Revisit your Case Study system with the course's complete ten-mechanism toolkit, separating what you can classify from how strongly your evidence actually supports it.
- Waiting Field Notes15% · due Week 4
- Queue Case Study20% · due Week 9
- Waiting Experiment25% · due Week 10
- Waiting Audit40% · due Week 12
Due: 10 May 2027 · Weight: 40%
The brief#
Revisit the system you analysed in the Queue Case Study, and evaluate it against the course’s complete ten-mechanism taxonomy — separating what you can classify as present in the system from how strongly your evidence actually supports any claim about it.
This is not a second case study. You are not starting over on a new system, and you are not re-running the five-mechanism analysis from scratch. You’re extending the same analysis to the five mechanisms your Case Study didn’t cover, and re-examining everything — old and new — through a stricter question than “is this present”: how strong is the evidence for what I’m claiming about it?
The same system, twice#
Use the system you analysed for the Queue Case Study. A different system is permitted only where the original has become genuinely inaccessible, unsafe, or no longer observable — and if you make that change, say so briefly and explain why. Carry your Case Study’s five findings (interface, environment, queue structure, priority, value) forward into this assessment’s structure rather than re-deriving them; your work here is to extend the analysis to time, attention, expectation, information, and control, and to apply the evidence-tier layer below across all ten.
Any continued or new observation of that system stays bound by the same Experiment ethics policy that governed the Case Study and the Waiting Experiment.
Two questions, kept apart#
Every mechanism gets answered twice, and the two answers are not the same question:
- Mechanism classification — present, absent, ambiguous, or cannot determine. This asks: what appears to be in the system?
- Evidence tier — directly evidenced, plausible interpretation, or undetermined. This asks: how strongly can what I have actually support this claim?
Directly evidenced does not mean caused. A feature or result can be directly evidenced without establishing its causal effect on experience. A countdown timer’s presence on a checkout screen is directly evidenced — you can see it, document it, quote it. Whether that timer actually changes anyone’s judged wait is, without a controlled test, at most a plausible interpretation. A mechanism classified “present” can sit alongside a claim tiered no higher than “plausible interpretation” — that’s a common, correct combination here, not a weak submission.
Where the experiment fits#
State plainly whether your Week 10 Waiting Experiment is relevant to the system you’re auditing. If it is: name the exact mechanism it tested and the conditions it tested under, and tier that specific claim according to how strong your own experimental design actually was — a thin, mixed, or uncontrolled result doesn’t automatically earn “directly evidenced” just because you ran an experiment. If it isn’t relevant, say so directly rather than forcing a connection. Either way, your experiment’s result speaks only to the one mechanism it tested — it must never be stretched to support a claim about any of the other nine.
What you submit#
The ten-mechanism classification, with the five from your Case Study carried forward and re-tiered rather than redone; an evidence tier for every claim you make about a mechanism; the experiment cross-reference above; a whole-system synthesis pulling the ten mechanisms together; and a limitations section addressing the scope of a ten-mechanism, single-system audit — what you could and couldn’t observe, the time window, the impossibility of inferring anyone’s psychology from what you saw, and the absence of a controlled comparison beyond whatever your one experiment actually tested.
What you must not claim#
No redesign, no recommendation, no fairness or efficiency verdict. No claim about what anyone felt, satisfaction, willingness to pay, or why the system’s owner built it that way, without evidence for it. “Undetermined” is a legitimate, correct answer wherever it’s honest — for a mechanism classification, or for an evidence tier, or both.
How this is marked#
Marking rewards accuracy and honest tiering across all ten mechanisms, not the number of mechanisms found present or the number of claims pushed to “directly evidenced.” A system correctly and thoroughly classified, with most claims honestly sitting at “plausible interpretation,” will score better than one that overreaches its evidence to look more conclusive than it is.
Afterwards#
This is the course’s last piece of work. Waiting is an experience, not merely a duration — the claim Week 1 made testable rather than merely plausible. This audit is where that claim gets applied in full, across every mechanism this course named, on the one real system you’ve now examined twice.
The spec
The brief poses the problem and leaves room for a response. The spec is the fixed part: what a marker checks. Some lines can be checked mechanically; the rest call for judgement.
- you have revisited the same system you analysed in the Queue Case Study; if you used a different system, you have briefly explained why the original became inaccessible, unsafe, or no longer observable
- you have classified each of the ten mechanisms — time, attention, expectation, information, interface, environment, queue structure, priority, value, control — as present, absent, ambiguous, or cannot determine
- for the five mechanisms your Case Study already covered, you have carried forward and re-tiered that evidence rather than re-deriving it from scratch
- for every claim you make about a mechanism, you have separately stated an evidence tier — directly evidenced, plausible interpretation, or undetermined — and you have not treated "directly evidenced" as equivalent to "caused"
- you have stated whether your Week 10 experiment is relevant to this system; if relevant, you have named the exact mechanism it tested and the conditions, and tiered that claim according to how strong your own experimental design actually was; if not relevant, you have said so directly
- you have not extended your experiment's result to any mechanism it didn't test
- you have included a whole-system synthesis and a limitations section addressing the wider scope of a ten-mechanism, single-system audit
- you have not offered a redesign, a recommendation, a fairness or efficiency verdict, or any claim about feelings, satisfaction, willingness to pay, or organisational motive without evidence for it
How it is marked
| Criterion | Weight |
|---|---|
| Accuracy across all ten mechanisms | 30% |
| Correct separation of classification and evidence tier | 25% |
| Honest, appropriately limited use of the Week 10 experiment | 15% |
| Quality of the whole-system synthesis | 20% |
| Honesty and specificity of limitations | 10% |