Section Sixpretest cannot be used because it may sensitize employees to a
subtle objective of the training and therefore defeat its
purpose.^58 An example might be where “asking questions of the
members of the control group regarding their management
style might cause them to become sensitized to this aspect of
their behavior and, inadvertently, cause them to change their
style in some systematic way.”^59 Further, the statistical testing
with this design is straightforward, as the significance of the
difference in the means between the experimental and control
groups is determined with a t-test. Unfortunately, the
researcher may have nagging doubts as to whether the
experimental and control groups are truly equivalent.^60 This is
an important point because decision makers, who must be
convinced of the value of the training, may not truly
understand randomization and research design.
Although these two experimental designs should be
adequate for most evaluation needs, the Solomon four-group
design can eliminate other threats to the validity of
evaluation.^61 However, the exceptions that require this level of
analysis are probably beyond the scope of interest for most
general managers.