Chart III
(page 25) shows the actual distribution of grade-point averages for
the 1952 Scholars and Comparisons. This group was chosen for
illustrative purposes because it is the largest for which
comparable data are available (414 Scholars and 431 Comparison
students at 11 colleges in the freshman year, and 277 Scholars and
309 Comparison students at 11 colleges in the senior year).
As the chart
indicates, a substantially larger proportion of Scholars than
Comparisons ranked in the top fifth of their class in all four
years of college, while the situation at the bottom end of the
scale was mixed. In the freshman and junior years, a slightly lower
proportion of Scholars than Comparisons ranked in the bottom fifth
of the class, but in the sophomore and senior years the situation
was reversed.
Scholars with
11 years of previous schooling tended to do slightly better than
those with only ten, but the latter tended to do slightly better
than those with 12. Among all four groups of Scholars, those with
only ten years of previous schooling tended to rank in the top
fifth of their class with greater frequency than the Comparison
students. (See Appendix Table V, B.)
AREA
TESTS OF THE GRADUATE RECORD EXAMINATIONS
Grade-point
averages are a reasonably reliable yardstick for comparing the
academic performance of individual students or groups of students
within a college or university, but they are not very
reliable in measuring the comparative performance of students in
several institutions, because each institution may be using
a different yardstick.
In the spring
of 1954, however, the Educational Testing Service of Princeton, New
Jersey, made available a new battery of tests that provided a much
broader basis for measuring the comparative performance of the
Scholars and Comparison students at the 12 participating colleges
and universities. These new tests were the Area Tests of the
Graduate Record Examinations,