Library Pilot: More Visits or Just Later Visits?
The Westbrook Public Library extended its Friday closing time from 6 p.m. to 9 p.m. for a 12-week pilot. During the pilot, staff counted an average of 84 visits after 6 p.m. each Friday. The Eastside branch, which kept its 6 p.m. closing time, averaged 31 visits in the same Friday evening window. At the end of the pilot, the library left a short survey at the front desk; 210 patrons responded, and 78% selected 'extended Friday hours are important to me.' The board's draft report states: 'The pilot increased total library use and has strong public support, so Friday hours should be made permanent at all branches.' A board member notes that Westbrook's total Friday attendance, including before 6 p.m., was 410 per Friday before the pilot and 418 per Friday during it. The pilot cost $9,600 in overtime and utilities.
1. Does the evidence support the board's claim that the pilot increased total library use? Explain.
Read the worked answer
No. The relevant total is Westbrook's own Friday attendance: 410 before versus 418 during the pilot, an increase of only 8 visits, or about 2%. Meanwhile the branch recorded 84 visits after 6 p.m. That pattern is consistent with visits shifting from earlier in the day rather than new use. The Eastside comparison is weak because the branches may differ in size, location, and baseline evening demand, and the memo gives no pre-pilot after-6 baseline for Westbrook. The board's claim about total use is therefore not supported; the pilot may have changed timing more than volume.
2. What is the most serious weakness in using the survey to claim 'strong public support'?
Read the worked answer
The survey is self-selected. Only 210 patrons responded across 12 Fridays, while Friday attendance was roughly 418 per Friday, so responses represent a small fraction of visits and an even smaller fraction of residents. People who feel strongly about the library are more likely to answer, and the item wording ('important to me') invites agreement. The result can describe the views of respondents, not 'public support' in general. A random sample of residents, or at least a response-rate calculation with a neutral question, would be needed.
3. What additional evidence would best distinguish increased use from shifted use?
Read the worked answer
The best evidence is hour-by-hour visit counts for Westbrook before and during the pilot, covering the whole day and the same Fridays. If after-6 visits rose while pre-6 visits fell by a similar amount, the pilot shifted use; if total daily visits rose, it increased use. Also useful: compare the same branch across seasons, check whether other days changed, and use Eastside as a control only if the branches had similar pre-pilot evening patterns. A patron survey asking why they came at that hour would help but is secondary.