Performing Subjectivist Studies in the Qualitative Traditions Responsive to Users - Evaluation Methods in Biomedical Informatics - page 296

grouped together, but the larger theme became clear in what he calls a
“thunderbolt” moment. 4 The central organizing theme describing the lives
of residents, he realized, was “surviving.” He then revisited his data and
became more and more confident that this was the central issue. His sub-
jects did not talk directly about survival, so it was not until he reached the
highest level of interpretation that in the investigator's view, full under-
standing emerged. The insight emerged directly from the data and is an
excellent example of the outcome of a grounded theory approach.
Another approach, one that grew out of the organizational development
movement, is participatory action research (PAR). It is receiving increas-
ing attention in medicine. A recent Institute of Medicine workshop
explored the role of the public in the design, review, and setting of the
research agenda for clinical research. The workshop built upon Green's def-
inition of PAR, reported as “systematic inquiry, with the collaboration of
those affected by the issue being studied, for purposes of education and
taking action or effecting change.” 5 Participatory action research , participa-
tory inquiry , and action science are all terms that refer to the cooperative
efforts of researchers and those who work inside an organization to
improve the situation. Qualitative evaluation studies that involve the stake-
holders in the design, execution, analysis, and determination of what will be
done with the results fit this definition.
Another approach is to conduct the evaluation as a case study. Case
studies are detailed descriptions of real situations in which the context
and decision-making processes are explored. Case studies are an ideal
way of summarizing the results of an evaluation study by combining
what has been learned using different methods. They can include one or
more sites and can be either quantitative or qualitative, but are generally a
blending of both. They can be used for evaluation purposes or as teaching
aids, but would be written differently for those different purposes. Evalua-
tion or research case studies must be accurate and rigorous; teaching cases
need not be, as long as they stimulate dialogue. We will limit our discussion
to evaluation cases.
Case studies can be descriptive or explanatory and, like the qualitative
methods described above, they can help to answer research or evaluation
questions that ask why or how something happened. Also as outlined above,
there must be a clear strategy for selecting sites and informants and spe-
cific foci or measures appropriate for the research questions. Cases can be
either representative of the norm or unique. Evidence for the study can
come from any of the qualitative methods described and from any docu-
ments, including those presenting facts and figures from quantitative data.
The goal is to gather enough information in different ways so that the mul-
tiple sources of evidence help the investigator paint an accurate picture.
Analysis consists of gaining a broad view of the data as well as finding pat-
terns and themes. Again, it is recommended that multiple researchers