Testing
Let's talk about the point of tests and exams. That's always a nice, calm, unemotional topic, right? One of my children and I were just talking about this a bit ago, and I thought it might make an interesting discussion. Now, before I begin, I need to say up front that I test extremely well. This post is in no way sour grapes. Taking tests and doing academic work is definitely a strength and I played the school game quite well. Perhaps I was able to play it too well, and that has more than a little bearing on my thoughts about tests and testing now.
The stated purpose for tests and exams is to determine how much of the material the student learned and understood, often so that a grade may be given for the class. On the face of it, it doesn't sound too bad. Students are in class to learn, and it is not unreasonable to ask them to show that they've done that. Seems reasonable, huh? In fact, for a very long time, I was right there. Teachers teach. Students learn. Tests are taken. Grades are given. That's education.
Or is it?
There was that moment while I was researching homeschooling, before we had actually taken the official plunge, where I came across this statement by John Holt**, "If a student fails a test, it was the wrong test." Unless you are a fairly radical homeschooler, I imagine that you are currently reacting rather negatively to that statement... possibly even sputtering a bit. I did. I sputtered a lot. It was wrong. I just knew it was wrong. There was a lot of cognitive dissonance going on.
Here is the crux of the problem. Testing assumes that students are not doing their best. It is the entire foundation behind testing. Think about it. The 'best' in a testing situation is 100%, because that is when the student gets everything correct, thus they did their job to learn the material. But everyone does not get 100%, do they? In fact, if everyone does get 100%, then immediately the accusations are leveled that the test was too easy, and the tests are adjusted accordingly for the next time. What kind of system sets out to discover what the students know, then rigs the system so that not everyone can succeed? It becomes clear that it stops being about what the students know or not, and becomes about something else. I'm not even sure what the something else is, but it certainly isn't about learning.
I'll let you in on a secret. Those of us who are good, very good, at taking tests, can usually do quite well even if we don't know the information. This is especially true if the test is multiple choice or true and false. (This is also why straight fill in the blank tests were never my favorites.) I'm great at the informed guess. Even as a child, though, I knew that my test results had nothing to do with what I actually knew. I wasn't fooling myself into thinking I had the information down pat; the test was just another game that was played in school. Every so often when I would voice this thought to a teacher, the inevitable reply was, "You need to stop being so hard [critical] on yourself." Pfft. If they only knew. I tried to tell them, but since they wouldn't listen and take me seriously, I just stopped trying and continued to study the least amount possible.
So what is the correct test? It's not one that's out to get the student; to show them exactly how ill prepared they are. It's not a measure of commitment to the information, and it's certainly not a measure of intelligence or potentiality. It is a window into a sliver of time, when, assuming the student was in top form and not distracted by other areas of life, the teacher gets to find out how well he communicated the information and how well the student was able to guess at what was deemed important. Sometimes a test doesn't even give that much information.
The right test (if we still have to use a testing model), is one that sets the student up for success. It is one that allows the student to share what they have learned, what they find interesting about the material, what they want to focus on next. It would be more of a personal diagnostic tool than anything. A bad test assumes the student doesn't want to learn and needs to be threatened into doing so. A good test, would aid both teacher and student to discover what is clear and what still needs to be worked on. It represents a partnership and not a dictatorship.
But I'm still not convinced that we need to have them at all. And if you're curious, I've never given a test to any of my children in the name of education.
**I was in one these books, I can't remember: How Children Fail, How Children Learn, or What Do I Do Monday?
The stated purpose for tests and exams is to determine how much of the material the student learned and understood, often so that a grade may be given for the class. On the face of it, it doesn't sound too bad. Students are in class to learn, and it is not unreasonable to ask them to show that they've done that. Seems reasonable, huh? In fact, for a very long time, I was right there. Teachers teach. Students learn. Tests are taken. Grades are given. That's education.
Or is it?
There was that moment while I was researching homeschooling, before we had actually taken the official plunge, where I came across this statement by John Holt**, "If a student fails a test, it was the wrong test." Unless you are a fairly radical homeschooler, I imagine that you are currently reacting rather negatively to that statement... possibly even sputtering a bit. I did. I sputtered a lot. It was wrong. I just knew it was wrong. There was a lot of cognitive dissonance going on.
Here is the crux of the problem. Testing assumes that students are not doing their best. It is the entire foundation behind testing. Think about it. The 'best' in a testing situation is 100%, because that is when the student gets everything correct, thus they did their job to learn the material. But everyone does not get 100%, do they? In fact, if everyone does get 100%, then immediately the accusations are leveled that the test was too easy, and the tests are adjusted accordingly for the next time. What kind of system sets out to discover what the students know, then rigs the system so that not everyone can succeed? It becomes clear that it stops being about what the students know or not, and becomes about something else. I'm not even sure what the something else is, but it certainly isn't about learning.
I'll let you in on a secret. Those of us who are good, very good, at taking tests, can usually do quite well even if we don't know the information. This is especially true if the test is multiple choice or true and false. (This is also why straight fill in the blank tests were never my favorites.) I'm great at the informed guess. Even as a child, though, I knew that my test results had nothing to do with what I actually knew. I wasn't fooling myself into thinking I had the information down pat; the test was just another game that was played in school. Every so often when I would voice this thought to a teacher, the inevitable reply was, "You need to stop being so hard [critical] on yourself." Pfft. If they only knew. I tried to tell them, but since they wouldn't listen and take me seriously, I just stopped trying and continued to study the least amount possible.
So what is the correct test? It's not one that's out to get the student; to show them exactly how ill prepared they are. It's not a measure of commitment to the information, and it's certainly not a measure of intelligence or potentiality. It is a window into a sliver of time, when, assuming the student was in top form and not distracted by other areas of life, the teacher gets to find out how well he communicated the information and how well the student was able to guess at what was deemed important. Sometimes a test doesn't even give that much information.
The right test (if we still have to use a testing model), is one that sets the student up for success. It is one that allows the student to share what they have learned, what they find interesting about the material, what they want to focus on next. It would be more of a personal diagnostic tool than anything. A bad test assumes the student doesn't want to learn and needs to be threatened into doing so. A good test, would aid both teacher and student to discover what is clear and what still needs to be worked on. It represents a partnership and not a dictatorship.
But I'm still not convinced that we need to have them at all. And if you're curious, I've never given a test to any of my children in the name of education.
**I was in one these books, I can't remember: How Children Fail, How Children Learn, or What Do I Do Monday?
Comments
Consider an actor learning a long monologue. The actor learns it by reviewing the lines in the script, and then closing the script and testing herself on how much she recalls and how far she can get. Then she returns to the script and works on the spots where she got stuck, closes the script, tries again, and repeats the process. This is self-testing at its best, as a learning tool. The test doesn't carry any significance, except as a way for the actor to assess her own progress.
This kind of informal formative assessment isn't stressful for the learner, because nothing is riding on the outcome... no grade or score or reward. And unlike the high-stakes testing games, these kinds of tests are authentic, because their only purpose is to really figure out where to push myself. We do this kind of informal testing unconsciously all the time whenever we are learning a new skill or topic in which we are truly interested. We push ourselves to see how far we can go, and then we use that knowledge to guide our future learning. A knitter tries out a new pattern as a way to force herself to learn new techniques; an athlete pushes herself to see how well she can perform, and then uses that knowledge to figure out what new skills to work on.
J.
Once again, I contend the best learning is self-motivated and self-directed... including self-testing to be sure the information can be recalled if needed. But then that would be the choice of the learner and as a diagnostic, learning tool rather than an arbitrary exercise with no benefit.
e
(And now all of you can imagine better what our conversations at home sound like. J. and I like to discuss things and will often play devil's advocate for the other's position.)