Test anxiety: I call B.S.

Great Debates
61

[QUOTE=amarinth]
And people who are good at test taking (or who have taken classes from people who were good at test taking) know that that’s horrible advice. Because if you can eliminate at least one answer, mathematically you’re better off guessing than leaving it blank. The benefit from the percentage of right guesses outweighs the penalty for the wrong guesses. Even with the penalty, scores favor the bold.
[/quote]
Actually, it’s only mathematically in your favor if you can eliminate two answers (assuming five possible answers, as on the SAT family of tests). But yes, the principle is sound.

The way it works is if they have a certain grade in certain core classes, and took the AIMS test at every single opportunity, then they get a point bump based on this or that or the other thing, which might give them enough points to pass. It’s got nothing to do with the teachers, unfortunately, just raw grades and attendance.

62

[QUOTE=Diogenes the Cynic]
I mean that it makes teachers teach to the tests. That’s sort of the opposite side of the grade inflation/standardized test divide, but it still has the net result of neglecting educational foundations.
[/QUOTE]

I never understood the objection, in principle, to teaching to a test. Now, the particular test might be a bad test that is not crafted well, but a comprehensive test that tests not only knowledge of a particular area, but also comprehension and application of that knowledge should test exactly what was supposed to be taught.

63

[QUOTE=Diogenes the Cynic]
I guess I can’t see how answering a question requires a skill. You either know the answer or you don’t. I also can’t understand why pressure (I never actually felt this pressure. I can’t ever remember being the least bit nervous about a test) would make somebody not be able to anser a question if they know the answer.
[/QUOTE]

“I guess I can’t see how falling asleep requires a skill. You either are sleepy or you’re not. I also can’t understand why stress (I never actually felt stress. I can’t ever remember being the least bit stressed) would make somebody not be able to fall asleep if they are tired”

“I guess I can’t see how taking a dump requires a skill. You either want to take a dump or you don’t. I also can’t understand why side effects of some medications (I never actually felt such side effects. I can’t ever remember having the least bit of side effects from any medication) would make somebody not be able to take a dump if they want to.”
This is ludicrous. Have you never experienced trying to remember the name of a person or some other fact and, even though you know you know it, you can’t for the life of you remember it at that moment? (Maybe not, since you’re a “certified genius” :rolleyes: )

Obviously, it is possible for the brain to know a certain piece of information and for it to not be able to recall it at some point in time.

In addition, some tests put so much pressure on people (essentially, in some cases, the next 3 hours will have a huge impact on the rest of your life), that it is not unreasonable to think that some simple acts (like answering questions) might become difficult.

It’s like walking on a 1-foot wide plank of wood.

If I ask you to do this task while the piece of wood is on the floor, no problem.

If I ask you to do this 10 stories above the ground (in a hangar, so wind will not be an issue), most people would have a huge problem with that.

I can see some adrenaline junkie say “Either you know how to walk across a 1-foot wide plank of wood or you don’t. I have never experienced any anxiety from doing this at 10 stories above ground, so I can’t see why anybody would have a problem with it”

FTR, I have not had any stressed-induced inability to answer tests, but I’m sane enough to realize that it is something that others may go through.

64

[QUOTE=Diogenes the Cynic]
Yes, but I think the disservice is in passing students along dishonestly and not giving them genuine correction and assistance when they need it. The standardized tests expose the kind of dishonesty and laxity that BlinkingDuck is talking about.

This is the fault of the high schools, not the kids. The kids are essentially being shined on and told they are doing better than they actually are. Your kid could definitely pass these tests if he had a real foundation.
[/QUOTE]
There’s a simple remedy for this. Post all grades in 2 forms: actual and curve. If you see grades of 60 (D) and 90 (A) then the student and parent will know the true level of achievement.

It’s very common in my area to have a student moving from the city to a suburban school go from A’s to D’s. The parents are, of course, shocked.

65

[QUOTE=ArizonaTeach]
Actually, it’s only mathematically in your favor if you can eliminate two answers (assuming five possible answers, as on the SAT family of tests). But yes, the principle is sound.
[/QUOTE]
It’s mathematically in your favor if you can eliminate even one answer. Since the negative points are there to counteract the effectiveness of random guessing, any guess that is statistically better than random is in your favor.

66

[QUOTE=iamthewalrus(:3=]
It’s mathematically in your favor if you can eliminate even one answer. Since the negative points are there to counteract the effectiveness of random guessing, any guess that is statistically better than random is in your favor.
[/QUOTE]
No, it’s not. If you choose randomly between four answers, you stand a far, far better chance of gaining negative points overall. With three answers you score about a 75% chance of gaining points when the test is taken as a whole. It’s an exercise I do with my students every single year, and it’s actually a part of most SAT prep courses curriculum (which I teach). I can demonstrate it for you, if you want.

67

[QUOTE=ArizonaTeach]
No, it’s not. If you choose randomly between four answers, you stand a far, far better chance of gaining negative points overall. With three answers you score about a 75% chance of gaining points when the test is taken as a whole. It’s an exercise I do with my students every single year, and it’s actually a part of most SAT prep courses curriculum (which I teach). I can demonstrate it for you, if you want.
[/QUOTE]

I suppose I must be missing something. I’d like to see that demonstration.

My understanding is that an incorrect answer is -1/4 points, while a correct answer is +1 point. Thus, to break even, you need to get about 1/5 of all the questions correct.

If you have an independent 1/4 chance of getting each particular question correct, taking the number of questions to be large, the binomial probability distribution of your percentage score can be well approximated by a normal distribution with mean 1/4 and variance 1/4 * (1 - 1/4) = 3/16 (thus, standard deviation sqrt(3)/4). In this case, the probability of getting more than 1/5 of all the questions correct is approximately 54.6% (obviously, it’s going to be more than 50%, as 1/5 is less than the mean).

Now, of course, there are discretization effects lost by the approximation with the bell curve, but I can’t imagine they could possibly salvage any analysis which suggests that the probability of scoring below the expected value is significantly greater than the probability of scoring above it, so much so that the median score of the probability distribution is much less than 20% while the mean score remains at 25%. So what is the element of reasoning here which I have missed?

68

[QUOTE=ArizonaTeach]
No, it’s not. If you choose randomly between four answers, you stand a far, far better chance of gaining negative points overall. With three answers you score about a 75% chance of gaining points when the test is taken as a whole. It’s an exercise I do with my students every single year, and it’s actually a part of most SAT prep courses curriculum (which I teach). I can demonstrate it for you, if you want.
[/QUOTE]
I must have missed something, too. Taking a much more simplistic approach than Indistinguishable did:

  1. There are five answers to each question

  2. The test-taker can eliminate one answer, leaving four potentials

  3. A correct answer is worth one point, and incorrect answer, -0.25 points

Given a random distribution, the test-taker could expect to get 3 wrong answers (-0.75) for every one right answer (+1), so guessing with one answer eliminated would, on average, provide +0.25 points per question.

Where’s my error?

69

Well, it’s not always true that a positive expected score implies a greater than 50% chance of a positive score; for example, suppose the test just had one question. In this case, the chance of a positive score is just the chance of guessing that one question correct, which is much less than the chance of guessing it wrong and thus getting a negative score. (Similarly, we could consider a bizarre raffle which awarded a million dollar prize, with entering the raffle costing one dollar, and odds of winning being one in a thousand. In this case, certainly, the mean profit from entering the raffle is very large, and yet, all the same, the probability is 99.9% that that one will just lose money in so doing.)

So it’s not obvious that we can make that leap, if the probability distribution is significantly asymmetric with respect to its mean. But the particular details of this distribution suggest to me that, as the number of questions is large, the resulting probability distribution is well-modeled by, and thus very close to being, a symmetric one, and thus the leap can indeed be made, as you suggest.

70

I would say I am relatively smart and I certainly enjoying testing in the classroom and standardized tests throughout almost all of high school. But I started getting panicky on tests when I took comm coll classes at night during high school. And if I read a question and didn’t get it immediately (like I would have in high school), I would panic and everything I knew and had studied would fly out of my head. I compensated by writing down formulas and outlines of material before reading the test; I knew they were true when I wrote them down, even if I lost it when reading the test. I attribute it to being afraid to fail in the more rarifed atmosphere at college, when things were more difficult and I would have been in big trouble with the family if I messed up.

71

[QUOTE=duncfoo]
First Post in GD!
[/QUOTE]
Uh, cite? (just messing with you duncfoo, glad to have you posting here in GD)

How does test anxiety compare to perfomance in athletics? Say your kid is on the basketball team, and dominates in practice, virtually scoring at will. But in a real game, couldn’t put the ball in the hoop to save his life. Sure, the kid most likely has game anxiety, but does that mean the coach should keep letting him shoot those bricks throughout the season? Or should lesser talented players get a chance who can actually perform during games?

72

[QUOTE=ArizonaTeach]

Weeelll…NCLB has actually made grade inflation a hell of a lot more difficult, if that’s what you mean.
[/QUOTE]

In one sense, yes. In another sense, it’s made grade inflation mandatory.

You know, and I know, and anyone with the most basic statistical knowledge knows, that half of a group is gonna be below average. If we set our expectations for competency at a given grade level at what we think an average student at that grade level should be able to accomplish, then roughly half the students should be below grade level, and half should be above. Simple stuff.

This year, I think, NCLB requires that 84% of students be at or above grade level. By something like 2014, NCLB will require that 98% of students be at or above grade level.

There are exactly three ways to accomplish this:

  1. Wreck the curve, by devoting virtually 100% (far more than 98%) of your resources toward those 50% of students below grade level, in order to bring them up just barely to grade level. This one is theoretically possible, but I’m not sure that our entire country has the resources to do it, nor do we necessarily want to ignore students at or above grade level (and I do mean ignore: we’d need to give them the minimal possible resources in order to devote everything else to the low-level kids).
  2. Set grade level artificially low, essentially 3 (I think) standard deviations below the median, so that 98% of students will be at or above the grade level we set.
  3. Prevaricate, manipulating statistics so that it appears people passed.

Right now, states are frantically doing all three in order to avoid the NCLB penalties. There are no other choices. Grade inflation, in the sense of #2 in which we set the equivalent of a “C” where we really should be setting a laughably pitiful “F”, is virtually mandated by the law.

Now, as for test anxiety. I got some questions.

  1. Are phobias real? I certainly have a nasty phobia of getting shots; when they’re unavoidable, I turn into a humiliating moaning, shivering, writhing, vomiting wreck. It’s the end of a nasty week-or-so process of stressing about it, nightmares, trying to psych myself up, and finally entering the doctor’s office in a state of terror. If I were asked to take a test while I was in that state, I probably wouldn’t be able to say what day of the week it was. It sucks, but it happens.
  2. Is it plausible that someone would have a phobia about high-stakes tests? It certainly seems plausible to me.
  3. Would having such a phobia be triggered be enough to make it very difficult to think through a test? I can’t imagine thinking clearly when my phobia is being full-on triggered: phobias are by definition irrational fears, and they sabotage the ability to think clearly.
  4. Would taking a high-stakes test trigger a phobia of taking high-stakes tests?

If you answer questions 1-4 as “yes,” then I can’t possibly see how you’d find the idea of “test anxiety” to be bullshit. If you answered “no” to any one of them, I’d appreciate an explanation.

I don’t know how common a phenomenon it is, and I’m certain that some people use it as an excuse for lack of studying or lack of cognitive ability. But that doesn’t mean the phenomenon itself is a fiction.

Daniel

73

[QUOTE=ArizonaTeach]
Oh, you have no idea how pissed off I am by this grade augmentation crap, that was supposed to be over and done with. Kids have FIVE chances to take the AIMS test. It’s been something they’ve known about for about a decade – AIMS is a big, big deal here in Arizona. The test is designed so sophomores can pass it, and we have a senior with a 3.2 GPA failing math five times? Something’s messed up, but I really have to say, if a kid can’t pass the extremely simple multiple choice test that they’ve been prepping for since the third grade, symptoms would have shown up prior to taking the damn test.

Aigh. This whole thing bugs the hell out of me.
[/QUOTE]

I’ve never been the type to choke on a test. I am the sort that doesn’t go back and recheck my answers, because that sort of second guessing generally flubs me more than just going with it and being happy. Usually I am one of the first finished with the test, whether I pass it or fail it. When I took the New Mexico Sophomore competency, I remember everyone freaking out about it, and I thought, “How hard can it be? I’ve seen some of the idiots who graduate.” In my opinion when it comes to tests like the AIMS test, being able to take the test is as much a part of the test as knowing the answers to its questions. If you don’t realize that those tests are pretty damn easy, then you’ve got some serious problems that you need to address. I blew through that test like it was nothing, and I got a higher score than anyone I knew.

I definitely agree that life is a series of tests, if you can’t take the test, you’ll be floored when a real challenge comes your way.

74

[QUOTE=Pashnish Ewing]
How does test anxiety compare to perfomance in athletics? Say your kid is on the basketball team, and dominates in practice, virtually scoring at will. But in a real game, couldn’t put the ball in the hoop to save his life. Sure, the kid most likely has game anxiety, but does that mean the coach should keep letting him shoot those bricks throughout the season? Or should lesser talented players get a chance who can actually perform during games?
[/QUOTE]

If these high-stakes tests were an extra-curricular activity–if they weren’t essential for passing one of our society’s biggest hurdles toward material success–then this would be a good analogy. Unfortunately, these tests aren’t just given to the academic elite, those blessed souls like Diogenes: they’re given to every single person who tries to graduate high school. It ain’t a team sport, it’s a necessary passage.

Daniel

75

Grades n’ tests n’ schooling don’t mean nuthin’. They’re jest elitist.

If yer so smart, why ain’t ya rich? :smiley: :confused: :confused: :smiley:

76

Is this thread for tests like the ACT/SAT only? Because I know quite a few people who’ve succumbed to test anxiety over the Law School Admission Test :wink: Granted, it’s a reasoning test through and through.

Anecdotally, I knew just about everyone at my university in the Pre-Med Biology path, and not a single one of the people who were seriously thinking about going to medical school did poorly on the MCAT due to test anxiety. The ones who did poorly were victims of their own unpreparedness.

However, it’s probably a bit different for graduate school tests. At that point, those guys had been through thousands of hours of tough biology, chemistry, math and other subjects that all required testing to pass. The ones who weren’t prepared (or the ones who suffered most from test anxiety) would have undoubtedly been filtered out by the time of the MCAT.

In high school, I don’t think it’s necessarily different. Grade inflation isn’t the case in every instance, after all. Smart people in high school just get used to being in classes with other smart people. The rest of the students might hardly exist. Through that, you might just develop a bent perspective of the personal abilities and motivations of teenage human beings.

Another part of test anxiety is that tests aren’t patterned to account for specialization. This makes sense, to a degree, because specialization isn’t really a focus of public education, but there are numerous opportunities for students to find themselves specializing in a certain field in our current systems. Looking back at high school, I had a huge lean towards “liberal arts” and such. It was easy to take advanced courses in those subjects, but only take the minimum required of math and science. By the time the ACT rolled around, I’d not seriously done any math work in at least a year! I scored high on that test, but only barely passed the math section. The only thing that saved me was getting everything else perfect.

I could see how other people I consider intelligent would choke up in a situation like that. Heck, what shook me up the first time on the LSAT was when it tells you to write out the section about not cheating in cursive! I hadn’t written in cursive writing in so long that I sat there completely stumped for two minutes. It wasn’t a great start to an already stressful test :smiley: If I’d been a person more prone to panic, it might have affected my score and psyche to a dangerous extent.

77

[QUOTE=Indistinguishable]
I suppose I must be missing something. I’d like to see that demonstration.
[/QUOTE]
PM’d to you and Wombat!

78

I think exemptions are not the best approach. But I think test anxiety does exist and I think accomodations are appropriate for some people.

My own lovely Fourth Sister has it and it was a puzzle to me. If you place a test in front of her and she knows it’s Very Important that she pass, she will blow it every time. I studied with her regularly when she was in school and even when she knew the material cold she blew the tests. I don’t understand it myself, her brain just choses flight instead of fight under strss and runs away. Or something.

This was apparent quite early, her standardized testing scores in grammar school were so low that they implied that she was minimally mentally metarded. Only she isn’t, she is a person of ordinary intelligence. Remediation was tried, by means of special “how to take a test” classes and so on, which helped but not much. I think it is fair to say that she has never had her actual level of knowledge reflected on a test unless it was an oral exam.

She learned to cope, to overstudy enough that she could squeak by essentially. She managed to pass the exit exam for high school (mass family sigh of relief) and took a degree in photography at which she is very talented. She now works for the county in CSI of all things.

Her performance, both in school and since, is abundant evidence that her dismal performance on written tests is neither a function of not knowing the material or just being stupid. She just shrug can’t take tests.

I disagree with some of you about the importance of test taking in life, since leaving school a very long time ago the only test I have had to take was the theory portion of the driving exam here in Holland. And that was in Dutch, so my language skills were more to the point than my test taking ones. Other than that, my life has been nothing like a series of written tests. Test taking is an important skill during school years and not after that, at least IME.

79

I believe test anxiety exists, because I’ve experienced it.

Just once.

On one of my Ph.D. comp exams.

I can’t remember why I got so rattled (this was nearly 20 years ago), but I remember sitting there in a state of panic, being unable to think coherently about things I really knew quite well.

Fortunately, the comps at my school concluded with an oral exam, which gave you an opportunity to show that you really did understand the stuff you botched on the written exams.

I don’t have any great insight from this experience about what allowances to make for test anxiety, or whether its incidence is exaggerated, but it’s real.

80

[QUOTE=Diogenes the Cynic]
I guess I can’t see how answering a question requires a skill. You either know the answer or you don’t. I also can’t understand why pressure (I never actually felt this pressure. I can’t ever remember being the least bit nervous about a test) would make somebody not be able to anser a question if they know the answer. If they really do freeze up so much that they can’t remember anything they’ve learned, then I would consiuder that, in itself, to be a cognitive deficiency.
[/QUOTE]

You really should learn to accept that not everyone works like you do. :rolleyes:

I recall vividly when taking the California Bar Exam the fellow student, who I knew had done quite well in classes (B’s or better the whole three years) who I found puking his guts out in the bathroom after the first session the first day. He failed the exam five times running, before finally giving up and doing something else with his life. And I can guarantee that he “knew the answer.” He simply was overwhelmed by the anxiety of having to demonstrate his capabilities on a high-stakes test.

And I love how all the people here say how they think that a high school diploma should be dependant upon passing some graduation test of knowledge, despite the fact that none of us ever had to do it ourselves. Easy to say that something should be required when you don’t have to do it.

Me? I’ve always considered myself fortunate for the fact that standardized testing doesn’t bother me; indeed, I excel at it. But that doesn’t mean I don’t have sympathy for those for whom it is an experience on a par with how I feel when standing at a cliff’s edge. :eek: :eek: :eek:
ETA: And let’s dump that stupid meme in this thread about “life is a series of tests.” Life rarely let’s you know when you are being “tested.” Often, when it does, people sieze up and get afflicted by all sorts of anxieties. Like when you find out your spouse wants a divorce. Or when you find out your boss wants to fire you. The analogy between standardized school tests and “life” is quite inexact.