Quick answer
Calculate exact age on the cutoff date, then apply the current published rule for the specific authority. Both halves matter.
Two things, calculated separately
The arithmetic is: how old is this child on the cutoff date. The policy is: what does the authority do with that answer. Keeping them apart is what makes the result checkable, because the arithmetic is stable and the policy is not.
Collapsing them — deciding a child 'is a September starter' without writing down the age and the rule — produces a conclusion nobody can verify, including you in twelve months.
Use this guide with the right tool: Open the School Starting Age Calculator for age on a chosen school-entry date compared with a custom minimum age. If the question shifts, compare it with the Kindergarten Eligibility Calculator or the Age Range Calculator.

Cutoffs vary more than people expect
England uses 31 August, so the school year runs from September to August. Scotland uses late February with a deferral right. Many US states use 1 September but a substantial number use other dates, and some leave it to individual districts. Australia varies by state, with several offering a window rather than a fixed date.
The practical consequence is that a rule from a neighbouring area, an older sibling's experience, or a general article is not evidence. Only the current published policy of the responsible authority is.
The summer-born question
In systems with a fixed cutoff, children born just before it start school as the youngest in their cohort — nearly a year younger than the oldest. There is a substantial body of research on relative age effects, and several systems now offer deferral or delayed-entry routes in response.
Whether such a route exists, who decides, and what evidence is needed varies entirely by authority. If this is your situation, the question to ask is procedural — what is the process and what is the deadline — rather than whether the child is 'ready', which is a separate conversation.
Boundary cases
A birth date within a few days of a cutoff is a signal to verify rather than conclude. Check the date of birth against the birth certificate, check the cutoff against the current published policy, and check whether the wording is 'on or before' or 'before'.
That last one is not pedantry: a child born on the cutoff date itself qualifies under one reading and not the other, and both wordings appear in real policy documents.
Worked example
Check your child against the cutoff.
One September baby, four reference dates
The worked example separates two children by a single day of birth under a 31 August cutoff. Following one of those children — born 1 September 2021 — through a full year of plausible reference dates shows the same arithmetic producing a different school career depending on nothing but the date a rule happens to choose.
On 31 August 2026 her exact age is 4 years, 11 months and 30 days: one day short of five. Under a rule requiring five by 31 August, she does not start in September 2026. She starts in September 2027, on which date she is exactly 6 years 0 months 0 days old — a whole year of schooling decided by the single day between her age and the threshold. One day later than the first reference date, on 1 September 2026, she is exactly five, and under a rule written as "five by 1 September" she starts at once, the youngest possible reading of the same child.
Systems that start children in January rather than September meet her at a different age again. On 31 December 2026 she is 5 years, 3 months and 30 days; on 1 January 2027, 5 years and 4 months exactly. Under a January-entry rule with a minimum of five, she qualifies comfortably and starts a term later than the September systems would have taken her — not as a deferred or early case, but as the ordinary outcome of a rule whose reference date falls four months after her birthday instead of on its eve.
The deferral outcome deserves its own arithmetic, because it decides who her classmates are. If she waits and enters in September 2027 at exactly six, the youngest child in her cohort — born 31 August 2022, entering at 5 years and 1 day — has been alive for 1,827 days to her 2,191. A gap of 364 days inside one classroom, nearly a fifth of the younger child's entire life at that point. Had she started in 2026 under the 1 September rule, she would have been at the old end of a younger cohort instead. Her birth date, her readiness and her exact age on every one of these dates are identical in both stories; only the reference date moved, and it moved her by a year in one direction or the other.
That is the whole of the method this guide recommends, seen end to end. The exact ages — 4 years 11 months 30 days, exactly 5, 5 years 3 months 30 days, 5 years 4 months, exactly 6 — are stable facts that any calculator will reproduce. Everything that follows from them belongs to the authority whose rule supplies the reference date and the threshold. Record both halves, and a decision that looks arbitrary from outside becomes a straightforward reading of a stated age against a stated rule.
Common mistakes to avoid
- Using today's age instead of age on the cutoff. The cutoff date is the reference date. Today's age is irrelevant to the rule.
- Assuming a neighbouring authority's cutoff. Cutoffs are local and change. Check the current published policy for the specific authority.
- Rounding age up near the boundary. Rules use exact dates. Rounding is how a child ends up in the wrong cohort on paper.
Use the right calculator
Use the School Starting Age Calculator for school-entry age, the Kindergarten Eligibility Calculator for cutoff eligibility, and the Age Range Calculator for range checks.
Frequently asked questions
Can a child start early or late?
Some systems allow deferral or delayed entry with a process attached. Whether one exists, and who decides, is entirely authority-specific.
Does the calculator determine eligibility?
No. It gives the exact age on the date you specify. The authority's current rule determines what that age means.
What if two sources give different cutoffs?
Use the responsible authority's own current publication. Secondary sources go out of date and are frequently wrong.


