The current study compares analyses of data collected using both constructed-response and interview formats across a common pool of subjects. At its core, the current study explores the degree to which the data sets collected through the constructed-response and interview formats support similar inferences when coded with the rubrics from a recent series of studies on knowledge structure coherence.