Gifted Issues Discussion homepage Forums Identification, Testing & Assessment Erratic scores: which ones do you go with?

My youngest is 2e, which likely accounts for her rather erratic test scores, but I am wondering which scores one usually "believes" as accurate when achievement scores go up and down regularly.

For instance, on NWEA MAPS, her numbers may go up or down by 20 or 30 points from one testing session to the next. I recall, in 2nd grade, she was quite high in both reading and math in the spring, and then tanked in 3rd with her fall and spring #s lower than her spring 2nd grade scores.

This year was another one of those with fall scores high, but not outstanding and a bit lower than the spring prior. In reading, she came up something like 11 points from fall to spring of 6th grade, which her teacher is seeing as amazing growth, not seeming to notice that the #s are no higher than they were at the start of 5th, just that they had fallen a lot in spring of 5th and fall of 6th grade.

Math, if she's giving me the right #, came up more like 25 pts from fall to spring of 6th. She was higher in math than reading in the fall and may be really quite high on math for spring, again if she has the # right.

What we've been told in the past is that, whenever her scores are high, she has guessed well (we've heard that a lot re IQ, MAPS, and various other tests). I've also read on NWEA's site that it is impossible to guess well and get a too high #. However, if the # is high and then goes down a bunch, which # do you believe is more accurate?

I'm not always sure that the highest number is the most accurate either, though. Thoughts?

With a 2E kid myself, how I look at this is that the high number is real - it shows the minimum that your kiddo can do on a good day when everything lines up for them - real ability might even be higher. Ideally, you want to work with your child on developing strategies, coping skills, and accomodations to get to the point where most days are good days, achievement-wise. The erratic nature of the performance is telling you that the disability is real and has real effects, and the magnitude of the swings lets you know the minimum impact that the disability has on a bad day - it could be worse, because you don't know for sure that the low score is reflecting a worst-day scenario. So the combination of results tells you the floor of the ceiling and the ceiling of the floor, if that makes any sense. Working on bringing the floor up closer to the ceiling is the goal of IEPs and 504 plans.

One of my boys is 2E, and our school uses a STAR test for reading. His scores go up and down 1/4 to 1/2 a grade level each time. When I expressed concern about that to the teacher, she said every child's scores vary like that, but as long as they're on a strong upward path, there's no concern.

My other son, who has dys scores, varies as much as a grade level (but his start 7 grade levels ahead), and it's not a concern.

I would try to look at the trend, rather than the actual score.

PS I also wanted to add that being tired really affects my 2e son's scores, as does his mood and what he's eaten. When we had a private tutor, she would try to only test him when she could tell he was having a great day, and that really affected his scores. Obviously that can't happen in a school setting.

Even if a child is in equally good test-taking form for each test, there are random errors in test scores, which argues for averaging the scores, unless you have strong reason to believe that one set of scores is unrepresentative.

Cricket, our experience with our 2e ds is that achievement test scores can be erratic for different reasons. I am not familiar with MAP testing, so I can't help you with that, but fwiw, my random thoughts:

1) Our ds has had experiences where we know his scores don't reflect his knowledge, but other odd little quirky things happen related to 2e-ness. For instance, he had an extremely lower-than-expected score on one test which, we learned after that the school did not give him his accommodation of writing in the answer book, and his neuropscyh suspects he got off track following the bubbles to fill in on the answer sheet. We were told on his state testing last year that he missed answering some of the questions in the booklet because he missed the pages while thumbing through the book.

2) Timed testing is sometimes an issue for our ds, sometimes not - but when it's an issue it can be a huge issue (as in, he has had tests where he has answered every question correctly but only answered half the questions because he ran out of time).

In the cases of 1 and 2, we (or his school staff) went back and looked at answer sheets where we were able to and also talked to the test proctors to get an idea of what had happened. We questioned the scores, went back and looked etc because we knew that the scores didn't reflect what we (teachers, parents) knew about ds in terms of the academic level he was working at and academic knowledge he had at the time. So - for a 2e kid - when a test doesn't make sense, it can help to go back and look at the actual test, as well as look at things like the environment etc (was he tired, was the room noisy etc - whatever impacts your child).

Next thing - when achievement test scores are up and down and up and down, yes, I would tend to either believe the higher scores and look for test-specific things that might have impacted the lower scores as long as I didn't have any questions about how my child was performing in general in the subjects tested. However, with our 2e ds we've also had a situation where achievement scores in reading fluency have decreased over time. They were very high when he was very young and they are still much higher than average, so on the surface they don't look worrisome to a teacher. They haven't flip-flopped up and down, but instead just gradually gone down when you look at his percentile score. He also has some odd quirks associated with reading - he clearly prefers to learn via video or auditory and he has difficulty summarizing what he reads. So - there, we have test scores going down with time plus things we are seeing in real life that are puzzling re reading, so in that case, I tend to believe the test scores are relevant and we're trying to look into what's up.

polarbear

This is just a guess on my part, but I suspect the type of error you're referring to Bostonian is much smaller than the type of erratic score differences that parents of 2e kids see on testing. The example I posted about my ds getting off track on the bubble sheets was a score that was 40 percentile points lower than what we would have anticipated his score should be. With a 2e kid, it isn't terribly surprising to see errors much larger than the standard error in the test due to challenges that have nothing to do with the child's knowledge of the subject matter being tested.

polarbear

FWIW, dd did have the right # for her math MAPS b/c they just posted the official scores. She's really high at this point. Math has been her more consistent score on MAPS although it too fluctuates, just not as much as reading.

I'm not talking about little variations that fall within the 90% confidence interval of the standard error. I, and the OP, are talking about really wild swings. If your child has a known disability that impacts testing performance, you have reason to believe that all of the scores might be unrepresentative of their knowledge, honestly.

For us with NWEA, the testing environment dictates which score I count on more than anything. For example, my son's language usage went down 24 points this last test. After asking what happened and discussing it with him, I discovered that a 1st grader sat next to him and was taking the syllables test at the same time. The kid was clapping out syllables for 20 min of the 30 min test. My kid's brain was mushy at that point :-)

If it's a disability, then you want different testing accomodations for sure. But we have seen wild swings that vary just based on what the teacher says. Things like "You're the last one so hurry up!" and "When everyone is done we'll go have snack!" make a major difference in NWEA testing.

Also, the test starts down one level from the previous score. If the child misses those first three questions, for whatever reason, the test drops down an additional level. From what I've been told, if they drop down after the first section, they cannot get back to the previous high level because they've technically dropped down two levels.

So there are lot of variables! I pretty much just look at growth across three tests at any point and time. If you take three tests and they met the growth goal, then good. If not, then why?

Oh and one more thing, from my observation of several HG+ kids scores, they tend to score well in the fall and their scores drop slowly over the year. They jump back up again in the fall. I have no research to back this up, but I think it's the summer-affect on gifted kids.