01 · What breaks
Design for the failure modes, not only the happy path.
Early-warning lists fail quietly. Teachers stop entering data, flags turn into blame, transfers inflate dropout figures and a home visit reveals more than a family wanted shared.
Paper registers entered twice
Where government schools keep a paper register, asking teachers to type the same marks into an app doubles their work, and the digital copy is the first thing to lapse. Photograph the register and extract it instead, with a volunteer checking every extracted tick against the image before it counts.
Punitive use of risk scores
If a list of high-risk children reaches a head teacher as a performance measure, the easiest fix is to drop names from the register. Use flags only to start support. Keep them out of staff appraisals and school rankings, and never show a child's flag to classmates or the wider community.
Missing transfers counted as dropout
A child who moved with family to another district may already be enrolled elsewhere. Counted as a dropout, they distort the figures and use up follow-up time. Keep the case open until the new enrolment is confirmed, and record transfer as its own closure state with the evidence behind it.
Contact that exposes sensitive family information
A volunteer asking about absence at the door, within earshot of neighbours, can expose debts, illness or a marriage being arranged. Agree who may contact a family, through which channel, and what gets written down. Sensitive reasons go in a restricted field that only the case owner and supervisor can read.
The Right of Children to Free and Compulsory Education Act, 2009 gives every child aged 6 to 14 the right to free and compulsory education in a neighbourhood school, and a child who dropped out has the right to be admitted to an age-appropriate class with special training. The RTE rules require local authorities to keep a record of children up to age 14, drawn from a household survey updated every year. A programme's follow-up records can support that work but do not replace it. Take legal advice before sharing case data with any authority.
02 · Minimum viable model
Treat each spell of absence as an episode with a start and an end.
The core record is a follow-up episode that opens when a threshold is crossed and closes with an outcome. The learners are under 18, so under the Digital Personal Data Protection Act, 2023 a parent or lawful guardian must give verifiable consent before their personal data is processed; most of these duties apply from around May 2027, as of October 2026. Capture it at enrolment, not at the first home visit. A data collection app that works offline usually feeds the attendance side.
| Closure state | What it means | Evidence to attach | Counted as dropout? |
|---|---|---|---|
| Returned | Attending again for an agreed period, for example two full weeks | Attendance records for that period | No |
| Transferred | Enrolled in another school or an open-school programme | Admission confirmation from the new school, or a document the family shares | No |
| Exited | Left education after follow-up, with the reason recorded in the family's words | Case owner's note and supervisor sign-off | Yes, with the reason |
| Unresolved | Family not reached or situation still unclear after the agreed attempts | Attempt log with dates and channels | Reported separately |
03 · Implementation sequence
Move through reviewable gates.
Build the system in the order a field team will use it: decide what counts, collect it cheaply, flag sparingly, check the flags, follow up safely and close honestly.
Define absence and dropout carefully
Write down what counts as a day absent, a late mark and a school holiday, and when a run of absences becomes a case. Dropout should be a dated closure state with a reason, never a box someone ticks. A child stays an open case, not a statistic, until that case is closed.
Use the least burdensome attendance source
Use what already exists: a facilitator's phone register in community centres, a photographed government register in schools. Enter or extract it within 48 hours so a run of absences is visible while something can still be done. A monthly tally arrives too late for a child who stopped coming three weeks ago.
Set review thresholds with field teams
Start with a few plain rules and let field teams adjust them after a term. Programmes often begin with three consecutive absences creating a follow-up task, and attendance under 60% for two weeks alerting the mentor. Add closer attention in June and July, when the Class 5 to 6 and 8 to 9 moves happen.
Test false positives by subgroup
After the first term, look at who the rules flagged and what follow-up found. If girls in one block, or children from migrant families, are flagged far more often but mostly turn out to be fine, the threshold is wrong for them. Change it, and record the change and the reason.
Assign safe follow-up
Give each task to a named person trained for the conversation, with a due date and a channel the family has agreed to. Seasonal migration, sibling care and safety worries need different responses. Anything that suggests harm to the child goes to a safeguarding lead the same day, not to the next review meeting.
Close with evidence and reason
Close a case only with an outcome and its evidence: back in class, enrolled elsewhere, left after follow-up, or unresolved. For a transition case, keep it open until the child appears on the new school's register. Review unresolved cases monthly, because a growing pile usually means follow-up capacity is too thin.
04 · Evidence
Measure whether the work became clearer.
The test is whether a flagged child gets a timely, safe conversation and a clear outcome, not how many flags were raised. The education sector playbook shows where these measures sit in a wider school programme.
05 · Responsible automation
Let systems prepare. Keep consequence with people.
Systems can support
- Extracting ticks from photographed paper registers
- Creating follow-up tasks when an agreed threshold is crossed
- Comparing flags with follow-up findings by subgroup
- Reminding case owners about overdue and unresolved cases
People should decide
- Whether and how to contact a family
- Safeguarding response
- Confirming transfer or dropout
- Changing support for a child
06 · Common questions
Practical questions teams ask.
Questions school programme teams and field coordinators ask before switching on attendance alerts.
What is a school dropout early-warning system?
It is a set of attendance records and simple rules that put a child on a review list when absence patterns suggest disengagement, such as a run of missed days or a falling trend before a transition year. A trained person then follows up and records what happened. It is a prompt for a conversation with the family, not a prediction or a label attached to the child.
What data does an attendance early-warning system need?
Daily or session attendance with its source and entry date, the child's school and class, the reason for absence as the family gave it, known context such as migration season, each follow-up attempt and its result, and a closure state with evidence. Identifiers that help match a child who changes school, such as a PEN where it is shared, cut down false dropouts.
Who should follow up when a child is flagged?
A named, trained person who already has a relationship with the family, such as a community mentor or facilitator, backed by a supervisor. Teachers can share what they know, but they should not carry every home visit. Safeguarding concerns go to a designated lead straight away, and the programme manager confirms any closure as a transfer or a dropout.
What mistakes make attendance alerts fail?
Asking teachers to enter attendance twice, so the data stops arriving. Using flags to judge schools or staff, which encourages names to vanish from registers. Counting transfers as dropouts. Setting thresholds once and never checking false positives. And recording sensitive family reasons where too many people can read them. Each of these erodes trust faster than any software fault.
Can a small nonprofit run an early-warning system without custom software?
Yes, for a few schools or centres. A shared sheet with attendance, a formula that highlights three consecutive absences and a follow-up log can work if someone reviews it every week. Move to a proper system when several field teams are involved, entry has to work offline, family details need restricted access or children must be matched across schools.