Empirical Accuracy and Consistency in College Students' Knowledge of Classical Astronomy
This study focuses on how college students explain basic astronomical phenomena and the nature of those explanations, moving beyond existing research of astronomy misconceptions and shedding new light on the nature of knowledge representation and the stability of knowledge. Fifty college students answered a questionnaire that asked for explanations of six classical astronomical phenomena (the solar system, day/night cycle, seasons, lunar phases, and solar and lunar eclipses) and were re-tested in an interview. The participants gave confidence ratings for their explanations, and tried to apply their explanations to several hypothetical scenarios. The participants also completed a general spatial ability test and a spatial ability test specifically designed for the classical astronomy domain. Explanations that were more scientifically accurate were the most consistent over time, were given higher confidence ratings, were better applied to the hypothetical scenarios, and were the most internally consistent with explanations for other phenomena. These explanations showed characteristics most like the theorized form of knowledge representation of a mental “theory.” In contrast, other non-scientific explanations were more likely to change over time, had lower confidence ratings, and were often internally inconsistent with other knowledge or were so primitive that they provided no conceptual connections to explanations for other related phenomena. Only one non-scientific explanation showed traits similar to more scientifically-accurate knowledge, a common explanation of the lunar phases based on Earth occluding light from the Sun. The nature of knowledge representation of novices, kinds of inconsistencies in knowledge, and the hypothetical relationship between the presence of inconsistencies and stages of learning are discussed.