CRITERIA FOR NATURALISTIC CASE STUDIES B3

As I review manuscripts or read student project papers, I try to consider each as a totality and as an individuality. Only secondarily do I try to look at them analytically, satisfying this or that criterion. It is the totality that gets priority. What kind of creature is it? I try to realize the special uniqyeness of each paper, realizing how it might be different if done at a different time or by a different researcher. At least at first I resist the idea that the quality of the case study is some aggregate of scale values. I look for the quality of the whole.

But as I look back over my reviewer remarks, it is apparent I am analytic too. I do have criteria. I note certain "shortcomings" occurring over and over. The derivative criteria are listed below. Any one of them might be unimportant for a particular case study. But again, any one might almost completely determine a study's lack of quality.

Conceptual Structure. When I have finished reading, I want to feel that I know the structure of ideas, the theoretical framework or emic issues, around which the report was organized. Mere chronological or institutional or goal structure usually is not sufficient. Still, I do not want to feel that the conceptual structure is so strong that it has kept key observations from being made.

Editorial Organization. I want to feel that each paragraph has been reread critically, finally assembled in a way that contributes to the whole. The study should have a sense of closure, probably ending on the ideas of its opening; with good format.

Author's Viewpoint. I prefer to be told, one way or another, the author's experience and feelings. If these do not nicely fit into the report, then footnotes, a preface or even a cover letter can be used. I often am dismayed, however, with extensive subjective ruminations or with an author's frequent references to self.

Sense of Audience. I like to feel that I know to whom the author is writing.

Over interpretation. Naturalistic case studies are interpretive research. Still, I want raw observations available for the reader's interpreting. To say "the father was indignant" or "it was an efficient office" is not an observation, it is an interpretation. Facts should come first. Over interpretation is one of the most common and distressing failings of case study writing.

Failure to Interpret. After the observations are presented, perhaps at the very close of the report, the author should give additional interpretation of things according to his or her special purview or competence. For example, an educator should acknowledge relevant educational writings. (Passing judgment or indicating causality are seldom necessary interpretations.)

Validation. I want the author to let me know, directly or indirectly, how the observations were validated, and even what efforts were made to disconfirm principal conclusions, to tell what deliberate efforts there were to triangulate.

Multiple Realities. I hope the study acknowledges alternative ways of seeing.

Proportion. I want to find a sense of proportion, that the author is in touch with how people value things. To do this requires a good balance among descriptions of ordinary events, issues, contexts, and assertions. The trivial can be reported but not idolized. The momentous should not be ignored.

Stake, 2/86