Risk-Signal Frequency Baseline Across Five Jurisdictions, 2023–2026

Note on scope and schema

The shared data-report schema has no risk-signals report type. This report uses the openings-closures type as the closest valid value and frames its content as a baseline of risk-signal frequency across the five host jurisdictions. The signals below are the registry’s own computed flags — derived from public register fields such as approval validity, intake windows and award configuration — not independent audit findings. They indicate where a record diverges from a “clean” baseline, and they are starting points for scrutiny, not verdicts.

What a risk signal is

Each registry record carries a set of flags. One flag, verified_ok, marks a record that raised no computed concern. Of 4,351 records, 1,728 (39.7%) are flagged verified_ok only. The remaining 2,623 records (60.3%) carry at least one risk signal. A record may carry several signals at once, so the per-signal counts below sum to more than 2,623.

Frequency of each risk signal

Ranked by number of records carrying the signal:

1. intake_window_expired · 1,171 records · The advertised student-intake window has passed its end year. Common in older programmes whose approval permitted intake only through a specified year.
2. no_foreign_award · 990 records · The programme does not confer the foreign partner’s award; students receive only the local credential. Relevant to students expecting a dual or foreign degree.
3. approval_expired · 746 records · The approval certificate’s validity period has lapsed without a recorded renewal or termination in the registry.
4. hk_exempted · 436 records · A Hong Kong course operating under an exemption from registration under Cap. 493 — lawful, but outside the standard registration framework.
5. no_cn_award · 261 records · The programme does not confer a Chinese award; the local credential is absent, which matters for recognition in mainland China.
6. not_in_national_plan · 209 records · The programme is not included in the national enrolment (gaokao) plan; admission is by separate application rather than unified enrolment.
7. sg_no_edutrust · 179 records · A Singapore record without an EduTrust certification marker in the registry.
8. sg_no_foreign_award · 125 records · A Singapore delivered-award record that does not confer the foreign award.
9. distance_no_f2f · 107 records · The programme is delivered at a distance without face-to-face teaching, a modality flag rather than a defect.
10. non_degree_level · 14 records · The programme sits below degree level.
11. branch_no_programme · 6 records · A branch-campus record with no associated programme listed.
12. ae_khda_only · 6 records · A UAE record known only to Dubai KHDA, not to the federal CAA.

Risk-signal frequency by jurisdiction

1. Mainland China (CN) · 3,277 records, of which 1,951 (59.5%) carry at least one risk signal · The largest absolute population of signals, driven by intake_window_expired, no_foreign_award and approval_expired against an ageing approval base.
2. Hong Kong (HK) · 723 records, 463 (64.0%) with a signal · Heavily shaped by hk_exempted (courses lawfully exempt from registration) and no_foreign_award.
3. Singapore (SG) · 317 records, 203 (64.0%) with a signal · Led by sg_no_edutrust and sg_no_foreign_award.
4. Malaysia (MY) · 11 records, 0 with a signal · All 11 branch-campus records are flagged verified_ok in this dataset.
5. United Arab Emirates (AE) · 17 AE records, 0 with a signal; the 6 AE-DXB Dubai free-zone records all carry ae_khda_only · Within the strict ae code the branch campuses are clean; the free-zone subset is flagged for federal/CAA absence.

How to read these signals

A risk signal is a prompt for a question, not a finding of harm. For example, intake_window_expired most often means a programme’s approved intake years ended and it is no longer admitting — which may be a planned wind-down rather than a failure. no_foreign_award and no_cn_award describe the credential structure and are material to a student’s recognition expectations but are not defects. Researchers should treat the frequencies here as a map of where scrutiny is most likely to be rewarded, then confirm each record against the live source register.

Data limitations

1. Flags are computed from register fields, not from fresh audit. They can lag reality: a programme renewed yesterday may still show approval_expired until the registry rebuild captures it.
2. verified_ok is a baseline, not a guarantee. A clean flag means no computed divergence was found; it does not certify quality or financial viability.
3. Signal meaning differs by jurisdiction. hk_exempted is lawful in Hong Kong; the same “exemption” concept does not transfer to China or Singapore.
4. Counts are not mutually exclusive. A record with three signals is counted once in the 2,623 total but three times across the per-signal list.
5. The registry is a 2023–2026 snapshot. The period fields denote the observation window of the current rebuild, not the dates on which risks first arose.

---

This report is compiled from the shared TNEstudy programme registry. Flag definitions and frequencies are computed from public register fields at the current rebuild. Researchers should confirm individual records against the live source registers before drawing conclusions about any specific institution or programme.