KarenNC Posted June 20, 2009 Share Posted June 20, 2009 My daughter took the ITBS and CogAT for her standardized testing this year. Last year, based on her performance on the WJIII achievement test previously, we gave her an out of level test (4th grade test while in 2nd grade) and she came out with a composite grade equivalent of 5.2, percentile rank 57, which is about where we thought she would be overall. This year, because we planned on applying for the Duke TIP next year, we gave her the third grade test (she's in 3rd so we wanted age level) and the CogAT for our own info. Her composite GE came out at 6.1, PR 93, but there were some definite oddities in the subtests. In vocabulary her GE dropped from 7.9 to 6.6, spelling 6.0 to 4.5, language usage and expression 13+ to 8.9, sources of information total from 6.2 to 5.3. I could see that she might not progress a year worth in all areas, but to have such a dramatic drop is odd to me. Is this usual? Is her actual level somewhere between the two? In the CogAT, how much does not attempting all questions affect integrity of the scores? She was given a profile of 8E (Q-), so there was an extreme discrepancy between verbal/nonverbal and quantitative. On verbal she attempted all the questions, in nonverbal 63 out of 65, but only 51 of 60 on quantitative. Her PR were verbal 95, nonverbal 89, quantitative 57 (this was on the B version of the test, recommended by the publisher for average to above average 3rd graders, but not the one most of the testing companies send out for this age group). The scoring profile recommended checking the integrity of the scores, but I'm not sure I can do that without purchasing their manual for teachers. I was not surprised that she was lower in quantitative, as math has consistently been her worst area overall (as well as the capitalization and punctuation in language). I am concerned, however, whether this large a discrepancy is indicative of a possible relative learning disability (for which we should pursue other testing) or partially an artifact of the number of unattempted questions. She did score about a year ahead of grade level in math on both years on the ITBS (3.4 in 2nd grade, 4.8 in 3rd grade), but much lower than in all other areas. Any ideas? I have tossed around the idea of formal IQ testing for a while, but it is pricey and we didn't have a clearly defined need to spend that money for a score. If I need to identify and deal with a possible learning disability, however, that would be a different story. Is it worth purchasing the manual on interpretation for teachers (about $25 that we really don't have extra at the moment) first? Quote Link to comment Share on other sites More sharing options...
EKS Posted June 21, 2009 Share Posted June 21, 2009 When you tested out of level, the ceiling was much higher, and so she could get the higher scores. Giving her the on level test lowered the ceiling so when she missed any questions it lowered her score substantially more. What would have been better to do is to give her the 5th grade ITBS and have it scored against 3rd grade norms. Then the ceiling effects would be less pronounced and you would still get the percentile rank against other 3rd graders. As for the CogAt, I wouldn't put much stock in the results, except as general trends. IQ testing is much more informative. Quote Link to comment Share on other sites More sharing options...
KarenNC Posted June 21, 2009 Author Share Posted June 21, 2009 When you tested out of level, the ceiling was much higher, and so she could get the higher scores. Giving her the on level test lowered the ceiling so when she missed any questions it lowered her score substantially more. What would have been better to do is to give her the 5th grade ITBS and have it scored against 3rd grade norms. Then the ceiling effects would be less pronounced and you would still get the percentile rank against other 3rd graders. As for the CogAt, I wouldn't put much stock in the results, except as general trends. IQ testing is much more informative. I stepped back to the grade level test figuring that it would show up the higher percentiles for the Duke application. So it sounds then like I actually did exactly the wrong thing? She still has the scores to get in at this level with her verbal scores anyway, so at least I didn't totally shoot us in the foot:tongue_smilie:. Last year was my first experience with the ITBS---we'd used the WJIII before and it is done so very differently, including scoring. I didn't know you could ask that the ITBS be normed against a level other than one chosen. I will have to ask about that next year. It was a bit of a chaotic time for us this year anyway as our regular provider was disqualified to sell the tests at the last minute and we had to scramble for another source. Then I didn't exactly endear myself to the new provider by asking them to send me the B level of the CogAT (publisher recommended for average to above average 3rd graders) rather than the A level (recommended for average to below average), a choice I wasn't aware needed to be made until I started reading the directions on administration. Her response that basically homeschoolers were too ignorant to be able to read a chart and pick the right level, so the testing company didn't want to go to the trouble of offering them choices didn't sit very well with me. Oh well, guess it was a good lesson in buyer beware. I had only read the provider's website info, not the publisher's, or I would have been aware of the potential issue at the beginning. Quote Link to comment Share on other sites More sharing options...
EKS Posted June 21, 2009 Share Posted June 21, 2009 I have used BJU's testing services in the past and they have always been very accommodating of my "odd" requests. They will even hand score against one set of norms and then send it in for machine scoring against the other set. The CogAT also has a low ceiling and rewards children who are fast with their answers. A child with issues with being timed, for whatever reason, is not going to do as well on the CogAT as on an individually administered IQ test. It's good that you went up a level with the CogAT (level B is intended for 4th graders). Quote Link to comment Share on other sites More sharing options...
KarenNC Posted June 21, 2009 Author Share Posted June 21, 2009 I have used BJU's testing services in the past and they have always been very accommodating of my "odd" requests. They will even hand score against one set of norms and then send it in for machine scoring against the other set. The CogAT also has a low ceiling and rewards children who are fast with their answers. A child with issues with being timed, for whatever reason, is not going to do as well on the CogAT as on an individually administered IQ test. It's good that you went up a level with the CogAT (level B is intended for 4th graders). I'm guessing that they charge twice for the double scoring? She definitely is not a speedy tester when it comes to math. We've been working on getting multiplication facts memorized, but it is a bit slow going. Quote Link to comment Share on other sites More sharing options...
EKS Posted June 21, 2009 Share Posted June 21, 2009 They do charge extra for double scoring, but I think it wasn't quite double. If she is not speedy with math then the quantitative part of the CogAT will not reflect her true potential (except the potential to take timed tests!). My son who scores 99+ percentile whenever he is given a standardized math test *untimed* (like the WJ-III or ITBS with accommodations for time) only scored at the 80-something percentile on the CogAT. Every problem he attempted he got right, but he ran out of time. I really don't like the CogAt as a test for giftedness. When given on level it has a low ceiling and for 2E kids, it doesn't show their abilities at all. Our district uses it for entry into their gifted program (nonexsitant as it is) and will not accept any other scores (like the WISC) for entry. 2E kids who are unidentified as either gifted or LD (because they are compensating and averaging out to average) don't have a chance. It makes me so mad! Quote Link to comment Share on other sites More sharing options...
TravelingChris Posted June 22, 2009 Share Posted June 22, 2009 I just used the Iowa and Cogat this year. My mathematically gifted child, usually scoring 99 percentile on conceptual math and very scientifically inclined, scored very low (47) on her quantitative and lower than her not mathematically gifted but older and faster on computation sister. I agree that the test doesn't measure IQ or giftedness. This kid solves math puzzles for fun and the test says she is below average???? I don't think so. Quote Link to comment Share on other sites More sharing options...
klmama Posted June 24, 2009 Share Posted June 24, 2009 Did you go to the CogAt site and enter the info to receive a description of this type of student? If so, did you think the description of your dd was accurate? She may not have a learning disability at all, but simply have less skill in math at this point. That may change as she practices more. Quote Link to comment Share on other sites More sharing options...
TravelingChris Posted June 24, 2009 Share Posted June 24, 2009 I just went to the site. It showed my older to be high average on verbal although she has always been much above average in writing and reading once her eye problems were solved. So basically it showed her too low but correct on the part which she is most talented. On the other hand, my youngest is topping out math concept sections and is definitely much more math oriented and talented than her sister. However she scored lower than her sister on both and they said she was equally good or bad at everything. That just isn't true. As the interpretation said, N is significantly higher on all the non Language Arts and Reading portion of the test than predicted. That is because the test was not accurate. Now I will have to decide when we will have enough money to test her accurately with a professional. Quote Link to comment Share on other sites More sharing options...
frogpond1 Posted June 29, 2009 Share Posted June 29, 2009 are the best way to interpret these scores. Quote Link to comment Share on other sites More sharing options...
KarenNC Posted June 29, 2009 Author Share Posted June 29, 2009 are the best way to interpret these scores. Suggestions on a better way? :bigear::001_smile: For me, it was not the specific grade equivalent or that it's a grade equivalent per se, it is the dramatic drop in grade equivalent between what she scored as a 2nd grader on a 4th grade test and what she scored as a 3rd grader on a 3rd grade test from the same company. I would have expected that the same measure would be more consistent between levels in the same product. Quote Link to comment Share on other sites More sharing options...
frogpond1 Posted June 30, 2009 Share Posted June 30, 2009 are better to watch, but there is a margin of error in those too. I just know that Grade equivalents are a little fickle as indicators. Quote Link to comment Share on other sites More sharing options...
frogpond1 Posted June 30, 2009 Share Posted June 30, 2009 test probably is norming things according to her birthdate. I'd do better seeing the whole test score sheets to get it all figured out. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.