Section Six
A final research design appropriate to the assessment of
training programs is the quasi-experimental design based on a
time series. This design may be appropriate in situations in
which no control group can be obtained. The essence of the
approach is to obtain measures of the evaluation criterion at
repeated intervals, such as every month for six months, prior to
the conduct of training. After the training, the measures would
be collected again at repeated intervals. Unfortunately, analysis
of the significance of the results is not as straightforward as
with the two experimental designs because of the
nonindependence of observations.^62 In some cases, the
interpretation of the results may rest on a simple visual
examination of observation trend lines.
The CIGNA Corporation has used the quasi-experimental
design to assess the effectiveness of its basic management
skills training program. In one unit, repeated measures of
productivity were collected for several months prior to and after
the training in order to assess the program’s effectiveness. In
another unit, measures of outstanding collectibles were
collected in the same manner. Additionally, projections of
collectibles were also developed in this unit to indicate the
performance that would have occurred in the absence of the
training. These projections allowed the calculation of financial
returns on the training investment. Other aspects of its
program provide an indication of the comprehensiveness of