A more systematic approach to assessing the enhancement of student learning about stellar distributions
(which I have not done yet) would be to have students draw what they think the model will look like
before they build it. They might include details about what type of stars would be found at various
positions within the model, as well as whether there would be equal or different numbers of stars of each
spectral class. At the end of the exercise, the students would be given back their original drawings and
asked to comment about whether their drawing of the model was consistent with what the actual model
looked like, what the significant differences were between their model and the real model, and whether
they were surprised by those differences.