Gifted Issues Discussion homepage Forums Identification, Testing & Assessment WISC testing?

Long time lurker here; new poster.

DS(6.5) is getting to the age where we would like to test him. We suspect he will be at least HG-plus. I have read (I think on this forum) that the new WISC is coming out this fall and that there is significant difference between scores of those who test when the test is new verses at the end of its test-life. The question is-- when should we test? If we do it now, his scores will be inflated? But if we wait a few months the scores will be deflated. I don't want to wait years just to get a "more accurate" score. Why do the scores fluctuate? Is it only because the answers/test format leak out over time? I'd like (and honestly easily expect him to) qualify for Davidson, but I am nervous about the timing of the new test.

Also we are planning on testing at local university (we are NOT in Montana now). Will this be a good place to do it? We can't spend thousands right now to test. I'd like to keep it under 500$. I've been told we may get a worse score if they don't have experience with gifted kids. We are in the mountain west but can not afford the Denver gifted center testing. Advice?

To clarify, the -most- accurate IQ score is obtained in the first year of a norm's life. All scores after that are slightly, and increasingly, inflated. It's not that there's a sweet spot a few years down the line when the scores are accurate. The -least-accurate scores are obtained on a test that is being replaced (or afterward, obviously). The reason some people want their children to be tested on nearly-obsolete norms is that they suspect that their children are borderline for qualifying for a higher classification, and are taking advantage of score inflation to artificially boost their apparent IQ scores.

A more innocent reason to use the old WISC-IV when the WISC-V is already out would be that the examiner is experienced with it, and might be more skilled at administering it, interpreting it, or both. A reason for giving the WISC-IV that has nothing to do with the client is that the new WISC-V test kit costs around $1000 to purchase, and maybe you as an evaluator aren't in a business position to incur that expense immediately, or you still have a pile of WISC-IV record forms lying around, and tossing them unused is a bit painful, at about $10 a pop. It is considered ethical, though not optimal, to continue administering an old test for up to a year after the new version has been released. And for kids near the population mean, the Flynn effect (norm obsolescence that results in score inflation) is not marked enough to affect high-stakes decisions (estimated to be 0.3 points per year, or 3 points per decade). There is some research that indicates that the Flynn effect is more significant further away from the mean, such as for the gifted population.

If you want more detailed info on score inflation over time, google "IQ scores Flynn effect", and you'll find several articles on it. But, in short, no one really knows why (although you will find a number of ideas floating around in the field), but IQ scores have been steadily rising over the past 100 years or so (which is how long the history of IQ testing is), at a rate of roughly 3 points per decade. As modern IQ scores are deviation IQs, meaning they represent your relative standing in a population of your age and cultural peers, this score inflation forces publishers to re-standardize tests periodically, so that you don't have more than 50% of people performing above the 50th %ile, which would make no sense, statistically. Various theories have been suggested, ranging from the idea that people in the industrialized nations are actually getting more intelligent on the average (generally dismissed by the most reputable researchers), to changes in what is considered common knowledge, as cutting edge knowledge or technology becomes everyday. An example of this that gives face validity to this reasoning would be the difference between a 70 year-old struggling with using a new smartphone, and a 2 year-old's seemingly natural facility with an iPhone. Is it true that the 2 year-old is more intelligent than the 70 year-old? Not necessarily, but their prior experiences and cognitive expectations/schema are quite different, which affects how they approach the same problem-solving situation. This is part of what is captured in using the most recent norms possible. When you use old norms (WISC-IV), you are comparing your six year old to the population of six year olds who were born in 1996 (currently 18 year-olds). You may contemplate whether you think their experiences and cultural context are comparable.

According to the publisher, the new WISC-V has a more robust floor and ceiling, so it should be better at distinguishing both the top and the bottom ends of the population. The factor structure of the WISC-V is also more in line with current intelligence theory. All other things being equal, I would go with the WISC-V (but then, I am an evaluator myself, and I like new tests!).

Depending on your state's regulations about giftedness, you may be able to ask your local public school district to test at no or minimal cost. In some states, giftedness is considered a special population, with educational protections. Check your state DOE's website for info on gifted ed in your state. If you go the private route, you should be able to get an IQ only for under $500, but not much under, depending on where you are. Thousands of dollars usually comes in if you want other things, like achievement or neuropsych testing. If your local university has any research projects about giftedness, the graduate students administering the tests will have had some training on evaluating the gifted, and will be supervised by faculty members. But you will probably be better off if pre-doctoral students who are already master's level-qualified, or post-doctoral fellows working on licensure, are giving the tests, than if you have second or third year grad students.

Aeh,
Thank you so much for clarifying. So it looks like going with the Wisc-v is the thing to do. It sounds like it maybe out already? I have not contacted the local university yet. I am hoping to get the WISC and the WJ or similar achievement testing. Hopefully I can find it inexpensive enough. I don't think our state does testing--besides we homeschool.

About how much variance in score do you think you will get between using a experienced tester with gifted children and an inexperienced one?

Whether you homeschool or are enrolled in your local public school, your children have a right to any gifted or special education evaluation resources that other residents of your town have. Whether they test for gifted is another matter, having mostly to do with funding (or lack thereof) for Javits.

You can get substantial variance between overall experienced and inexperienced testers, mainly having to do with knowing when to query ambiguous responses, and in maintaining an appropriate and engaging pace. Also, some newbies have not yet mastered neutral responses to children, that don't telegraph the strength or weakness of the performance. Depending on the child, these tells may discourage or encourage them...or merely be distracting.

The difference between experienced testers in terms of their specific experience with gifted children mainly affects engagement, pace, and interpreting low-frequency responses. Gifted kids often come up with responses that are correct, if you think about it differently, but are not in the manual. An examiner unfamiliar with divergent thinkers may score it exactly according to the manual, and just move on, rather than probing to see what the reasoning is, and how it might actually be a higher-level correct response. Many gifted children are more perceptive about the examiner's own level of engagement, skill, confidence, or intelligence, which can affect their willingness to produce optimal performance for the examiner. In rare cases, you may encounter an examiner who is vaguely resentful, or insecure, about testing a child who may be markedly more cognitively-able than themselves. Obviously, that would have implications for the validity and accuracy of your results.

The pre-publication purchase price for the WISC-V is good until October 18th, so I would imagine they are planning to start shipping around then.

There is no retest period between -different- editions. Just if you are re-testing using the exact same test. E.g., WISC-IV to WISC-IV. So if your child just had the WISC-IV a few months ago, you could certainly have them re-tested on the WISC-V this fall.

Do you have any idea what the floor and ceiling of the test is likely to be? It seems that every time a new test comes out the floor and ceiling increases by 5-points (e.g., 40 to 60 on the WISC-IV may become 35 to 165 on the WISC-V)?

It's not totally clear how the IQ/Index ranges will change, as they don't appear to have released that information yet, but there have been efforts made to strengthen the subtest floors and ceilings, which can only be to the good, in the event that extended norms are developed. Keeping scores in perspective, if you read the WISC-IV Technical Report #7, which has the extended norms in it, you'll note that exactly one child in 2,200 members of the standardization sample obtained a GAI of 151, and none obtained an FSIQ of 150+. So from a general utility standpoint, there is no particular value in going above 160 on the FSIQ chart, but maybe the stronger subtest ceilings will allow for a better extended norm chart.

I asked because I believe it was the N.A.G.C. that requested extended norms to differentiate gifted children. In a sample of 157 moderately gifted children (the average IQ was 135.1) the highest IQ and GAI attained were 159 and 176, respectively. Although the highest score one would expect in a sample of about 2,000 is about 150, I did not know that such scores were so rare. I know that the distribution among races is not necessarily normal (e.g., blacks have a pearson type IV distribution according to Jensen). Thank you for the information.