764 THE STRUCTURE OF EVOLUTIONARY THEORY
more common than fraud)—for scientists affected by publication bias do not
recognize their errors (and their bias may be widely shared among colleagues),
while a perpetrator of fraud operates with conscious intent, and the wrath of a
colleague will be tremendous upon any discovery.
Begg and Berlin (1988) cite several documented cases of publication bias. We
can hardly doubt, for example, that a correlation exists between socioeconomic
status and academic achievement, but the strength and nature of this association
can provide important information, for both political practice and social theory.
White (1982, cited in Begg and Berlin) found a progressively increasing intensity
of correlation with prestige and permanence of published source. Studies published
in books reported an average correlation coefficient of 0.51 between academic
achievement and socioeconomic status; articles in journals gave an average of
0.34, while unpublished studies yielded a value of 0.24. Similarly, Coursol and
Wagner (1986, cited in Begg and Berlin) found publication bias both in the
decision to submit an article at all, and in the probability for acceptance. In a
survey of outcomes in psychotherapy, they noted that 82 percent of studies with
positive outcomes led to submission of papers to a journal, while only 43 percent
of negative outcomes provoked an attempt at publication. Of papers submitted, 80
percent that report positive outcomes were accepted for publication, but the figure
fell to 50 percent for papers claiming negative results.
In my favorite study of publication bias, Fausto-Sterling (1985) tabulated
claims in the literature for consistent differences in cognitive and emotional styles
between men and women. She does not deny that genuine differences often exist,
and in the direction conventionally reported. But she then, so to speak, surveys her
colleagues' file drawers for studies not published, or for negative results published
and then ignored, and often finds that a great majority report either a smaller and
insignificant disparity between sexes, or no differences at all. When she collated all
studies, rather than only those published, the much-vaunted differences often
dissolved into statistical insignificance or triviality.
For example, a recent favorite theme of pop psychology attributed different
cognitive styles in men and women to the less lateralized brains of women. Some
studies have indeed reported a small effect of greater male lateralization; none has
found more lateralized brains in women. But most experiments, as Fausto-Sterling
shows, detected no measurable differences in lateralization at all and this dominant
relative frequency (even in published literature) should be prominently reported in
the press and in popular books, but tends to be ignored as "no story."
Paleontology's primary example of publication bias—the nonreporting of
stasis under the false belief that such stability represents "no data" for evolution—
illustrates a particularly potent form of the general phenomenon, a category that I
have called "Cordelia's dilemma" (Gould, 1995) to memorialize the plight of King
Lear's honest but rejected daughter. When asked by Lear for a fulsome protestation
of love in order to secure her inheritance, Cordelia, disgusted by the false and
exaggerated speeches of her sisters Goneril and