Bede families in Bangladesh face school access barriers from fixed-address rules

Project hypothesis (v3): Offline School Enrollment Checklist Tool for Mobile Bede Families Target users: Education volunteers preparing generic school-information questions for services used by mobile Bede families in Bangladesh; no eligibility screening or individual family record is part of the prototype. How it would work: An offline-first website helps volunteers prepare questions for schools serving mobile Bede families in Bangladesh. A blank checklist records public school contact, verification date and unanswered questions about fixed-address assumptions. Volunteers test fictional scenarios and school-confirmed public information, storing no family records or recommended alternative documents. First version: A printable offline website checklist operated by a volunteer coordinator with 6 adult volunteers at a consenting education partner over two weeks. Fields cover public school contact, verification date and unanswered enrollment questions. Tests use fictional requests, not children or family dossiers. Alternatives to investigate: Existing NGO-led enrollment drives, paper-based checklists, and informal advocacy by community leaders. Investigate if any schools already accept alternative documentation or if similar digital tools exist for mobile populations in Bangladesh. Difference to test: Compare the offline checklist with the partner's current paper notes for finding unanswered school-information questions in fictional requests. Any improvement would concern volunteer workflow, not enrollment rates or children's outcomes. Human responsibilities: A volunteer coordinator obtains partner permission and confirms public school information with educators. Adult volunteers test fictional scenarios and record errors; schools remain responsible for their actual requirements and admission decisions. AI contribution and limits: AI is not required for the first version. If expanded, AI could help match checklist items to varying school policies, but this is not in scope for the initial pilot. Data and access needed: Partner permission, public school contact information, a dated list of school-confirmed requirements and fictional test scenarios. Do not collect family identities, children's records or claims about accepted alternatives that schools have not verified. Source-supported setting and capacity: Built for boats, adrift on land: Bangladesh’s Bede community struggles with a fixed-address system Researchers from the study published in Cambridge Prisms: Global Mental Health contrast it with refugees who flee a single event: Bede families are almost always on the move, which makes them hard to reach for clinics, schools, and civil registers that assume a fixed address. Ordered steps and owners: 1. The coordinator confirms a willing education partner and public school-information fields before recording baseline task times. 2. A volunteer developer builds the offline checklist and tests it with fictional scenarios. 3. The coordinator compares completeness, errors and logged minutes before deciding whether to continue. Minimum resources and dependencies: Use an existing laptop, browser, paper notes and a volunteer coordinator; confirm partner permission, public-information verification and volunteer time before starting. No school system integration or family database is required. Costs and affordability unknowns: Costs are unknown or not reported; local partners must obtain quotes for volunteer time, printing, and minimal tech support before committing. How a newcomer can help: A newcomer volunteer can test blank checklist fields on fictional examples in a supervised session and report confusing labels. No contact with children, document handling or access to family identities is required. Pilot and success criteria: Proposed pilot target: 80% of fictional requests have all unanswered questions identified in two weeks, compared with the same volunteers using the partner's existing paper notes. Measure correctness and minutes per request using coordinator-reviewed logs; measure baseline before the trial rather than assume its value. Risks and safeguards: Stale school information could mislead volunteers, and an accessible interface may still be confusing. Date every confirmed item, leave unknown requirements unanswered and test with fictional data. Device access, maintenance effort and local affordability remain unverified. When to stop or change direction: Stop if public requirements cannot be confirmed, the partner cannot appoint a reviewer, the checklist increases errors or any personal family records are requested. Do not publish unverified alternatives or infer that a checklist changes admission rules.