Results of this approach, however, have been disappointing. The sorting out of
leaders with various leadership uaits from those without them has been notoriously
ineffective. One early study (Bird, 1940) extensively reviewed the relevant research
and compiled a list of traits that seemed to differentiate leaders from nonleaders
in one or more studies. However, only five percent of the traits listed appeared
in four or more studies; many of the other traits listed appeared in only a single
study. Mann (1959) reviewed 125 leadership studies searching for a relationship
between personality and performance in small groups. His search yielded 750
findings about personality traits, but no traits as conclusions. He found a lack of
consistency among traits described as significant for leaders and, further, found
that some traits listed as significant were diametrically opposed to significant traits
listed in other studies. Researchers continue to search for the behavioral scientists’
(if not the alchemisrs') gold, and with similar results. In studying discussion leaders,
Outer (1978) found no statistically significant relationships among traits of dis-
cussion leaders, student evaluations of them, and the grades received by students
of discussion leaders. His conclusion: “Attention to personality traits . . . would
have been of limited value in the selection of discussion leaders" (p. 697).