There has not been, to my knowledge, any research aimed at answering this question. However, Harold A. Edgerton's 1973 study of Westinghouse Science Talent Search participants (Identifying High School Seniors Talented in Science, Washington, D.C., Science Service) is relevant. The study looked at that year's 300 recipients of honorable mentions, who had been selected on the basis of the quality of their research papers. The names of these 300 young people were checked against a list of 300 names selected from the total pool of Westinghouse entries on the basis of their high academic achievement (grades, SAT scores, and class rank).
The study compared the two groups to see how many students were in both. There was only one-third overlap. Two-thirds of the students chosen on the basis of scientific research performance did not have high enough scores to be in the top 300 if academic achievement had been the ...
