Worked example
From attendance reporting to earlier investigation.
Northstar Learning Trust is an invented trust with ten invented schools. Everything on this page is fictional.
Fictional demonstration data
Fictional demonstration scenario. Predictive indicators support professional review; they do not determine interventions.
Trust attendance trend
Trust average compared with one school over six fictional weeks.
| Period | Trust | Hawthorne Academy |
|---|---|---|
| Week 1 | 94.6% | 94.2% |
| Week 2 | 94.5% | 93.6% |
| Week 3 | 94.4% | 93.1% |
| Week 4 | 94.3% | 92.6% |
| Week 5 | 94.4% | 92.1% |
| Week 6 | 94.2% | 91.8% |
School comparison
Change over the same fictional period.
- Northstar Central Academy95.2%+0.1 pts
- Oaklands Community School94.1%-0.2 pts
- Hawthorne Academy91.8%-2.4 pts
- Riverbank Primary96%+0.3 pts
- Westbridge Learning Centre93.4%-0.4 pts
Cohort view
The overall school percentage hides a larger change inside one cohort.
- Year 793.9%-0.3 pts
- Year 888.7%-4.1 pts
- Year 992.8%-0.6 pts
- Year 1093.1%-0.2 pts
- Year 1192.4%-0.5 pts
Emerging pattern
Fourteen fictional pupils whose patterns may warrant professional review. Pseudonymous labels only.
- Pupil 8-014
- Pupil 8-027
- Pupil 8-041
- Pupil 8-052
- Pupil 8-063
- Pupil 8-071
- Pupil 8-088
- Pupil 8-094
- Pupil 8-102
- Pupil 8-115
- Pupil 8-121
- Pupil 8-137
- Pupil 8-146
- Pupil 8-158
AI-assisted explanation
A draft explanation, with everything behind it visible
Simulated demonstration response. No live AI service is called on this page.
Hawthorne Academy's attendance appears to have declined for six consecutive weeks, from 94.2% to 91.8%. The largest change sits within the fictional Year 8 cohort (-4.1 percentage points) and is concentrated in one authorised cohort group. Fourteen pupils show patterns that may warrant professional review. Related factors, including incomplete reason codes, may affect this analysis and warrant investigation.
Why am I seeing this?
- Sources
- Fictional MIS attendance extract, Fictional cohort register, Fictional data-quality log
- Reporting period
- Six teaching weeks (fictional)
- Last refresh
- Simulated — 06:00 today
- Data quality
- Two days of incomplete marks; 6% of absences without reason codes
- Confidence
- Moderate. The direction of change is consistent, but the observation window is short and reason-code data is incomplete.
- Reason for prioritisation
- Prioritised because the change is the largest in the trust and has continued for four consecutive weeks.
- Records considered
- 5 schools, 5 cohorts, 6 weekly periods
- Next human review
- Fictional Director of Education, within five working days
Data provenance
Where these fictional numbers came from
| Source system | Fictional MIS extract (Hawthorne Academy) |
|---|---|
| Source field | Session attendance marks |
| Reporting period | Six consecutive teaching weeks (fictional) |
| Extraction date | Simulated nightly extract |
| Last refresh | Simulated — 06:00 today |
| Transformation applied | Sessions aggregated to weekly percentage by cohort |
| Data owner | Fictional Trust Data Lead |
| Quality status | Two days of marks incomplete in week 4 |
| Missing information | Reason codes absent for 6% of absences |
| Manual adjustment | None |
| Confidence limitation | Small cohort; short observation window |
Decision and outcome
Approved investigation, then measurement
A leader reviews the evidence and approves an attendance-team investigation. Progress is reviewed after four weeks and compared with the original baseline.
| Baseline | 91.8% at the point of review (fictional) |
|---|---|
| Approved action | Attendance team investigation across the Year 8 cohort |
| Owner | Fictional Attendance Lead, Hawthorne Academy |
| Review date | Four weeks after approval |
| Evidence recorded | Meeting notes, contact log summary, revised support plans |
| Later position | 93.0% at follow-up (fictional) |
| Interpretation | Position appears improved; the cause cannot be attributed with certainty. |
An improvement following an action does not prove the action caused it. The platform records the sequence and the evidence so leaders can form their own judgement.
Could this help your trust see concerns earlier?
Book a discovery call, or request a MAT Data and AI Readiness Review.