Gifted Issues Discussion homepage Forums Identification, Testing & Assessment Different versions and uses of CogAT

I'm not sure if anyone here would know, but thought I'd throw it out there. I know that the CogAT for younger grades (like K-2) is not timed, but is it always timed for 3+? The reason I ask is because our school district is using it as a test to determine cognitive ability for special education eligibility. If there is a large gap betweeen cognitive ability and achievement, then the student could be classified as LD. I don't understand how they could give a timed GROUP test to kids who potentially have disabilities like ADHD or autism or certain learning disabilities like dyslexia and wouldn't do well with a timer. The district special ed director thinks that if the test is given to one child at a time it's "individually administered" and therefore legal. It's possible that they are giving the test untimed with the questions being read to the student, but then I don't see how the results would be valid. I'm trying to put together something for the superintendent and/or school board but want to make sure my facts are straight. One parent asked if her child being evaluated could take a normal IQ test instead and they refused. Has anyone ever heard of this being done in other schools, or a similar group test being given that is normally used just for gifted screening?

I was able to get extended time for my son when he took the CogAT for GT identification.

ETA: This was the the CogAT intended for 5th graders that was administered to 3rd graders.

When DD took the CogAT6 version as a screener in 2nd, I got the sense from what she told me afterwards that it was read to the class. Also, the kids sometimes called out for the teacher to stop or slow down, but she could not always do this. It sounded like the teacher was sticking to a strict format. This format sounds like it would be a nightmare for many 2e gifted kiddos - it would probably miss them completely.

Our schools use this test for the screening phase of G&T selection only. There are 4 criteria for screening and a child needs to qualify in 3 areas to more onto the "selection" phase. So technically, a kiddo could do poorly on the CogAT and still qualify on the other 3 areas to make it to the next "round." I believe that this is more how the designers of CogAT suggest that it be used.

Our students need to be reassessed for G&T before 6th grade (I guess to see whether they are still gifted? LOL). So they all take the CogAT again in 5th. For some reason MANY more kiddos do well on the 5th grade test - I have no idea why this is so.

I think that might be how it was with DS--the questions were read to him but he still said he felt rushed. DD was given the third grade version and it was student-paced with a time limit. She did horrible because she was so slow and left so many questions blank. I asked the g/t coordinator (who isn't there anymore) if DD could take the CogAT untimed the next year, and she said No, that it ruins the validity of the test, but she could take a different test, like an IQ test (then she proceeded to quit 1 month later leaving me to figure it all out on my own). I'm wondering if ALL 3+ versions of CogAT are timed according to the test manufacturer recommendations. I think if it is given to kids as recommended it can pick up discrepancies or the fact that they are slow, indicating a possible 2e issue or disability, but it's not going to be a true measure of cognitive ability if the kid can't handle the timing due to their disability. For DD it picked up the fact that she is incredibly slow and that's about it, it didn't pick up the giftedness (her WISC score was a lot higher). So the fact that the district is using this to measure the gap between ability and achievement (which needs to be 1.75 standard deviations) doesn't make sense. If a kid scores 100 on the CogAT because they can't handle the timing and they are only slightly lower on achievement, it's not going to show the gap they want to see for a SLD. I also think it's illegal for schools to give a group-administered test for spec. ed evaluations. Even if it's not given in a group it's still a group test.

Loy--I think our district gives the CogAT repeatedly as well, for instance when they transition to junior high. They refused to put "no timed tests" into DD's 504 and now I think I know why. They are leaving the door open to give her the CogAT when/if she needs to be re-evaluated.

blackcat - it sounds like one or both of your DC should be getting accomodations like Kai's DC. ADD/ADHD Dx or similar exceptionalities would sound like they absolutely justify more time on this test. Then again, I'm not sure how well CogAT EVER works for a 2e kiddo (anyone know?).

What's most troubling in your scenario is that it sounds like your district uses it for SELECTION - not simple screening. The CogAT is simply not an individual IQ test. I think your best argument for how CogAT is not appropriate lies in the test maker's data itself. I cannot recall where I saw it, but I seem to recall that CogAT is best used as a screening device.

I think it's used completely inappropriately for gifted identification (because they have a cut score of 139 composite), but it's even worse that they are using this to qualify or disqualify kids for special ed. If they use DD's CogAT score she probably would not have the discrepancy they need according to State law, but if they use her private WISC score, there is going to be a large gap between the WISC and Woodcock Johnson or WIAT writing clusters (I don't have her eval results yet so we'll see). If they use her WISC score they would need to use her GAI of 150 (I found a State document stating this). But with the CogAT the timing was a big factor dragging her score down. I have no idea how they are even going to figure out what is 1.75 standard deviations below 150 because the regression table only goes up to 129 IQ.

The standard deviation for the WISC is 15, so 1.75 standard deviations= 26.25. 150-26= 124. Or am I missing something?

This happens in our district too, and I've always wondered why - first grade has hardly anyone in GT, which slowly rises to about five or six in third and all of a sudden twelve to fifteen in fourth and fifth (though I think the percentage stays the same in middle school). This is for a pullout one day a week. I think an IQ test is the largest part of the puzzle, so there's retesting issues, and the later kids are noticeably 'less gifted' or more hothoused - perhaps because in our district kids can ask to be reevaluated each year I believe, so maybe they're actually violating retesting regulations?! But it's definitely strange - I hope this isn't too far off topic, sorry.

blackcat, my ds was given an untimed CogAT with the questions read out loud to him when he was tested for the gifted program - he was given those accommodations because they were written into his 504 plan.

I would *not* consider the CogAT a valid IQ test of the type that is needed for a SLD/SPED. Rather than advocate from the point of timed vs untimed and group vs individual test, I'd focus on the type of ability test it is - the WISC/WJ-III Cognitive/etc are *innate* ability tests, the CogAT test is a learned ability test. As mentioned above, I'd look to the CogAT documentation re the intended use as a group screener rather than a test to use as part of a diagnostic package.

There is (possibly) a potential upside for some students to use the discrepancy gap and CogAT vs achievement - if a student is struggling and has a gap between CogAT and achievement tests that meets your school district's guidelines, they can be qualified possibly more expediently than requiring a full individual IQ test be administered. The downside to that, of course, is that getting support for academics as quickly as possible is important, but truly understanding what is behind the issues that are resulting in the academic struggle is very important for the long term success of any support/interventions. I suspect the individually administered innate ability IQ tests are more beneficial in terms of understanding root cause of the issues, but even those can't tell you the full picture (usually) - our experience has been that they can show you that yes, there is an issue - but more testing in the area of the potential issue is usually necessary to understand it (things like executive function testing, OT or SLP, vision, etc).

Good luck with your advocacy!

polarbear

Discrepancy in predicted achievement, properly done, is not based on a simple standard deviation difference. Firstly, both instruments have standard errors, so you have to take two sets of standard errors into account. Secondly, in linking studies (when available), it is not unusual to find more significant regression to the mean than would be expected simply by examining the range of overlap of the two confidence intervals. Such linking studies are available for the SBV/WJIII, WISC-IV (and -V)/WIAT-III, and WISC-IV (and -V)/KTEA-3. A true predicted achievement standardization set is available for the WJIII/WJIII. Correlational tables derived from unlinked standardization data are available for the WISC-IV/WJIII.