Two Accounts of the Four-Day Week
Two documents describe the same six-month pilot at a factory.
Document A (company newsletter): "Our four-day-week pilot raised output per worker 9% and cut sick days by one third. Voluntary turnover fell from 14% to 9%. We are extending the four-day week to all departments in January."
Document B (letter to the editor, from a machine operator): "Output rose, but not because we worked smarter. My line dropped two weekly meetings totaling 90 minutes and stopped a nightly changeover. Two of the eight people on my line left and were not replaced; the rest absorbed the work. Our pay is unchanged because the pilot froze salaries. The 14% turnover figure counts only people who quit, not people who transferred out."
1. On what point do the two documents agree, and where do they part company?
Read the worked answer
Agreement: Both say output per worker rose. Document B opens with "Output rose," so the operator is not disputing the measurement - only its cause and its meaning.
Disagreement:
- Cause. A attributes the gain to the four-day week itself. B attributes it to specific removed tasks (90 minutes of meetings, a nightly changeover) and to unfilled vacancies.
- Fairness. A frames the result as a benefit delivered. B frames it as work compressed onto fewer people with frozen pay.
- Measurement. A reports turnover as a single number. B argues that number is defined too narrowly.
Reading caution: the two documents cover different facts, not contradictory ones. A says nothing about meetings, changeovers, or salaries; B does not deny that output rose. A careful reader should not treat this as a lie-versus-truth pair. It is one set of numbers with two explanations attached, and the explanations are the real dispute.
2. Document B offers an alternative explanation for one of Document A's results. Identify the result and the explanation, and describe the evidence that would help decide between them.
Read the worked answer
Result: the 9% rise in output per worker.
B's explanation, in its own terms:
- Two weekly meetings (90 minutes total) were eliminated - time returned to production.
- A nightly changeover was stopped - less downtime per day.
- Two of eight positions went unfilled - the same output divided across fewer workers raises the per-worker figure without anyone working faster.
Why point 3 is the strongest: "Output per worker" is a fraction. Removing workers from the denominator raises the number even if total output is flat. B's claim is arithmetically possible, which is what makes it a serious alternative rather than a complaint.
Evidence that would decide it:
- Total output, not per-worker output, for the pilot months and the same months a year earlier.
- Headcount by line, showing whether total staff fell.
- Whether other departments, which did not drop meetings or changeovers, showed the same 9% gain. If they did, the task changes are not the explanation.
- Whether the changeover was eliminated or moved to another shift - moved work is not removed work.
3. The operator says the 14% turnover figure "counts only people who quit, not people who transferred out." Why does that matter for judging the claim in Document A?
Read the worked answer
What A claims: "Voluntary turnover fell from 14% to 9%," offered as evidence that employees were happier under the new schedule.
What B's objection does to that claim:
- Turnover is a definition before it is a number. If only resignations count, then an employee who leaves her line for another department inside the same factory disappears from the statistic while still leaving the job.
- If B's line lost two of eight people who "transferred out," the line's effective loss is 25% even though none of it appears in the 14%-to-9% trend.
- This does not prove A is wrong. It shows A's number measures one kind of departure and is being used to support a broader conclusion about satisfaction. That is a gap between the evidence and the claim, not a falsehood.
What a reader should do with it: treat the turnover figure as consistent with the pilot helping retention, but not as proof of it, until the definition of turnover is stated and transfers are counted. Note also that A never says the drop was caused by the four-day week - the newsletter implies it by placing the two facts side by side.