The Vaccine Carrier Retest: Rubric Weighting and an Invalid Probe Reading
An 11th-grade engineering design class is prototyping insulated carriers for a scenario in which a rural clinic must move 12 vaccine vials at 2–8 °C for six hours. Each team receives one 250 mL frozen gel pack, 1 cm foam, cardboard, reflective bubble wrap, tape, and a digital thermometer; the prototype budget is $8 per team. On the first test, three of six carriers warm above 8 °C within 90 minutes. One team records 11 °C at every reading, including time zero, because the probe was taped to the outside of the box. Another team's carrier was never sealed during the run, though it was sealed in the practice build. One carrier holds 4–6 °C for the full six hours. Two days before the unit ends, the team whose carrier failed asks to abandon its concept and start over. The teacher has one class period available for retesting and a rubric that awards most points for the final temperature.
1. Should the teacher permit the restart, and how should the rubric be reweighted? Justify both decisions.
Read the worked answer
On the restart: allow a bounded revision, not a fresh concept. The team may change one variable - seal method, insulation thickness, or gel-pack placement - and must justify the choice with its own test data. Rationale: iteration is the core of design, so refusing revision teaches the wrong lesson; but an unlimited restart two days before the end erases the real constraint of time and discards the data the team already paid for. One period supports a single-variable retest at parallel stations with timers and a shared recording sheet.
On the rubric: the current weighting rewards a lucky result. Rebalance toward process evidence: criteria and constraints, a test log with time-stamped readings, failure analysis, a justified revision, and final performance against the 2–8 °C threshold for six hours. Award credit for a well-reasoned decision not to change a design when the data support it - otherwise students learn to tinker for its own sake.
On the successful team: do not let success end the work. Ask them to explain why the carrier held temperature and to test a second variable, so the result is attributed to design rather than luck. Their evidence earns credit, but the explanation is what is assessed.
2. The team whose probe was taped outside the box recorded 11 °C at every reading. How should the teacher treat that run and use it instructionally?
Read the worked answer
Treat the 11 °C flatline as an instrument problem, not a design result. A reading that begins at 11 °C at time zero cannot describe a carrier packed with a frozen gel pack; it describes a probe taped to the outside of the box, measuring room air. The data are precise and useless - a clean teachable case of valid measurement versus repeated measurement.
Classroom move (20 minutes): have teams place probes in an ice-water bath and in room air, compare readings, and write a one-sentence rule for probe placement. Then have the affected team rerun with the probe inside a water-filled vial standing in for vaccine, and record both the air and vial temperatures. Require every team's method section to state probe location.
Grading: do not count the invalid run as performance evidence, but do count the team's diagnosis and rerun as evidence of analysis. The other two failures need the same treatment - one unsealed carrier is a procedural deviation, not a design flaw, and should be reported separately.
Broader point: the discrepancy is information. Students should learn to ask whether a surprising number means the design failed or the measurement did, and to design tests with redundancy - two probes, one in the payload and one in the air.
3. Propose how the unit should be restructured for next year, tying each change to a specific failure observed in this cohort.
Read the worked answer
Structural changes, each tied to a failure observed this year:
- Separate the run from the concept. Give teams a mandatory shake-down test in week one with a non-graded carrier, so seal and packing errors surface before the graded run - this addresses the unsealed carrier.
- Standardize instrumentation. Issue two probes per team, require calibration in ice water, and specify probe placement in the protocol - this addresses the outside-the-box probe. Build a shared data table so all teams record at 0, 30, 60, 120, 240, and 360 minutes.
- Publish the threshold in advance. State the success criterion (all vials 2–8 °C for six hours) at the start and show the rubric, so teams design to the specification rather than to the ranking.
- Schedule iteration. Put a revision cycle and retest day on the calendar instead of improvising one at the end.
- Keep the scenario honest. The clinic constraint - 12 vials, six hours, $8 - is realistic and should stay; the failure was in test design, not the premise.
Assessment implication: the unit should be judged on whether teams can explain the gap between predicted and measured performance, which is the same reasoning used in structures and thermal units.