Three other limitations of this study lay outside
our taxonomy. First, our coding was limited to
every third volume of AMJ. It remains an empirical
question whether the trends observed in our data
would hold with all volumes coded. Second, it may
be that the trends observed in our data would have
differed if other top management journals had been
coded. As noted previously, journals develop their
own particular cultures, which may alter the levels
(and impact) of theory testing and building over
time. Third, we utilized citation rates as a means of
capturing the impact of empirical articles. A key
limitation of citation counts is that they weigh each
citation equally, regardless of the importance of the
cited article to the citing manuscript (Kacmar &
Whitfield, 2000). Citation counts are also driven by
a number of factors that were not captured in our
study, including specific methodological and article
characteristics (Judge et al., 2007).
Conclusion
In their discussion of “what theory is not