By: Dr. Roy Freedle
Abstract by: Allen Chein
Standardized multiple-choice tests, such as the Law School
Admissions Test (the “LSAT”), systematically underestimate verbal
intelligence of minority individuals. As a result, the validity and
also the reliability of the scores used to assess individual test
performance are questionable. Studies on the verbal SAT have shown
African-American students will actually score better than expected
on the hardest items and worse than expected on the easiest items,
a phenomenon that cannot be explained by “lucky guessing.” This can
be attributed to factors such as different uses of semantic and
syntactic cues between the races. In addition to bias based on
total scores achieved by each racial group, an examination of each
test item also reveals bias. Jay Rosner explained that when a new
test form is constructed by selecting test items from a large
potential test items pool, test assemblers routinely select items
that seem to favor Whites over African-Americans.
The validity of the LSAT as a predictive tool for student
performance as well as the dominant ranking criteria for law
schools is also uncertain. Some studies show the LSAT correlates
more strongly with grades for the first law school year and then
systematically diminishes over each of the next two years. As to
school rankings, how can the quality of education offered by the
faculty of an educational institution be primarily dependent upon
the test scores of an incoming student body? Better methods for
ranking would clearly separate a law school faculty’s status
ranking from any ranking associated with student LSAT scores.
Another superior option would rank schools by region and areas of
specialization.
The Fagan-Holland test should be adopted to achieve a truly
valid standardized test of verbal intelligence. No evidence for
ethnic differences in intelligence occurs using their procedures.
Fagan-Holland teaches that if all individuals are given an equal
opportunity to learn crucial standardized test concepts, there
would be no significant difference in how the races respond to that
test. Adopting this test in lieu of the LSAT will have a
complementary benefit. The race neutral Fagan-Holland model will
remedy biased LSAT test scores, which are contributing directly to
decreasing minority enrollment in many law schools.