The Fifth Norming of the Mega Test

Ronald K. Hoeflin

          The fourth norming of the Mega Test was based on scores reported by Mega Test participants on five previously taken tests: the Army General Classification Test, the California Test of Mental Maturity, the Wechsler Adult Intelligence Scale, the Stanford-Binet, and the Langdon Adult Intelligence Test. Three commonly reported test results -- on the Scholastic Aptitude Test, the Graduate Record Exam, and the Miller Analogies Test -- were discarded on the grounds that their means and standard deviations with respect to the general population are quite uncertain.

          Recently, however, fresh information on the distribution of scores on the Scholastic Aptitude Test has been obtained by Keith Raniere, a member of the Hoeflin Research Group, from senior statisticians at the Educational Testing Service. Using this information, it is now possible to norm the Mega Test by comparison with a test for which an unusually large sample is now available. For whereas most conventional intelligence tests are normed using a sample of just 3,000 or so participants, the new SAT data consists of the performances of 964,739 individuals who took the test as high-school seniors in 1984. Moreover, this data shows combined verbal plus mathematical aptitude scores, whereas all previous data consisted of separated data for the verbal and mathematical aptitude portions of the SAT, which left the distribution of combined scores a matter of conjecture.

          There is still some uncertainty as to how the new data relates to the general population, since not all high school seniors attempt the SAT, but using the Educational Testing Service's conjecture that more then 95% of the most able one percent of high school seniors attempt the SAT each year, it is possible to arrive at some fairly plausible norms for the Mega Test, which reaches the 99th percentile at a fairly low raw score. In fact, the results are almost identical to those arrived at in the fourth norming despite the strikingly different sources of these two normings.

          The present norming is presented in the form of seven charts, whose contents can be summarized as follows:

          Chart A: This chart shows the distribution of scores on the SAT for 964,739 high-school seniors in 1984.

          Charts B, C, and D: These charts show (A) the distribution of SAT scores reported by Mega Test participants, (B) the distribution of Mega Test raw scores of these same participants, and (C) a smoothed distribution of Mega Test raw scores for these participants.

          Chart E: This chart shows the scores on the SAT and on the Mega Test that are equivalent to various percentiles vis-a-vis the general population based on comparisons of Charts A, B, C, and D as well as estimates (reported orally to Keith Raniere by Educational Testing Service officials) concerning what percentage of high-school seniors at various ability levels attempt the SAT and what percentage of SAT data is from foreign students.

          Chart F: This is a graph of the results from Chart E.

          Chart G: This chart gives IQ's and percentiles for each raw score on the Mega Test as read from Chart F.

Reprinted by permission of the Educational Testing Service, the copyright holder.

[ the rest of this norming is still under construction ]

Return to the Uncommonly Difficult I.Q. Tests page.