evidoria

← Back to browse

Good practice Imported

SiPKPM AI Early-Warning System — Malaysia's National Dropout-Prevention Network

Malaysia · Putrajaya · See the Malaysia profile · See the Putrajaya profile

Evidence: Observational / pre–post Top 31% 60/100 · Ask Evidence Copilot about this practice

Malaysia's Ministry of Education added AI to its national Student Tracking System (SiPKPM), scoring attendance, grades, disability status, household income and more across roughly 5 million students. The AI-enhanced version helped over 9,000 at-risk students, including 3,287 girl

8,000+ candidates
SPM candidates who did not sit the exam (no-shows) (2025 SPM cohort)
4,758 students
Dropout students helped back into school
9,000+ students
At-risk students re-enrolled in 2025 (2025)
3,287 girls
Of whom girls re-enrolled (2025)
SiPKPM AI Early-Warning System — Malaysia's National Dropout-Prevention Network

Details

Maturity
Established
Promoter
Ministry of Education Malaysia (KPM) / UNICEF Malaysia
Period
2024-2025 (AI-enhanced rollout); SPM 2025 cohort
Keywords
education administration, student welfare, early-warning analytics, public sector AI

Context

Malaysia's Ministry of Education (KPM) added an AI layer to SiPKPM, its long-running national Student Tracking System, to flag students at risk of dropping out or missing the SPM school-leaving exam before they disappear from the system.

Activities

The AI model scores each student on seven factors — attendance, academic performance, disciplinary record, disability or learning-support needs, household income, distance from school, and parents' marital status — then sends structured alerts to teachers, counsellors, district and state education offices, and the ministry itself. Flagged students can be matched to one of 18 forms of ministry assistance, from financial aid to learning support.

Results

Government and press reporting in February 2025 said the AI-enhanced system had already cut the number of no-show SPM candidates from tens of thousands to just over 8,000, and a follow-up report put the number of dropout students helped back into school at 4,758. UNICEF's own account of the 2025 rollout cites more than 9,000 at-risk students re-enrolled that year, of whom 3,287 were girls.

Conclusions

These are ministry- and UNICEF-reported figures rather than an independent evaluation, and none of the sources reviewed publish a data-protection or bias-audit policy for a model that profiles children on income, disability and family status.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Medium (1–3 years)
Staffing & skills
Ministry of Education Malaysia (KPM), UNICEF Malaysia (partner)

Conditions for success

  • Built on the pre-existing SiPKPM national student tracking system already covering roughly 5 million students
  • Structured alerts routed to teachers, counsellors, district/state offices and the ministry, linked to 18 forms of ministry assistance

Common failure modes

  • No public algorithm documentation, bias audit or data-governance policy published for a system profiling students on income, disability and family status

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful