13,892: how to read the student suicide figure without overstating it

The National Crime Records Bureau recorded 13,892 student suicides in India in 2023. It is the highest figure in a decade, and it represents about 8.1 per cent of all suicides recorded in the country that year.

The number is now quoted in policy documents, news coverage and training decks, including ours. It deserves more care than it usually gets, because the way it is framed often outruns what it can support.

What it counts

NCRB compiles figures reported by state police. The category is occupational: a death is classified under "student" based on the person's status, not on any finding about cause. The dataset records the occupation of the deceased. It does not record why, and it does not attribute the death to academic pressure or to any institution.

This distinction is routinely lost. A rise in student suicides is not the same finding as a rise in exam-related suicides, and the data cannot separate them.

What it undercounts

Suicide is under-reported in India for reasons that are documented and largely structural: stigma, the historical legal position before the Mental Healthcare Act 2017, and inconsistency in how cases are certified and recorded across states. Research consistently suggests that official counts sit below the true figure.

So 13,892 is a floor rather than an estimate. That cuts against the temptation to read year-on-year movement too finely, because part of any change may be reporting behaviour rather than incidence.

Three things the number does not tell you

It does not tell you a rate. 13,892 is a count, not a proportion of the student population, and student enrolment has grown substantially over the same decade. A rising count with a rising denominator is a different claim from a rising rate, and the two are often conflated.

It does not tell you which institutions. State-level breakdowns exist, but they reflect where deaths were recorded rather than where a student was enrolled, and no public dataset attributes deaths to individual colleges.

It does not tell you whether anything works. A count going up does not by itself establish that prevention efforts have failed, any more than a count going down would establish that they succeeded. Evaluating an intervention requires the intervention to be measured, which almost none of them are.

What an institution can reasonably conclude

That the scale is large enough to plan around. That the figure is understated rather than overstated. And that any campus of a few thousand students should assume it will encounter a student in crisis, rather than treating it as improbable.

That is a sufficient basis for having a referral route, trained staff and a visible helpline number. It is not a sufficient basis for claiming a specific number of lives saved, and we would treat any training provider making that claim with suspicion, including if it were us.

Source: National Crime Records Bureau, Accidental Deaths and Suicides in India, 2023 edition. Figures for earlier years drawn from the same annual series.