In a study of reading clinicians' collection and diagnosis procedures, three types of clinical agreement were examined: group agreement, intraclinician agreement, and interclinician agreement. Eight senior and respected practicing clinicians were involved in the study, which used simulated cases of reading problems based upon real cases. The problems represented included sight word deficiencies, inadequate structural and phonetic analysis skills, inadequate fluency of oral reading, and poor comprehension. The same four-step procedure was used for each case run. First, the subjects were given instructions and practice with a sample case. Second, there was an observational session in which the subjects collected data. Third, the subjects prepared a written diagnostic and remedial report, and fourth, there was a "debriefing" session in which the subjects underwent stimulated recall. Each clinician received three simulated cases at one-week intervals. The first and third were equivalent forms of the same case. The results indicated that in diagnosing reading cases, reading clinicians do not agree very well with themselves or with other clinicians. (Appendixes contain a discussion of the inquiry theory and the research agenda.) (FL)