Students' Conceptions of Research. I: A qualitative and quantitative analysis

Abstract

Relatively little is known about students' conceptions of research and, in particular, whether there are conceptually discrete patterns of variation that can be used to model this phenomenon in terms of research‐as‐learning outcomes. The present study explores the dimensionality of students' conceptions of research from two complementary research perspectives. The open‐ended written responses of students (n = 154) to questions aimed at soliciting variations in conceptions of what research is are analysed using a qualitative methodology to isolate “categories of description”. Findings are summarized in terms of eight such main categories, some of which are further internally differentiated. In terms of the main categories, research is conceived in terms of (variation in): (A) information gathering, (B) discovering the truth, (C) insightful exploration and discovery, (D) analytic and systematic enquiry, (E) incompleteness, (F) re‐examining existing knowledge, (G) problem‐based activity, and (H) a set of misconceptions. The substantive verbatim excerpts that formed the units of analysis in the qualitative analysis were used as a basis for item stems which were psychometrically operationalized into a Students' Conceptions of Research Inventory (SCoRI). This inventory was administered to a second heterogeneous sample of postgraduate students (n = 224) and the resultant data were subjected to exploratory factor analyses that provided empirical support (as dimensions of variation) for a smaller subset of the categories isolated in the qualitative analysis. Empirically, and in terms of additional psychometric considerations, there was support for five dimensions of variation (common factors) in terms of categories B, C, F, G, and H. These findings provide an initial conceptual basis for interpreting how students engaged in research activity may differ from one another in terms of their conceptions, as well as what the likely consequences of any such stable differences may be for research‐as‐learning outcomes.