach test-taking tricks and strategies rather than to demand true mastery of essential content and skills.
I bought this rhetoric for a long time. As a teacher, I always felt like I was taking a shortcut if I chose multiple choice tests over short answer questions or essays.
That was, until I started actually writing network-wide interim assessments and helping teachers use the data from these tests to drive daily instruction, one-on-one tutoring, and small group instruction. It was only then that I really began to realize the power of these frequently-maligned assessment tools.
To be clear, I wholeheartedly agree that multiple choice tests cannot and should not be the only means of assessing student knowledge and skills. But, they rarely are. For example, I can't think of a single instance where open-ended response questions aren't part of the state assessment system. Or when the best teachers don't pair these assessments with projects, essay tests, and other measure of student learning.
But in reality, there is much teachers can learn about student progress toward mastery of essential content and skills from multiple choice questions. In fact, I sometimes believe that you can learn more about where student understanding is breaking down by analyzing data from a well-designed multiple choice question than from many open-ended questions.
Unfortunately, most analyses of multiple choice tests begin and end with the student test score?how many questions did they get right and wrong. This type of superficial analysis ignores the most useful data that a multiple choice question can provide. Careful analysis of the distracters?the incorrect answers students select?can give teachers valuable insight into where student understanding is breaking down, and therefore can help teachers maximize their instructional time by better targeting whole-class and small group instruction and individual tutoring. What's more, these assessments can be scored quickly and their data used to drive instruction almost immediately.
Sadly, though, many people see multiple choice tests only for their limitations and not for their power.
Yes, these assessments?like all tests?are limited. Yes, teachers must ensure that they are using multiple measures to paint a complete picture of student learning. And yes, some (poor) teachers will look for shortcuts and use valuable instructional time to teach students tricks to try to ?game? the tests themselves. (Though, given the formulaic way most open-ended responses are scored by states, I would argue that short answer and essay questions encourage more such short cuts than multiple choice questions do.)
But, aren't we kidding ourselves by saying that teachers who take such instructional shortcuts would magically stop if it weren't for standardized assessments? And, are we sure that the trade-offs of opting for tests with fewer (if any) multiple choice questions?cost, timeliness of data reporting, subjectivity of scoring, etc.?would paint a clearer picture of student learning?