But it can also be seen as a process theory
waiting to be tested. In this case it is a theory
about demographics, communication, conflict,
and turnover, but many kinds of organizational
research have a similar logical structure. Antecedents
and consequences are measured, correlations
are computed, and results are reported.
Yet, as Abbott (1990) points out, a typical data set
contains no information on the causal chain of
events that explains why the variables are related.
Sutton and Staw (1995) have argued quite
forcefully that it is the responsibility of the researcher
to fill in the missing events, because
without explanation there is no theory. This conclusion
is echoed by Lawrence (1997), who suggests
that although examining the intervening
processes may be expensive or time consuming,
it is necessary if one wants to build explanatory
theory. A purely instrumental theory, which posits
a connection between variables but lacks
explanation, generally is not seen as a theory at
all (Lawrence, 1997).