WF-10 — Learner Progression, Remediation and Graduation Clearance

IoL cohorts are small, so this KPI is not a statistic. It is a list of names. With a cohort in single or low double figures, one learner is worth five to ten percentage points of retention, and the difference between a departure and a deferral is worth the same again. Treat KPI 2.2 as a case-by-case learner support process with a data consequence, not as a rate to be managed. The correct way to improve it is to notice a struggling learner in week four, and the correct way to record it is to classify what actually happened rather than what is easiest to enter.

1. Purpose

To ensure that every IoL learner's progression is monitored against their study plan, that difficulty is identified early enough for support and remediation to work, that any departure or interruption is classified accurately against the controlled reasons the Ministry publishes, and that graduation clearance produces a complete, reconciled graduate record capable of carrying the learner into the twelve months of outcome tracking on which almost half the programme-level framework depends.

2. Scope

Scope statement. This process manages learner progression from the first progression review after registration, through monitoring, support, remediation, interruption or departure, to graduation clearance and the creation of a complete graduate record.

Applies to. All registered PGDipHPE and MScHPE learners, including those on deferral, leave of absence or suspension, those transferring between IoL programmes or to and from other MBRU programmes, and those completing by dissertation or capstone.

Does not apply to. Admission and initial registration (WF-07), the conduct of assessment itself (WF-09), the investigation of an academic integrity allegation or an appeal against an academic decision (WF-11), or post-graduation outcome tracking (WF-25). This workflow hands over to WF-25 at the moment the graduate record is complete.

Applicable requirements. OBEF guide v11.5 KPI 2.2 including its counting rules and the one-year programme rule; KPI 3.2 and the Grad_Workplacement reconciliation; KPI 2.5 cohort definitions; Appendix B evidence for student outcomes; CAA progression and award standards; MBRU academic regulations; CHEDS Students - Attrition.xlsx and Students - Graduates.xlsx, with the HEDB lookups Leave of Absence Reasons, Suspension Reasons, Voluntary Withdrawal Reasons and Dismissal Reasons.

The counting rules that materially change the number, all of which apply here.

• Institution level: a learner who transfers from one programme to another is not attrition.
• Programme level: a transfer out is attrition, unless the destination programme has comparable admission criteria.
• Deferral of registration, or of a semester, is not attrition. Such learners remain in both the numerator and the denominator. Dropping them is the most common way a small programme understates its own retention.
• Where learners are not admitted to a specific programme in year one, progression into the intended major counts as retention.
• For one-year programmes this KPI is replaced by the one-year graduation rate. At institution level, learners in one-year programmes are excluded from both numerator and denominator, unless every programme at the institution is a one-year programme.

The last rule may apply to PGDipHPE and must be confirmed. A postgraduate diploma is commonly a one-year award. If PGDipHPE is flagged One Year Program = Y in Institute - Academic Programs.xlsx, then its OBEF 2.2 value is its one-year graduation rate rather than a year-one-to-year-two retention rate, and its learners drop out of MBRU's institutional 2.2 entirely. [IoL to confirm with WF-01] the nominal duration of both programmes and the value of that flag. This is not a detail: it changes what is computed, for which programme, and whether those learners appear in MBRU's institutional number at all.

3. Trigger, boundary and endpoint

4. SIPOC

5. Accountability

Process owner. IoL Programme Director, or the role able to approve remediation plans, decide progression and recommend an award.

Escalation.

6. Process steps

1. Monitor progression against the study plan. Track credits attempted and achieved, assessment outcomes, engagement and supervision milestones. [CAPTURE] progression status per learner per period. Serves 2.2. For a cohort of this size, monitoring is a conversation and a list, not a dashboard. The point is that the conversation happens on a schedule and that its outcome is written down.

2. Identify a learner of concern early. Triggers include a failed assessment, two weeks of disengagement, a missed supervision milestone, a placement difficulty, or a change in the learner's employment or clinical rota. IoL learners are practising professionals; the most common cause of difficulty is workload at work, not capability, and it is visible before it is decisive.

3. Agree a support or remediation plan. Named actions, named owner, review date. [CAPTURE] the plan, its date and its review outcome. Serves 2.2 and, through completion, 3.2 and 1.1. This is the step that actually moves KPI 2.2. Everything else in this workflow records what happened.

4. Deliver and review remediation. Record the outcome, including where remediation did not succeed.

5. Decide progression. Progress, progress with conditions, repeat, defer, or exit. [CAPTURE] decision, date, decision-maker. Serves 2.2.

6. Classify an interruption or departure against the controlled vocabulary. [CAPTURE] Attrition_Student_ID, Attrition_Academic_Period, Attrition_Category, Attrition_Reason, Attrition_Program_Code, using the HEDB lookups Leave of Absence Reasons, Suspension Reasons, Voluntary Withdrawal Reasons and Dismissal Reasons. Serves 2.2. Decision point, and the one with the largest numeric consequence in this workflow. Apply the counting rules stated in section 2:

   • a deferral or leave of absence is not attrition; the learner stays in both numerator and denominator, and must not be removed from the cohort;
   • a transfer between programmes is not attrition at institution level; at programme level it is attrition unless the destination programme has comparable admission criteria, tested against the transfer-equivalence table from WF-07;
   • a dismissal following an integrity case is attrition, and the classification comes from WF-11;
   • where a learner was not admitted to a specific programme in year one, progression into the intended major counts as retention. Record the rule applied alongside the classification, so the judgement is visible rather than embedded.

7. Confirm continuation into year two. Join the learner's registration in the current academic year to their registration in the previous autumn period. [CAPTURE] continuation flag. Serves the 2.2 numerator. The numerator is built from Students - Enrollment.xlsx in the following year, joined on Enroll_Student_ID, not from an absence of an attrition record. A learner with neither a continuation record nor an attrition record is a data gap, not a retained learner, and must be chased.

8. Verify award requirements. Credits, placement, dissertation or capstone, and any professional requirement. [CAPTURE] clearance decision and date.

9. Reconcile the placement flag. Compare the placement register from WF-13 against the intended Grad_Workplacement value for every graduating learner. [CAPTURE] Grad_Workplacement. Serves 3.2 and cross-KPI assertion C3. Exception route. A disagreement is resolved by investigating which record is wrong, never by overriding one to match the other. Every graduate flagged Y must have at least one matching row in Students - Internship.xlsx, and every learner with a completed placement episode must be flagged.

10. Complete the graduate record. [CAPTURE] Grad_Student_ID, Grad_Academic_Period, Grad_Program_Code, Grad_GPA_Cumulative, Grad_Master_Thesis_Title where applicable, Grad_Submission_Term, Grad_Student_Email_Address, Grad_Student_Mobile_Number. Serves 1.1, 1.2, 2.5, 3.1 and 3.2. Re-confirm the personal contact set at clearance, at the last moment MBRU has the learner's attention. This is the handover point to twenty percentage points of Pillar 1, and a graduate who cannot be reached scores nothing however well employed they are.

11. Define and release the graduate cohort. The cohort defined here is the population for KPI 1.1 and 1.2 twelve months later, for KPI 2.5's micro-credential component in the reporting year and its licence component three years later, for KPI 3.1's denominator among those who completed a placement, and for KPI 3.2's numerator and denominator. [CAPTURE] the cohort definition and the date it was fixed. Serves all of the above. Because KPI 2.5's two components use different cohorts by design, the reporting-year cohort for micro-credentials and the cohort of three years earlier for licences, the cohort definition must be stored per year and never recomputed from a current extract.

7. OBEF data generated

Capture rule. Three things must be recorded at the time and cannot be reconstructed afterwards: the classification of a departure, with the reason and the rule applied, because six months later nobody remembers whether a learner deferred or withdrew and the two are counted differently; the graduate contact set re-confirmed at clearance, because it is the last moment MBRU controls; and the cohort definition as fixed for the year, because KPI 2.5 uses two different cohorts and a recomputed cohort silently changes both. The continuation flag can be re-derived from registration data, provided nobody has overwritten Enroll_1st_Academic_Period upstream.

Reproducibility test. A second analyst can reproduce KPI 2.2 from Students - Enrollment.xlsx across two academic years joined to Students - Attrition.xlsx, provided three conditions hold: the one-year programme flag is correct, the transfer-equivalence table exists so the programme-level transfer rule can be applied, and deferral reasons are mapped to the HEDB lookups rather than free-typed. The second condition fails today, because no transfer-equivalence table exists anywhere at MBRU as far as this pilot can establish, and it is WF-07 action R6. Until it exists, any programme-level transfer must be documented individually with the reasoning recorded.

8. Max-score design

Why the two levels differ, and why that matters to IoL. At programme level, a transfer out counts as attrition unless the destination has comparable admission criteria. At institution level, the same transfer counts as retention. The same learner therefore appears in two numbers with two different treatments, which is exactly why the register warns that the institutional result is not a simple average of the programme-level results. IoL should compute and report its programme numbers, and let WF-26 compute the institutional number independently.

What this workflow must do to reach the target.

1. Classify every departure against the controlled HEDB reason lists, never free text. Owner: Data Steward.
2. Never record a deferral or leave of absence as attrition, and keep those learners in both numerator and denominator. Owner: Data Steward.
3. Confirm the one-year programme flag for both programmes, and apply the substitution rule if it applies. Owner: Programme Director with WF-01.
4. Reconcile Grad_Workplacement to the placement register before every graduation, with zero unreconciled exceptions. Owner: Data Steward.
5. Re-confirm graduate personal contact data at clearance. Owner: Registry.
6. Fix and store the graduate cohort definition per year. Owner: Data Steward.
7. Run progression review on a schedule, with a written outcome per learner, because this, and not any data rule, is what actually retains learners. Owner: Programme Director.

[REDESIGN] actions.

Sequencing note. R1 comes first, because until it is settled nobody knows which KPI PGDipHPE is being measured on. R3, R4 and R5 are free and immediate. R2 is the only action here that improves the underlying outcome rather than the record of it, and in a small programme it is worth more than all the data controls combined.

9. Indirect strategy where data cannot be collected

Category 4 and Category 5 both apply, and it is important to be clear about which is which.

Category 4, activity exists but is invisible. Most of what protects retention at IoL already happens informally. Programme leads know when a learner's clinical rota has changed, when a dissertation has stalled, and when someone is about to disappear. The support is real and is often effective. What does not exist is the record: the concern, the plan, the review and the outcome. Without it, the department cannot show a reviewer that retention is managed, cannot learn from the pattern across cohorts, and cannot defend a classification when a learner's departure is later queried. The response is capture, at steps 2, 3 and 4, and R2 and R7 make it routine.

Category 5, performance genuinely low, is the honest category if retention is genuinely poor. If a cohort loses learners because the programme demands more than practising clinicians can give, that is a real finding and no classification rule will improve it. The sanctioned response is to report it accurately and open a corrective action plan on the MoHESR CAP template. It is worth saying plainly that a small programme has a strong incentive to reclassify its way to a better number, because one reclassification moves the rate by several points. That incentive is exactly why R3 puts two people on the classification decision.

A word about the small-cohort arithmetic, because it cuts both ways. With twelve first-year learners, one departure is 8.3 percentage points. Two departures move a programme from 100% to 83%, from the top of the High band to the middle of it. This means the number is volatile and a single year tells you very little, which is one reason the KPI is a three-year rolling average. It also means a single misclassified deferral is worth more than any process improvement available in a year. Report the value with its cohort size, in every internal paper, so nobody reads an 83% as a decline when it is one learner who took a clinical secondment.

Category 2 does not apply. KPI 2.2 is Track A and computed from registration records, not from a survey, so Appendix C sampling thresholds are irrelevant to it. Do not group cohorts for this KPI. Grouping is a survey remedy and applying it here would be a misuse of an Appendix C provision.

Boundary check. This workflow must never:

• record a deferral, leave of absence or suspension as attrition, or drop such a learner from the denominator;
• reclassify a departure after seeing its effect on the retention rate;
• treat a programme-level transfer as non-attrition without a documented comparable-criteria judgement;
• compute the institutional retention figure as an average of the programme figures, since the two levels treat transfers differently;
• override Grad_Workplacement to match the placement register, or the register to match the flag, instead of investigating which is wrong;
• release a graduate record without valid personal contact data, or with an employer address as the only contact point;
• recompute a prior year's graduate cohort from a current extract;
• apply the one-year graduation rate substitution to a programme that is not flagged as a one-year programme, or fail to apply it to one that is.

10. Service standards

11. Records and evidence

Appendix B readiness. Today, IoL could most likely produce registration and progression records, and transcripts evidencing placement completion, within 15 working days. It could not produce transfer-equivalence documentation, because no equivalence table exists, and it could not produce a dated graduate cohort definition for a prior year, because cohorts are currently recomputed rather than stored. Both are addressed by R6 here and R6 in WF-07.

12. Risks and controls

13. Performance measures

The retention percentage is the measure everyone will look at, and it is the least informative row in this table for a cohort of this size. The capture and accuracy measures are what make it mean anything.

14. Change control