Published and source-checked on September 3, 2026.
Research question
Is access to a usable shift-transfer process associated with lower voluntary turnover among eligible frontline employees?
Start with access, not approval rate
An approval rate alone is a poor measure of shift-transfer access. Employees may never submit a request if the opening list is incomplete, the form is unavailable off shift, eligibility rules are unclear, or supervisors discourage requests informally. The denominator should begin with employees eligible to use the process, then trace awareness, submission, review, decision, and completed transfer as separate stages.
NIOSH's worker well-being and quality-of-worklife materials treat work schedule, autonomy, supervision, and flexibility as related but separate parts of work experience. They justify measuring schedule access carefully, not claiming a universal retention effect. Department of Labor guidance on hours worked helps preserve the distinction between schedule records and compensable time. BLS tenure and JOLTS sources provide population context but cannot serve as a control group for a company study.
A stepped access study
Data Sources and Methodology
The study uses eligibility rosters, opening notices, distribution logs, requests, decision records, effective transfers, schedules, and separation records. Before rollout, assign an owner to each source and publish a dictionary for notice, eligible, submitted, decided, approved, and completed. An independent reviewer should trace a sample from the roster through the final outcome. This matters because a request system can show a decision without proving that the employee ever moved.
The preferred design is a phased introduction with rollout dates fixed for operational reasons before results are available. Estimate changes within each site and compare them with sites waiting for introduction. Include site and period effects, state how uncertainty is calculated, and show raw counts beside adjusted estimates. If sites choose their own rollout dates, record the reason and test whether earlier sites had different pre-rollout patterns.
Publish a participant flow from all rostered employees to eligible employees, employees with confirmed notice, and those with observable follow-up. Missing notice records must not default to "not interested." For qualitative context, recruit across shifts through a neutral channel and report participation limits without attaching comments to individual scheduling records.
Where operations permit, introduce a clarified transfer process in phases across comparable sites. The intervention could make current openings visible, publish eligibility criteria, give employees a route that does not depend on the current supervisor, and require a dated response. Keep the underlying staffing and qualification rules unchanged unless the research explicitly studies them.
Record the rollout schedule before outcomes are reviewed. For each eligible employee-period, capture whether the process was available, whether the employee received notice, whether a request was filed, decision timing, stated decision category, actual transfer date, and employment status. Keep employee preference text out of the analytic table when a limited code answers the question.
The primary comparison is the change in voluntary separation for eligible employees before and after access, relative to sites not yet introduced to the clarified process. Check whether groups followed similar trends before rollout. If rollout timing was chosen because a site already had severe turnover, state that selection problem and consider a different design.
Reading the funnel
Report each stage with its own denominator. Awareness uses eligible employees. Submission uses employees who received notice. Timely decision uses submitted requests. Completed transfer uses approved requests. Voluntary separation uses the defined employee cohort. Mixing these denominators can turn a bottleneck into an apparently strong approval rate.
Study timing as a distribution. Median response can improve while a small group waits much longer. Show upper-tail age and unresolved requests at the end of each period. Breakouts by shift, site, and job group are useful only when sample sizes protect privacy and the categories reflect real operating differences.
Approved and denied requests should not be treated as equivalent exposure. A denial may follow lack of openings, qualifications, tenure rules, attendance rules, or a temporary coverage constraint. Predefine a small set of factual reason codes and audit a sample against the underlying rule. "Manager discretion" is not an analytically useful category unless the discretion itself is being studied.
Confounding and unintended effects
Shift openings respond to vacancies and demand. A site with more turnover may offer more transfer opportunities, creating a backward relationship between access and exits. Employees seeking a transfer may already be considering departure. A completed transfer also changes the staffing conditions of two teams. The research should show these pathways instead of collapsing them into one adjusted percentage.
Guardrails matter. Track involuntary schedule changes, overtime, understaffed shifts, and time between approval and effective transfer. Faster approvals that leave the receiving employee waiting for months do not represent completed access. Nor should better transfer numbers come from pressuring workers to accept openings they did not request.
Privacy and legal boundaries
Do not infer childcare, disability, religion, medical status, or other sensitive reasons from a requested shift. Protected leave and accommodation requests belong in their established channels and should not be recoded as transfer demand. EEOC guidance on prohibited practices is relevant to consistent employment procedures, but this operational study is not a legal compliance determination.
Managers should see process-level reports. They should not receive a list of people assigned a presumed flight risk because they asked about another shift. Participation in interviews or surveys must be voluntary and separated from scheduling decisions.
Interpreting site differences
Site-level results need an operating narrative prepared without knowledge of the final turnover estimate. Record opening volume, qualification mix, shift premiums, hiring activity, temporary closures, and major schedule redesigns during each period. These facts help reviewers decide whether the comparison still makes sense. They should not be added selectively to explain away an inconvenient result.
Also compare the proportion of employees who could see at least one suitable opening. A technically available process offers little practical access when no compatible role or shift exists. Report that constraint beside awareness and submission. If one site has few suitable openings, its low request count should not be read as satisfaction with current schedules.
Limitations
A phased rollout is stronger than a simple before-and-after comparison, but it remains vulnerable to concurrent staffing changes, seasonal demand, and communication differences. Notice receipt does not prove comprehension. Administrative data miss informal requests. Voluntary-separation codes may be incomplete or inconsistently applied. Small sites may offer too little information for stable estimates.
Evidence-led conclusion
The proposed study can show whether making the transfer path usable changes request completion and whether voluntary turnover changes alongside it. The result will be most credible when access, approval, and actual movement remain distinct. Even a strong association will not prove that transfer access caused employees to stay. It can support a bounded decision about whether to retain, revise, or stop the access change and what evidence to collect next.
Sources
- Employee Tenure in 2024, U.S. Bureau of Labor Statistics. Accessed September 3, 2026.
- Job Openings and Labor Turnover Survey Handbook of Methods, U.S. Bureau of Labor Statistics. Accessed September 3, 2026.
- NIOSH Worker Well-Being Questionnaire, NIOSH. Accessed September 3, 2026.
- Quality of Worklife Questionnaire, NIOSH. Accessed September 3, 2026.
- FLSA Recordkeeping Requirements, U.S. Department of Labor. Accessed September 3, 2026.
- Hours Worked Under the FLSA, U.S. Department of Labor. Accessed September 3, 2026.
- Employment Tests and Selection Procedures, EEOC. Accessed September 3, 2026.
- Prohibited Employment Policies and Practices, EEOC. Accessed September 3, 2026.
- Statistical Quality Standards, U.S. Census Bureau. Accessed September 3, 2026.
- Privacy Framework, National Institute of Standards and Technology. Accessed September 3, 2026.