| The science education community has long tacitly assumed that astronomy, in addition to other fields in the earth and space sciences, is a conceptual domain crucially dependent upon three-dimensional representations and thinking, and that learners need to utilize complex spatial thinking to develop deep understanding of the field. However, a review of the literature indicates that the astronomy education research community has yet to empirically establish the character of the relationship between students' spatial reasoning ability and their ability to learn astronomy content in college science classes in general, or the manner in which this relationship varies across the astronomy content domain. In order to determine the relationship between students' spatial reasoning skills and their ability to learn astronomy content, undergraduate students in a non-major introductory astronomy survey class at a medium-sized, research-extensive mid-western university were given astronomy knowledge diagnostics at the beginning and end of a semester-long introductory astronomy survey course, as well as spatial reasoning diagnostics in the middle of the semester (N = 86, 49 males, 37 females). All data were matched. Instruments used were the Test of Astronomy STandards (TOAST), a test of conceptual astronomy content knowledge, and What Do You Know? (WDYK), a test of students' understanding of astronomical geography, events related to Earth's rotation, and phenomena related to Earth's orbit and tilt. Spatial reasoning ability was measured with a three-part instrument designed to determine students' ability to perform mental rotations, transformations, and spatial self-efficacy. In addition, at the end of the semester 14 students were interviewed to explore and validate assumed thinking processes used during the assessments. Pearson r correlations were performed in order to explore potential relationships among incoming prior knowledge, knowledge after instruction, knowledge gains and students' spatial reasoning related to mental rotations, transformations, and self-efficacy. After receiving traditional lecture-based instruction, it was found that students' normalized gains for astronomy content knowledge were quite low (TOAST, = 0.26; WDYK, = 0.13). As a result of the small variation in gain score, the correlations between the gains and the spatial assessments turned out to be quite small. In contrast, the Pearson r correlations between astronomy pre- and post-course scores for the TOAST and WDYK tests, and two of the three components of the spatial assessment instrument show moderate to strong relationships. We observed strong and statistically significant relationships between the TOAST pre-test and spatial rotation score (r(84) = .40, p <. 01), the TOAST pre-test and spatial transformation score (r(84) = .36, p <. 01), the TOAST post-test and spatial rotation score (r(84) = .48, p < .01), the TOAST post-test and spatial transformation score (r(84) = .37, p < .01), the WDYK pre-course and spatial rotation score (r(84) = .31, p < .01), the WDYK pre- course and spatial transformation score (r(84) = .19, p < .05), the WDYK post- course and spatial rotation score (r(84) = .43, p < .01), and the WDYK post- course and spatial transformation score (r(84) = .36, p < .01). Taken together, spatial reasoning appears to be moderately to strongly related to astronomy knowledge, and rotation skills appear to be more dominant than transformation skills for astronomy. The correlation between astronomy post-course scores and spatial scores suggests that the relationship between spatial reasoning and astronomy ability explains about 25% of the variation in the data. We further disaggregated our data by gender and found moderate gender differences in some spatial reasoning scores, however, not always in the same direction. This data invites further investigation in a number of aspects, particularly a comparison with courses showing higher gains, and a more in-depth investigation of the observed gender differences. Our findings result in recommendations for astronomy instructors to recognize the need to support student learning by employing instructional techniques that support and scaffold student thinking related to astronomy concepts demonstrated to rely heavily on spatial reasoning. This work was supported, in part, by the Wyoming Excellence in Higher Education Endowment. |
