5.8.1 Tacit knowledge and experience

In most of the studies reviewed above, tacit knowledge was found to relate to experience, indicated either by group membership (expert versus novice), or the number of years in one's current position.

In several studies, Sternberg and his colleagues showed that individuals with less experience in a given domain exhibit lower tacit-knowledge scores (Wagner, 1987; Wagner and Sternberg, 1985; Sternberg et al., 1993). In Wagner and Sternberg (1985), for example, group differences were obtained between business managers, business graduate students, and undergraduates on 39 of the response-item ratings on a tacit-knowledge test for managers, with a binomial test of the probability of finding this many significant differences by chance yielding p < .001. Comparable results were obtained with Yale undergraduates, psychology graduate students, and psychology faculty on a tacit-knowledge test for academic psychologists. In addition, Wagner (1987) found that business managers obtained the highest tacit knowledge scores followed by business graduate students, and undergraduates, with comparable results obtained in a study with psychology professors, psychology graduate students, and undergraduates. Wagner et al. (1994) also found that scores on a tacit-knowledge test for salespeople correlated significantly with number of years of sales experience.

Williams and Sternberg (cited in Sternberg et al., 1995), however, did not find significant correlations between several experience-based measures, including age, years of management experience, and years in current position, and tacit-knowledge scores. But they did find that the importance of specific pieces of tacit knowledge varied across organizational level. Their findings suggest that it may not simply be the amount of experience but what a manager learns from experience that matters to success.

5.8.2 Tacit knowledge and general cognition

In proposing a new approach to measuring cognition, it is important to show that one has not accidentally reinvented the concept of "g," or so-called general skill, as measured by traditional cognition tests. We do not dispute the relevance of general cognitive skill to performance. Schmidt and Hunter (1998) have shown that g predicts performance in a number of domains. Our aim is to show that tacit-knowledge tests measure something in addition to g. In all the above studies in which participants were given a traditional measure of cognitive skill, tacit-knowledge test scores correlated insignificantly with g.

The most consistently used measure of g in the above studies was the Verbal Reasoning subtest of the DAT. The absolute values of the correlations between tacit knowledge and verbal reasoning ranged from .04 and .16 with undergraduate samples (Wagner, 1987; Wagner and Sternberg, 1985) and .14 with a sample of business executives (Wagner and Sternberg, 1990).

One potential limitation of these findings is that they were obtained with restricted samples (e.g., Yale undergraduates, business managers). However, similar support for the relationship between tacit knowledge and g was found in a more general sample of Air Force recruits studied by Eddy (1988). The correlations between scores on the TKIM and ASVAB scales were modest, and none of the four ASVAB factors correlated significantly with the tacit-knowledge factor.