Commute and schedule fit are often discussed as personal preferences, and they behave more like structural constraints. A shift that starts before public transit runs, or ends after the last reliable bus, excludes a group of workers regardless of motivation. A schedule that changes weekly makes childcare and a second job difficult to arrange. The evidence for these effects is partly public and mostly local, and the two should not be blended. This report describes what public data can support and how an employer can test the local question.
Published November 4, 2026
The question and the scope
The question is: for a defined group of frontline employees, does the fit between shift timing and available transportation predict whether they remain in the role? The unit is one employee and one schedule period. The observation window is a defined quarter, with a follow-up at the end of the next quarter. The scope includes commute time, transit availability at the shift start and end, and the number of schedule changes in the period.
The scope excludes a judgment about whether an employee should be willing to commute further. That framing treats a structural constraint as a personal shortcoming and produces advice that does not solve the problem. The review focuses on the conditions the employer can change, such as shift start times, schedule stability, and the location of the work. The employee's choice remains the employee's, and the employer's job is to make the choice workable.
Why public data cannot answer the local question
Public commute data describes averages across regions and industries. The Census Bureau publishes commuting characteristics, and the Department of Transportation publishes national travel statistics. Those sources can describe the share of workers who drive alone or the average travel time in a region. They cannot tell an employer whether the last bus reaches a particular site at the end of a particular shift.
The same limit applies to time-use data. The American Time Use Survey describes how people spend time across the population, and it does not observe a single employer's schedule. A team that uses a national average to set a local policy is making an analogy, not a measurement. The public source is useful for framing the issue and for comparing the local pattern with a broader context. It is not a substitute for the local record.
Define three fit measures
The first measure is scheduled start and end times compared with the available transit windows. Record the first and last reliable trip for the site and the shift. A shift that begins before the first trip or ends after the last trip has a structural gap. The gap may be small enough to solve with a adjusted start time and large enough to exclude a group of workers.
The second measure is the number of schedule changes after publication. A worker who can plan a commute around a stable schedule is in a different position from a worker who learns about a change the day before. Record the change date and the notice given. The measure is about the stability of the schedule, not about the worker's flexibility.
The third measure is the actual arrival pattern. Record whether employees arrive on time, arrive late, or report a transportation problem. A pattern of lateness on one shift may point to a schedule that does not match the transit window. The measure should be paired with a conversation, because the record alone cannot distinguish a transit problem from a personal one.
| Measure | Data source | What it can show |
|---|---|---|
| Transit window gap | Local transit schedule and shift times | Whether a shift is reachable |
| Schedule changes | Published schedule and change log | Stability of the working pattern |
| Arrival pattern | Time records and supervisor notes | Whether the gap is affecting attendance |
Build the local record
Start with the site's shift times and the transit options that serve it. Record the first and last trip for each day of the week, including the weekend and holiday schedules. Note the walk or transfer time between the stop and the site. A schedule that looks reachable on a map may be unreachable in practice because of a transfer that does not connect.
Add the employee's usual mode and the constraints they have shared. Do not collect more personal information than the review needs. A note that an employee relies on a specific bus is enough. The record should be stored with the schedule and reviewed by the people who can change the shift. It should not become a general file about the employee's life.
Compare like with like
Compare employees on the same shift and in the same role before comparing across the site. A difference between two teams may reflect the manager, the work, or the schedule. A difference within a team is easier to interpret because many other conditions are shared. Keep the comparison small and describe the groups rather than ranking individuals.
Use the local record to test a specific change, such as moving a start time by thirty minutes or publishing the schedule a week earlier. Measure the arrival pattern and the schedule changes before and after. Do not change several conditions at once, because the result will not tell you which change mattered. A single change with a clear before-and-after is more useful than a broad program.
Read the result with restraint
Report the transit gap, the change frequency, and the arrival pattern as separate counts. Avoid a single fit score, because it hides which condition is causing the problem. A site with a small transit gap and many schedule changes has a different problem from a site with a large gap and a stable schedule. The remedies are different, and the score would obscure them.
Compare the local pattern with the public context only to orient the reader. A regional average travel time can help a manager understand whether the site is unusual. It cannot establish that the site's schedule is the cause of a retention difference. Keep the public context in a separate paragraph with a clear label.
Where the evidence stops
The review cannot show that a commute constraint caused a particular employee to leave. The employee may have left for a different reason and mentioned the commute as a convenient explanation. The review can show that the constraint existed and that the employee's departure followed. That sequence is useful and it is not proof of cause.
The review also cannot capture the workers who never applied because the shift was unreachable. That absence is invisible in the employee record and can be larger than the effect among current employees. Pair the review with a check on how the shift is described in the job posting. A posting that does not mention the start time may attract applicants who cannot take the shift.
A protocol for one site
Choose a quarter and define the three measures, the data sources, and the comparison groups. Collect the transit windows and the shift times, and reconcile the schedule changes against the published schedule. Review the first ten arrival records with the supervisor and note the cases that need a conversation. Freeze the definitions and log exceptions.
Report the counts, the change frequency, and the arrival pattern. Add a short list of the changes the team intends to test, such as a later start or an earlier publication. Repeat the review after the change using the same definitions. The comparison is more useful than a single snapshot because it holds the site's context constant.
Limitations and next action
The sample is one site and one quarter, and the results depend on the local transit system and the labor market. A rural site may have no transit at all, and the review should say so rather than force a measure. The protocol should be repeated before it is used as a baseline. A single review is a description, not a benchmark.
The next action is to pick one schedule condition and test a fix. A thirty-minute start change or a two-week publication window is small enough to implement and specific enough to evaluate. The team can then see whether the arrival pattern moves and whether the employees who were excluded can now apply. That is a more useful question than whether commute matters in general.
Data Sources and Methodology
- Census Bureau commuting data
- Bureau of Transportation Statistics
- BLS American Time Use Survey
- BLS Job Openings and Labor Turnover Survey
- DOL Fair Labor Standards Act
- NIOSH Total Worker Health
- EEOC laws and guidance
- Census Bureau Quarterly Workforce Indicators
- Federal Reserve Economic Data
- U.S. Small Business Administration