No Student Fails Overnight. Here's How to See It Coming.

Every teacher knows the feeling: the mid-term results come in, and there's a name you didn't expect. A student who seemed fine — present, quiet, not obviously struggling — has failed. And in hindsight, the signs were all there. The slipping homework. The two small tests that dipped. The handful of absences that didn't seem to matter at the time.

Here is the hard truth about academic failure: it is almost never sudden. A failed mid-term isn't the moment a student fell behind — it's the moment their falling behind became impossible to ignore. The decline had been underway for weeks. So the question worth asking isn't "how do we respond to bad results?" It's "how do we see the decline while it's still happening — while there's still time to change the ending?"

The good news: this is a solvable problem — and the research on how to solve it is unusually clear.

What actually predicts academic risk

Here's the part that should change how every school watches its students: risk is predictable. Decades of research into student outcomes, much of it from large-scale dropout studies, has converged on a small set of signals that reliably flag a child heading for trouble — long before a grade does. They're usually summarised as the ABCs: Attendance, Behaviour, and Course performance.

These aren't guesses. In national surveys of schools running early-warning systems, the indicators that most commonly trigger an intervention are course grades, attendance, and truancy — with effective systems typically watching around half a dozen signals in combination rather than relying on any single one. The reason the ABCs work is that they are leading indicators: they start moving before the final grade does.

The numbers behind this are striking. In a landmark study following almost 13,000 students for eight years, Johns Hopkins researcher Robert Balfanz and colleagues found that four simple sixth-grade signals — poor attendance, misbehaviour, and failing maths or English — could identify 60% of the students who would later not graduate. In high-poverty schools the link was even sharper: a sixth grader who attended less than 80% of the time, earned an unsatisfactory behaviour grade, or failed maths or English had roughly a 75% chance of eventually dropping out without intervention. Perhaps the most telling finding of all: a student's attendance predicted dropout better than their test scores did. How present and engaged a child is tells you more than how they scored on the last exam.

Attendance is often the earliest signal of all. A student who quietly slips from near-perfect attendance to missing a day here and there is frequently disengaging before their marks reflect it. Small, scattered absences are easy to overlook precisely because no single one looks serious.

Behaviour — in the broad sense of engagement, not just discipline — is the second. Reduced participation, incomplete homework, a sudden withdrawal from a student who used to ask questions: these are early distress signals that a numbers-only view of a class will miss.

Course performance is the third, but the key is to watch the trend, not the threshold. A student doesn't go from fine to failing overnight. A gentle, consistent decline across several small assessments is far more informative than a single low mark — and it shows up long before the report-card grade does.

Why schools catch it too late

So if the warning signs are this well understood, why do capable students still slip through? The answer is almost never a lack of caring teachers. It's a structural problem — not with who's watching, but with when and how the information arrives.

In most schools, the three signals live in three different places. Attendance is in one register, marks are in another, and the teacher's sense of a student's engagement lives only in that teacher's head. No single person sees all three for the same student at the same time. The class teacher notices the absences but not the dipping marks in the subject teacher's gradebook; the subject teacher sees the marks but not the pattern of absences. Each signal on its own looks minor. It's only when you put them together that a clear picture emerges — and in a manual system, nobody is positioned to put them together until the report card forces the issue.

The result is a system that is, by design, retrospective. It waits for grades to be finalised, then reacts. By then, the student has often been struggling for a full term — and the gap is much harder and more expensive to close than it would have been a month earlier.

This silent accumulation is well documented in India. ASER, the nationwide survey by Pratham, has for years found that large shares of children reach upper primary still unable to read text several grades below their level — in 2024, only about two-thirds of Class 8 students in government schools could read a Class 2-level passage. These are not children who failed overnight; they are children whose gaps were never caught while they were small.

This matters more under NEP than it used to. Continuous, competency-based assessment generates far more data points per student than the old two-exams-a-year model. That's an opportunity — there are now many more early signals to read — but only if someone, or something, is actually reading them. More data collected and never synthesised is just more paperwork.

What early detection looks like

The fix isn't more data — it's a different way of reading the data you already have. Instead of one backward-looking number, a continuous, combined view. The demo below shows the same class two ways. Switch between them, and watch who appears.

In the report-card view, every student is "passing" — nothing demands attention. Flip to the early-warning view and the same data, read as trends rather than totals, tells a different story: two students whose attendance, performance trend, and engagement are all sliding together, weeks before any of it would reach a failing grade.

The goal isn't to rank or label children. It's to direct limited teacher attention to where it's needed most, early. A teacher has finite hours; a system that quietly says "these students need a closer look this week" is simply making those hours count for more.

A crucial caution: the flag is the start, not the answer

Here is where honest writing on this topic has to part ways with the sales pitch. An early-warning signal tells you where to look. It does not tell you what is wrong, and it certainly doesn't tell you what to do.

In at least one rigorous district study, automated alerts had no measurable effect on attendance overall. Knowing that a child has failed mathematics is not enough to help them; you have to talk to the student, the subject teacher, and the parents to understand the actual cause — which might be anything from a vision problem to trouble at home to a single missed foundational concept.

So the right mental model is simple: the system flags, the human decides. A good platform raises the signal early and automatically, and then gets out of the way so a teacher can do the irreplaceable human work of figuring out the cause and the response. Any tool that claims to do the deciding for you is overselling — and any school that treats a dashboard as a substitute for a conversation will be disappointed.

How Prepdha brings the signals together

This is exactly the role Prepdha is designed to play. Because attendance, continuous assessment, and engagement data already flow through the platform as part of everyday teaching, the three signals that research says matter most are, for the first time, in one place for every student.

Instead of asking teachers to manually cross-reference registers and gradebooks, the platform surfaces the students whose combined signals are trending the wrong way — early, automatically, and ranked by how much attention they need. The class teacher and subject teachers see the same complete picture. And crucially, it stops there: Prepdha tells you who to look at and why the flag was raised, then hands the judgement back to the educator who knows the child.

The effect is to move the whole school from reacting to mid-term results to acting on early-term signals — turning the rich data NEP already requires you to collect into the early-warning system it was always capable of being.

The takeaway

Students at risk almost always signal it first — through attendance, engagement, and a slow slide in performance — weeks before a mid-term confirms it. Schools miss it not for lack of care but because those signals sit in separate places and only get assembled when it's too late to matter. Bring the signals together early, let them point teachers toward the students who need a closer look, and then trust teachers to do the rest. The mid-term will still come. It just won't be the first time you find out.