Hi CM,
I'm not really sure what you're looking for. I skimmed through the paper (read more closely in parts), and my impression is that her study is flawed at best. Her writing is garbled in places, too, and I felt that the introduction rambled and failed to make her purpose clear.
This paper had a lot of flaws that have pointed out in this thread. I still find it very hard to believe that she could have had so many children at or above the 99th percentile, especially that so many could have had IQs of at least 170. I don't get the control group structure at all (esp. the second "control" group). It seems messy.
I don't know much about the Stanford-Binet test, but
this article says that the ceiling of the revision that was released in 1986 (they administered it in the early 90s, right?) had a ceiling of 148. So I'm confused about her assertion that the ceiling of the test was 170. Other versions seem to have ceilings that are much higher than 170. She should have made this clear.
In fact, in looking through the paper, I saw that it's heavy on generalizations and has no quantitative data, apart from the dubious reporting of IQs ("Most of them now say....", "Too many had dissipated....", "Others, though, felt that....). I didn't find a single table in the entire paper. She didn't provide a copy of her survey questionnaire (if one even existed) or her basis for making judgments about "success." There was no statistical analysis, no reporting of percentages of subject's answers to questions, no nothing. Just
lots-of-people-felt-this-way stuff. The entire thing seems to be subjective.
The study seems to be something of an edumacation project to me. I've seen these before (as a reviewer and reader): they're characterized by failure to use rigorous or even semi-rigorous methods, failure to structure the study properly, failure to provide data, and an apparent willingness to cherry-pick information that supports an original hypothesis, rather than allowing data to drive conclusions.
HTH....
Val
