The teaching of lunar phases to pre-service elementary teachers is important since lunar phases and the relationship of the earth-moon-sun system are specifically stated in the national science education reform documents. Moreover, Benchmarks for Science Literacy (Benchmarks) (American Association for the Advancement of Science, 1993), National Science Education Content Standards (NSES) (National Research Council, 1996), as well as Next Generation Science Standards (NGSS) (National Research Council, 2013) give specific recommendations for making use of physical models to help students understand the geometry of the sun-earth-moon system that is critical for understanding phases of the moon. Each of these documents view concepts of lunar phases as foundational for understanding the sun-earth-moon system. Additionally, these national reform documents have been modified by many states to serve as a basis for their state science education standards.Pre-service elementary teachers' conceptions of shape, sequence, and scientific understanding of the cause of moon phases improve both statistically and substantively after inquiry-based, conceptual change instructional intervention based on a computer simulation. More specifically, planetarium software was found useful to facilitate a constructivist approach to science education where the students constructed their own understanding of lunar phases by the use of this inquiry-based, conceptual change approach. Gains in conceptual understanding were measured by a semi-structured interview protocol as well as the LPCI and an Abridged-LPCI that is a version of the LPCI that includes only the questions that align directly with the interview protocol.The Abridged-LPCI results correlate better than the LPCI with the semi-structured interview protocol. Although there were statistically significant correlations, they were not practically significant. Student interviews, as well as student daily written reflections, provided data that helped to illuminate the conceptual change process and are not available from forced-choice instruments.Additionally, there are no statistical or substantive differences in achievement between students who collect moon observation data in a whole class setting versus those who collect this data working with the computer simulation in pairs. In general, the factors measured from the writing of student reflections were not significantly correlated with measures of conceptual gains although they can provide insights into the learning process.And finally, the Styrofoam™ ball (scientific model) activity was found to be a significant instructional component that positively impacted improvement of scientific conceptions of moon phases. Only 2 of 12 possible correlations between student reflection and LPCI were statistically significantly positively correlated and none of the 12 possible correlations for the Abridged-LPCI were statistically significantly correlated with student reflections, although they were not practically significant.Moreover, ...