The Lunar Phase Item That Passed
A district common assessment includes a multiple-choice item: 'Why does the Moon show phases?' A: Earth's shadow falls on the Moon; B: the Moon rotates on its axis; C: the Sun-Earth-Moon angle changes as the Moon orbits; D: clouds cover the Moon. Across 420 students, 68% chose C, 22% chose A, and 10% chose B or D. The item is considered successful. However, in 30 think-aloud interviews, 18 students who chose C explained that Earth's shadow 'moves across the Moon and makes a crescent,' and only 5 could predict where the Moon would appear at sunset. In one class, students used a lamp and a ball, but the lamp was 30 cm from the ball and students stood at a fixed spot, so they saw mostly full and new shapes. The science coordinator must decide whether to keep the item, revise it, and what to change in instruction. Teachers have two 50-minute periods before the unit test.
1. Analyze the item and the interview data. What does the item actually measure, and why is 68% misleading?
Read the worked answer
The item measures verbal recognition of a phrase, not a working model of phases. Option C contains the correct words, but a student can select C while still believing that Earth's shadow sweeps across the Moon. The interviews show exactly that: 18 of the 30 students chose C but explained an eclipse-like shadow mechanism. Only 5 could make a spatial prediction at sunset.
The distractor analysis is also misleading. Option A is the classic eclipse misconception, and 22% chose it. That is a warning sign, not a success. The 68% appears acceptable only because the item does not require students to use the model. A high score on a recall item can hide a persistent misconception.
The item should be revised or replaced. A better item would present a diagram or a phenomenon-for example, 'At sunset, a student sees a half-lit Moon high in the south. Draw the Sun-Earth-Moon positions and explain the phase.' That requires a model, a prediction, and an explanation. The coordinator should not keep the item as evidence of mastery.
2. Critique the lamp-and-ball model and propose specific fixes for the two remaining class periods.
Read the worked answer
The model has several scale and geometry problems. The lamp is only 30 cm from the ball, so the light source is not effectively distant; a real sunlight source produces nearly parallel rays. The lamp has a finite size, creating a penumbra and blurring the terminator. The ball is too close to the observer for students to see the full geometry, and students stand at a fixed spot, so they see mostly full and new shapes rather than the full cycle. The model also may not show that the Moon's orbit is nearly in the same plane as Earth's orbit, which is why eclipses are rare.
Fixes:
- Use a dark room and a small bright lamp placed far enough away that the ball is clearly lit from one direction.
- Put the ball on a stick and have students move around it, holding a marker for Earth.
- Mark the Sun direction on the floor and have students predict the phase before observing.
- Use eight positions around the orbit and ask for the phase name, the fraction lit, and the Moon's rise/set time.
- Explicitly contrast phases with eclipses: in a phase, Earth's shadow is not on the Moon; in a lunar eclipse, the Moon enters Earth's shadow.
With two periods, the first should be model repair and prediction; the second should be a transfer task, such as predicting the Moon's position at a given time.
3. Design a revised assessment task with a rubric that measures the three dimensions.
Read the worked answer
Phenomenon task: 'On a clear evening, a student sees a crescent Moon low in the west just after sunset. Three days later, the Moon is higher and more than half lit. Use a diagram and written explanation to show the Sun-Earth-Moon geometry at both times. Predict where the Moon will be at sunset three days after the second observation.'
Rubric:
- DCI (phases): 3-explains that half the Moon is always sunlit and the visible phase depends on the Sun-Earth-Moon angle; 2-correct sequence but incomplete mechanism; 1-names phases without geometry; 0-uses Earth's shadow or clouds.
- SEP (developing and using models): 3-diagram shows Sun direction, Earth, Moon at correct orbital positions, and observer view; 2-diagram mostly correct with one geometric error; 1-labels but no consistent geometry; 0-no usable model.
- CCC (patterns and scale): 3-uses the pattern to predict a later position and comments on the limits of the diagram's scale; 2-predicts with partial reasoning; 1-describes a pattern without prediction; 0-no pattern.
The task requires explanation and transfer, not phrase recognition. It also gives the teacher evidence to distinguish a correct model from a memorized label.