ABSTRACT
Chapters 11, 12, and 13 discussed and examined techniques applicable in general observational research settings. In each of these earlier chapters, constructing, using, and scoring some type of coding scheme or schedule of questions were seen as necessary features of the investigative process if the aim of the observational research was testing, rather than generating, hypotheses. Often, the psychometric quality of the coding systems used in an observational situations is not a major consideration. This lack of attention to the statistical or psychometric properties of the measuring device is attributable to the fact that the coding scheme used by the observational researcher is viewed as a “one-shot” instrument.1 Most classification systems are constructed to satisfy the needs of a specific investigative setting. Indeed, as we continually suggest throughout this book, observational studies should be designed with the research setting in mind. Thus, considerations of the specific sample of individuals under investigation, their limitations, the physical dimensions of the research context, the behaviors of interest, and so on should all be taken into account in constructing the coding system. This observation suggests that a system suitable for the study of a teenage motorcycle gang in Shaker Heights, Ohio, might not prove useful in studying the adjustment behavior of a group of first-year medical students in Pomona, California.
