Gifted Issues Discussion homepage Forums Identification, Testing & Assessment Misunderstanding Achievement Tests

Ever tried to get a diagnosis of ADD or AS for a female?

It presents so differently I'm convinsed that these problems are underdiagnosed in females, especially gifted females.

Girls are 'generally' better at covering up all sorts of difficulties.

((shrugs))
Grinity

If at age 9 my son scores as well as the average MIT freshman on the SAT I and the SAT II subject tests, then I will have him use MIT Open Courseware and other resources to study what those students are studying, at least if my son wants to do so.

I think grade equivalents are more meaningful and useful than Z-scores. If the 9yo scores better than the average 6th grader, maybe he should be studying 7th grade math. If all you know is that he scores 2 standard deviations above other 9-year-olds, what does that say about what he should study?

But...


if I offer a test that is intended to measure... um...



well, nevermind that for the moment.

If I offer my daughter a "test" and she scores as well as the next door neighbor (a fifty year old professional) on it, does that mean that she is qualified to be in that person's profession??



Of course not. That's silly. I didn't even tell you what was covered on the test. What if it's a test for-- er-- color blindness?

See what I'm getting at here?

A test that is INTENDED FOR and NORMED FOR a general population of third grade students is going to contain material that THIRD GRADERS are expected to be learning or already know. Yes?

Then it doesn't mean much to say that a person scores "well" on such a test, does it? Aside from the obvious-- which is that the person is obviously a master of the third grade curriculum.

Out-of-level testing is a different matter entirely. But then, I think that still doesn't mean that the GE's given with many standardized test results actually mean what they seem to imply. For example, if a five year old takes a 3rd grade battery, and scores as a "8.6" on the literacy portions of that battery, that does NOT mean that his or her literacy skill set is necessarily "at 8th grade level" so much as that most 8th graders would be expected to score the same on that portion of that test.

But it does mean that the student is probably beyond the readiness level intended by the tool, I suspect.

Percentiles are, IMO, probably a more useful thing overall-- because those indicate when a student is placed "with academic peers" in a more general sense.

At least that is my understanding-- that as long as a student is still scoring at the 99th+ percentile in out-of-level assessments, it's probably insufficiently challenging.

Thanks for clarifying! I am about to try to get my possibly gifted 9 yr old DD diagnosed with inattentive ADHD as it happens :-). And dyslexia. And CAPD.... Which is why I was making sure that is what you meant and that there wasn't something further I was missing...

I think stating that achievement test grade equivalents are of absolutely no value seems like a pretty broad and therefore erroneous generalization.

Surely some of the test GE numbers are more solid than others.

I am not sure I understand the conversation; I am not aware of a large contingent of people who would immediately equate GE=X results to meaning a child needed to be moved up to that specific grade. Nor would they anticipate their child being moved down if they score low.

Surely a difference of even a few years between a child's scored GE and actual grade will indicate a problem in academic fit. If ds was in 4th grade and score ge=2nd grade in reading I would be FREAKED.

What are we worried about here? Pushy people giving a bad name to folks who genuinely need a grade skip or two for their kids?

I don't want to read too much into a couple lines, but this sounds like falling into the calculus trap. Normal math curricula and tests like the SAT/ACT are very shallow. Going deeper may be better than going faster.

http://www.artofproblemsolving.com/Resources/articles.php?page=calculustrap

I would have assumed, incorrectly, that the score meant they did as well on their grade level questions as a student from the achievement score grade would have done. I know, I just have to adjust. "Time marches on".

Our district doesn't use STAR math, but they do use STAR reading, and DD's grade-equivalent scores are a good 4 grades above her actual reading level.

I had previously read this essay and do not agree with it. Most people who are very good at math are not going to be mathematicians but will use it as a tool. Calculus is a useful tool, and if a student takes calculus earlier, he will be able to take a "real" physics or economics course earlier. I have bought an AOPS algebra book, and if my son uses other AOPS books, including the one on calculus, that should alleviate the problem of alleged shallowness.

Actually, I think the bigger issue with the idea that the 9-year-old might be ready for MIT classes by virtue of having similar scores on SAT's and other tests is this:

The SAT's have a pretty low ceiling, and fail to capture a great deal of the knowledge that I would expect an MIT student to have. SAT math is not very high-level math. I would bet that most of those MIT students are adept at several years worth of math beyond SAT level and have a much deeper and better understanding of the topics that are on the SAT, plus many more.

Put me in the camp that is distinctly unimpressed by perfect SAT scores. Yes, if your 9-year-old gets a perfect score on the SAT he or she will need some pretty radical acceleration, but I'll bet there are quite a few of those MIT students who were exactly the same.