scored, home runs, and walks. Power pitchers with rising fastballs benefited the most from
the change, whereas many control pitchers, who liked to “nibble” at the corners of the plate,
could not consistently get batters out. Among hitters, Barry Bonds of the San Francisco Gi-
ants, who no longer had to chase outside pitches, went on a record-setting home-run tear,
whereas Chuck Knoblauch of the New York Yankees, who had a tendency to “pop up” high
strikes, was mired in a season-long batting slump. The rule changes affected players in differ-
ent ways and had nothing to do with how hard they worked or their skills or their moral
qualities as individuals. People who had honed their games to the old rules, and were unable
to adjust, had their careers threatened.
There is a comparable phenomenon of assessments influencing results in gymnastics.
Have you ever wondered why the highest scoring competitors among the men are in their
middle to late 20s, whereas the best performers among the “women” are teenage girls? It is
not because of a law of nature, but because of the way performances are evaluated. In men’s
gymnastics, the highest scores go to routines that emphasize strength, so older, stronger,
and more experienced men score higher. In women’s gymnastics, the highest points are
awarded to routines that emphasize balance and flexibility, so younger, but not necessarily
more talented, women win the events. If women were graded using the same scale and rou-
tines as men are, there would be different medal winners.
Arbitrary rules and assessments play a similar role in educational evaluation. For exam-
ple, in September 2001, the incoming freshman class at the University of California differed
demographically in significant ways from the previous year (Holmes, 2001). Part of the
change was caused by the university’s decision to count scores on the verbal and math
knowledge parts of the Scholastic Aptitude Test less, while giving heavier weight to student
performance on subject-specific tests, including foreign language examinations. As a result
of the new criteria, native speakers of Chinese, Spanish, and Korean earned admission to the
university at a higher rate than in the past. A fair change? Maybe. An objective law of nature?
No way!
The point I am making is that assessment is nearly always arbitrary and often unreliable.
The question for us as teachers then becomes: “Can we design assessments that facilitate
the ability of students to demonstrate what they have actually learned?” I believe the answer
is a qualified “yes.”
An additional problem with assessment has been described by Claude Steele (1999), a
psychologist from Stanford University in California. Steele has conducted a number of exper-
iments in an effort to explain the “performance gap” (the difference between expected and
actual scores) of African Americans and women on certain types of standardized assess-
ments and sometimes on school performance in general.
Steele (1999) argues that all people are members of some group about which negative
stereotypes exist and that they experience “stereotype threat” as a result—“the threat of be-
ing viewed through the lens of a negative stereotype, or the fear of doing something that
would inadvertently confirm that stereotype” (p. 44). His research shows that as a result of
the additional stress caused by stereotypes that claim that African Americans are not as
smart as Whites or that women cannot do complex math, even top African American and
women students score significantly below expectations on specific tests.
Steele’s studies suggest that repeated poor performance on tests and failure in school
feeds into student resistance to learning. As they seek to avoid uncomfortable situations
that provoke stereotype threats, they refuse to do work, do not study for exams, and may
even disrupt classes. The most important and hopeful aspect of Steele’s work is that he
found that when teachers were able to reduce the stereotype threat felt by students, the
“performance gap” virtually disappeared.
202 CHAPTER 8