Gifted Issues Discussion homepage Forums General Discussion Quick question about IQ scores in books

Hi, I'm new to this forum and have a (hopefully) quick question. I am doing some reading on parenting gifted children, and I notice that every time an IQ score is mentioned in books (and usually online as well), the author doesn't say for which test the score applies. But when I look at this page on hoagies, I notice that the scores vary widely between tests for similar percentiles or 'levels of giftedness'.

So when an IQ score is given without mentioning a test, to which test could I assume it is referring?

Would it be reasonable to look at the hoagies chart as a rough conversion chart between tests?

Thanks!
Dawn

I think you have to consider when the book or article was published. Scores in older books are usually SB L-M scores.

The most commonly used IQ test now is the WISC-IV, I think.

I'm not an expert by any means, but what I do is see how high the score is and how they're treating it as far as level of giftedness. Anything over 155ish is probably SB L-M. Even though the WISC-IV has extended scoring, it is not widely used yet for score reporting (I don't think).

If it's a score that's harder to gauge like between 140-155ish, then see how impressed the author seems by it. If they're talking about it being a more moderate level of giftedness, then it's probably SB L-M. If they treat it like an extremely high score, then it's probably one of the modern tests.

Thanks for the quick replies! Between these suggestions, I think I can figure it out. Wouldn't it be simpler if everyone referred to the percentiles instead? Would that even make sense? I noticed that the Mensa organization required their members to be above the 98th percentile on any of the standard usual tests in order to eliminate confusion about the actual scores.

Do some authors refer to "highly" gifted when they really mean EG or PG? These terms all seem to be thrown about much to casually.

Thanks!!
Dawn

Oh yes yes yes!

The problem with percentiles is that unless you go out to several digits, they sort of lump HG, EG, PG all together.

I think I'd actually be very cautious about books that talk in terms of high IQ numbers without specifying what test they're talking about - it's a red flag if the author doesn't realise it's important! Which book(s) did you have in mind?

Cathy, I understand that they would have to add in the decimal digits since HG seems to start at 99, and PG seems to start around 99.9? Such a difference within that one percentile point.

ColinsMum, It may be that in my sleep deprived state I just couldn't find where the author indicated the test used. (In addition to my DD5, I have baby twins that seem to think that sleep is for wimps.) I am reading A Parent's Guide to Gifted Children and many, many websites. The book was recommended to me by the psychologist that administered my DD's test. He also recommended Hoagie's Gifted website to me, so I'm hoping I can assume it is well regarded to be accurate.

As mentioned above, MG, HG, EG, and PG have no standard definition. The big problem is that score reports often list the percentile as ">99th percentile" or ">99.6th percentile" etc. So people don't really know the precise percentile anyway (assuming that kind of precision would be meaningful when you take the test ceiling into account.) Current tests are simply not designed to differentiate between these groups (however they are defined) whether the scores are reported as standard scores or percentiles.

If you want to figure out the percentiles from the standard scores you can use this

Score Conversion Table and this Normal Curve Calculator

For example, if I have a standard score of 149 for a test with a standard deviation of 15, I can look in the table to find the corresponding Z-score (3.27) Then I can put that into the normal curve calculator to get a cumulative area of 0.9994622 which corresponds to a percentile rank of 99.94622

Have fun with numbers

If your test has a different standard deviation or mean, the formula for the Z-score is:

Z = (SS - mean)/sd

where SS is your standard score, and sd is the standard deviation. Z is the number of standard deviations by which your score differs from the mean (usually 100).