Gifted Issues Discussion homepage Forums Identification, Testing & Assessment Deciphering Talent Search Percentile

I was looking at talent search percentiles for out of grade level testing (Explore ACT) and trying to apply to overall population. For example if the 4th grade percentile indicates 80% what would this in the normal class?

Are the talent search kids are in the top 75%, 50% or top 25% of the population? This would translate the percentile to the overall class very differently

It depends but I think that Duke (explore) targets the 95th percentile so 80th percentile of that would be the top fifth of the top 5%, i.e, the 99th percentile of the general population

I am not 100% certain of that 95%'tile bit, however, as I do not entirely recall where or when I saw it anymore.

I think I know what you are trying to do, but I don't think that you can get reliable numbers that way. For one thing, the different talent searches appear to have different threshold criteria. In other words, the population demographics as far as ability are not exactly the same. For example, I glance at that Duke chart another poster recently linked and was surprised that the highest Explore scores weren't as high as I would have expected based on stats from some other talent searches.

We have used three different talent searches over as many years and it may also be that the stats also shift over time. The criteria to participate may also change. I think at least one of the talent searches that we used (JHU CTY?) allowed entry based on state assessment in the advance category. In 3rd and 4th grade math, that would have included almost 2/3 of all the kids in those grades at DS/DD's school. For our first talent search in 2nd grade for my younger two kids, we did not have any of the listed tests but was still able to register based on parent recommendation.

I think some of these talent searches supposedly aim for the top 5% in reading or math, but in reality the net ends up being broader with state assessments and parent recommendations as just two possible explanations. I would not assume that the talent search kids are higher than top 10% although I would assume that they are mostly at least top 25%. So in your example with the 80 percentile, that would translate to top 2% to top 5% in a normal class.

I am going to answer your question, but first I am going to say that the school did not care about the results at all. They did, however, want a copy of the personalized sheet (based on results) that said what my child should be working on now and what needed improvement, and the projected path through subjects (how many years ahead in math and so on). These things really impressed them.

The way it was explained to me is that the test was for kids already identified as gifted (top 2%) OR top 5% as recommended by a teacher. YMMV, I am sure.

Random factoid in response to your question: DS10 just took the SCAT for grade 5, compared to grade 8 norms. You have to be "in the 95th percentile" to write, but there's a long list of ways to qualify, including both achievement and cognitive testing. (It's a low bar and pretty easy to qualify, unlike, say, Davidson).

We received percentiles for math and verbal comparing him to both CTY's grade 5 "talent search group" and national grade 8 norms. His verbal score was middling, but 12 percentile lower compared to CTY grade 5s than national grade 8s. His math was high, but still 5 percentile less compared to the CTY grade 5s than the older kids.

You obviously can't extrapolate much from this, but I wouldn't be surprised if a typical group of out-of-level test takers could consistently outperform national norms set for several years older. DYS would be a whole different level. I also suspect this effect may be exacerbated when they are younger and aren't learning a great deal of new math from one year to the next.