Method note

GovernorBrief – IDSR briefing for governors. How the briefing is made, in words. Applies to version 3.4.

What the tool does and does not do

It reads the tables in a school's Ofsted Inspection Data Summary Report (IDSR) and restates them as headlines and questions. Every step is a fixed rule. No AI is used, and the same file always gives the same briefing. It does not judge the school, predict an inspection outcome or explain why a result is as it is.

Reading the IDSR

Each measure table is matched by its column names. If the school-value, national or banding columns cannot be found, no briefing is produced.
A figure is read only if it is one plain number. Markers such as "x", "c", "SUPP" or "<5" are treated as "not published", never as a figure. Footnote marks after a number are ignored.
For each measure the newest published year is used. If that year is missing, the briefing says which year it used instead.
Part-year attendance figures are labelled as part-year.

Banding and significance are kept apart

For most measures the IDSR publishes two things:

What	Meaning	How the briefing shows it
Banding	Where the school sits among all schools: above, close to average, or below.	▲, ≈ or ▼ with the word.
Significance	Whether the difference from the national figure is larger than chance would explain ("sig+", "sig−", "non-sig").	▲▲ or ▼▼ in bold with "sig. above" or "sig. below".

A result is only called statistically significant when the IDSR itself marks it so. The tool performs no significance tests of its own.

Which results become headlines, and in what order

Broadly in this order:

Results that are significant on the IDSR's multi-year average. These come first because a multi-year figure is less affected by one small cohort.
Results that are significant in the latest single year, including for disadvantaged pupils.
Persistent absence, suspensions and exclusions that are above national.
The gap between the school's disadvantaged pupils and other pupils nationally. This is a gap to a national figure, not a significance test, and is labelled as such.
Attendance bandings and other context. Attendance has no significance test in the IDSR.

A result that is not significant is labelled "not statistically significant" wherever it is quoted. The order is fixed and cannot be changed by the user.

The overall picture line

This is a count of all-pupil results that carry a significance test: how many are significantly above national, significantly below, or neither, for the latest year and for the multi-year average. The measures overlap (for example, reading, writing and maths combined is built from the three subjects), so it is a count and not a score. It is left out when fewer than four results are tested.

Small cohorts

The briefing states how much one pupil moves a percentage: 100 ÷ the number of pupils. With 34 pupils, one pupil is about 2.9 percentage points.
A disadvantaged gap smaller than one pupil's worth is not reported as a concern.
Cohorts of 10 or fewer are marked †.

Hiding small numbers

With "Hide small numbers" ticked (the default):

counts and cohort sizes of 5 or fewer are shown as "5 or fewer", and rows for cohorts of 5 or fewer are left out;
suspension and exclusion reasons are left out;
a suspension or persistent-absence percentage is withheld when, together with its cohort size, it implies 5 or fewer pupils;
a total is not printed where subtracting a shown part would reveal a hidden one.

Attainment percentages are shown as the IDSR gives them. The threshold follows the IDSR, which suppresses cohorts of 5 or fewer.

Questions for governors

Each question is produced by a rule that fires on a pattern in the data, and is linked to a theme in Ofsted's State-funded school inspection toolkit. The short descriptions of what the toolkit looks for are paraphrases written for this tool, not Ofsted's wording. Questions do not assert a cause. Safeguarding and personal development and wellbeing are not evidenced by IDSR data, so no questions are generated for them.

Worked example: from an IDSR row to a sentence

This uses the built-in example school, whose figures are invented.

1. The IDSR row. In the table "All pupils – Writing expected standard", the multi-year row reads:

Year	Pupils	School	National	Banding
3-year	103	61%	72%	Below (sig-)
2026	34	65%	73%	Below (non-sig)

2. Reading it. School 61, national 72. The banding text contains "Below" (banding: below) and "sig-" (significance: significantly below national).

3. The rule. The multi-year row is significant, so it qualifies for the first group of headlines. It joins the other significant multi-year results that are below national: Writing greater depth (6% against 13%) and Writing progress (−1.9 against 0.0).

4. The sentence. "Below national over the IDSR's 3-year average, each statistically significant: Writing expected 61% vs 72%; Writing greater depth 6% vs 13%; Writing progress -1.9 vs 0.0."

5. The single year. The 2026 row is below national but "non-sig", so it produces no headline of its own. It appears in the table as "▼ below" in normal weight, and in a question labelled "not statistically significant". One pupil in a cohort of 34 is 100 ÷ 34 = 2.9 percentage points, so the 8-point difference is about 3 pupils.

6. The evidence trail. Appendix A of the full briefing lists the row, year, number of pupils, school and national figures and the banding exactly as published, against the headline's reference number.

Limits

Checked mainly on primary-school data. Secondary handling has had less testing; special schools are not yet supported.
Similar-schools comparisons in the IDSR are not yet used.
The briefing supports scrutiny. It does not replace the IDSR or professional judgement.