This research compares the methodological tools employed in NOS research, with analysis of what the comparison implies about the structure of nature of science knowledge. Descriptions of practicing teachers’ nature of science conceptions were compared based on data collected from forced choice responses, responses to a qualitative survey, and course writing samples. Participants’ understandings were scored differently on the Views of Nature of Science Questionnaire (VNOS) than the forced-choice measure, Scientific Thinking and Internet Learning Technologies (STILT). In addition, analysis of the writing samples and observations combined with interviews portrayed more sophisticated, but more variable, understandings of the nature of science than was evidenced by either the survey or the forced-choice measure. The differences between data collection measures included the degree to which they drew upon context bound or context general reasoning, the degree to which they required students to move beyond the simple intelligibility of their responses and allowed students to explore the fruitfulness of the constructs, as well as the degree to which they revealed the interconnection of participants NOS conceptions. In light of the different portrayals of a participants NOS conceptions yielded by these different measures, we call for the use of crystallization as a methodological referent in research.
Settlage, John; Southerland, Sherry; Johnston, Adam; and Sowell, Scott, "Perhaps Triangulation Isn’t Enough: a Call for Crystallization as a Methodological Referent in NOS Research " (2005). TERC Documents. Paper 1.