Gifted Issues Discussion homepage Forums Identification, Testing & Assessment Is the PSI of WISC-IV computed correctly?

Dottie or anyone else who can figure out the WISC-IV

I'm having trouble with the low percentile of my daughter's PSI. I always knew it was average, but this percentile seems wrong.

The psychologist got her birthdate wrong and a few other things, so I'm wondering if it's an error.

VCI 134 99%
PRI 133 99%
WMI 110 75%
PSI 94 34%
GAI 139 99.5%

It seems odd the percentile of the PSI is so low when the WMI is in the 75% and not that much higher point-wise.

Thanks for any help,

Lucijane

My DD's PSI of 94 is also listed at the 34th percentile so that looks right to me. Dottie will explain this more coherently, but there's a big bulge of scores between 90 and 110, with 100 as the mid-point/50th percentile. A standard deviation is 15 points above or below that--most people fall somewhere in this range of 85-115. (Dottie, please correct as necessary!)

Gifted kids can be funny, with scores in one or both of VCI/PRI at two or more standard deviations above 100 (130+) and with scores in WMI and PSI closer to the norm. The NAGC website has some good information on this, and recommendations for using the GAI rather than FSIQ for these types of kids.

FWIW, we spent some time worrying about this spread and threw in a bit of occupational therapy, but it's pretty clear at this point that the relatively slow processing speed and average WMI that my daughter has only slow her down in a small number of tasks and can be compensated for fairly easily. We've also wondered how much of this is "trainable"--apparently PSI improves with video game and other activities that train kids to have more rapid finger responses.

I don't know on your original question, but did she correct the birthdate issues? If she was off by any significant amount, that would change her scores/percentiles.

Westcoastmom is right about the percentiles.

In a normal distribution, you'd expect 68% to fall within one s.d. (standard deviation) of average (100). Since these tests are designed to have s.d. of 15, only 16% of the population scores below 85 and only 16% scores above 115 - on each individual subtest. If 85 = 16th percentile, it makes sense that 94 = 34th.

But get the birth date fixed before you worry about an average subtest score. besides, processing speed isn't a g-loaded scale.

Thanks everyone. He got the birthdate wrong by six days so it won't make much difference.

Dottie, I do remember that the GAI was originally calculated at 139.95 when he showed me the scores during the break.