The Cafeteria Hours Memo
Read this excerpt from a hospital's internal announcement, then answer the questions.
Cafeteria Hours Change - Effective March 1 Beginning next month, the cafeteria will close at 7:00 p.m. instead of 9:00 p.m. A staff survey conducted last fall found that only 12% of respondents purchased hot meals after 7:00 p.m., so the change is expected to affect few employees. Administration projects the shorter hours will save $18,000 annually in labor and food waste. The survey was distributed through day-shift departmental email lists and was completed by 214 employees. Several night-shift nurses have written to the employee council noting that they were unaware any survey had been circulated, and that the cafeteria is one of only two food sources open to them after 8:00 p.m.
1. Which conclusion drawn from the survey is best supported by the text, and which is overstated? Cite specific details from the excerpt to defend both answers.
Read the worked answer
Best supported: The 12% figure cannot be generalized to night-shift staff. The text states the survey went out on day-shift email lists and drew 214 responses from a hospital that clearly employs night-shift nurses. Night-shift employees are therefore likely underrepresented in the sample, and the statistic may describe a population that is not the one most affected. A cautious, supported statement: "The survey measured day-shift purchasing behavior and may undercount after-7 p.m. demand."
Overstated: Any claim that night-shift staff do not use the cafeteria after 7 p.m., or that the change will affect few employees. The text gives no night-shift purchasing data at all, and it supplies counter-evidence - nurses report the cafeteria is one of only two food sources available after 8 p.m. Absence of survey data is not evidence of low demand.
The key distinction the exam tests is between what a statistic measures and the population it can legitimately describe.
2. An employee council member wants to state the case fairly in a one-paragraph letter. Which of the following two drafts is more defensible, and why? Draft A: "Only 12% of employees eat here after 7 p.m., so the change harms almost no one." Draft B: "The 12% figure comes from a day-shift-only survey, so the actual demand from night-shift staff remains unmeasured and should be checked before the change is final."
Read the worked answer
Draft B is more defensible.
Draft A commits two errors at once: it converts a percentage of respondents into a percentage of all employees (214 people, not the whole staff), and it treats a day-shift sample as representative of the whole hospital. It also moves from "purchased hot meals" to "eats here," which is broader than what was measured.
Draft B stays inside the evidence. It correctly identifies the sampling problem as the central issue, describes the gap as unmeasured rather than small, and proposes an action (verify demand before finalizing) instead of asserting a conclusion the data cannot carry. It also avoids contradicting the memo's own numbers, which a careless rebuttal often does.
When comparing two summaries, ask: does the draft claim more than the source measured, or does it flag exactly where the source is silent?