tistical hypothesis to be tested.
The sample for the study reported in this article consisted of 48 "female subjects diagnosed with early stage breast cancer" (Pierce, 1993, p. 23). The subjects were selected "from a major medical center and community physician referrals" (Pierce, 1993, p. 23). According to the article, a "theoretical sampling procedure" was employed: only patients with early stage disease were included in the sample "because they were suitable candidates for a full range of treatment modalities" (Pierce, 1993, p. 23). Patients were asked to participate in the study subsequent to diagnosis and counseling by their physicians on treatment alternatives. The 48 women who agreed to participate in the study then constituted the research sample for the study. The age range in the sample was from 25 to 90 years old, with a mean age of 49.6 years old.
A "qualitative method derived from grounded theory . . . was used to study the clinical problem of decision making" in the study reported in the article (Pierce, 1993, p. 22). This method involved asking the subjects to "think aloud" as they reached their decisions, while the researchers recorded every word said by the subjects to be analyzed at a later time (Pierce, 1993, p. 22). No instrumentation was employed in the collection of data for the study, and the interviews included no structured questions.
The research design provided for the sorting and coding of the narrative material collected from the subjects "according to the constant comparative method" (Pierce, 1993, p. 23). Line by line analysis "was used to look for indicators to describe the decision making experience of the subjects" (Pierce, 1993, p. 23). Sorting was performed several times "to find the best fit until a category could be name and described with confidence" (Pierce, 1993, p. 23). Quantitative procedures were not used in the analysis of the data.
The study found that five "empirical ...