Secondary E-Math OS Wrapper — O-Level Reliability Under Mixed Sets

Unified Almost-Code Exam Wrapper Spec (Registry-Clean) v0.2

PAGE-ID: EDK-WRP-OLEVEL-EMTH-v0.2
Role: WRP.Exam + MixedTopicStability + MethodSelection + PaceBudget + Transfer
Purpose: convert E-Math skill into reliable O-Level marks when questions are mixed, time is tight, and method choice matters.

0) Definition Lock

O-Level E-Math Wrapper := the exam-load layer that upgrades Math OS into O-Level conditions:

  • mixed topics (high switching cost)
  • method selection under time
  • multi-step word problems
  • graph/representation tasks
  • high careless risk late in paper

Lock: O-Level E-Math collapse is rarely “don’t know topic.”
It is usually wrong method + slow pace + error recurrence.

1) Output Spec (P3-OLEVEL-EMTH)

Target state:

  • stable method selection across mixed sets
  • stable accuracy on fundamentals
  • stable representation (word → equations/diagrams/graphs)
  • stable pace budget across the full paper
  • stable checking protocol under fatigue
  • transfer resilience (unfamiliar forms)

2) O-Level Load Model (What makes E-Math hard)

Load sources:

  1. Mixed-topic switching (context shifts)
  2. Method ambiguity (several tools plausible)
  3. Multi-step questions (errors compound)
  4. Time compression (pace budget)
  5. Careless under fatigue (late collapse)

Wrapper adds controls:

  • Method Signal policy
  • Mixed-set routing
  • Pace budget strategy
  • Error recurrence governance
  • Transfer ladder enforcement

3) Exam-Time Execution Protocol (OLEVEL.EMTH.EXAM.LOOP)

Protocol ID: OLEVEL.EMTH.EXAM.LOOP

Scan → Classify → Choose Method → Execute Cleanly → Check → Decide (skip/return) → Final Audit

Stage 1 — Scan (first 60–120s)

  • identify easy marks quickly
  • mark hard / time-sink items

Stage 2 — Classify (method signals)

For each question write a 1-line:
Signal → Method

Examples (E-Math typical):

  • “linear relationship” → form equation + solve
  • “percentage change” → multiplier method
  • “rate” → distance/time table
  • “graph” → gradient/intercept reading
  • “geometry” → angle facts + proof chain

Stage 3 — Execute cleanly

Clean working is an exam strategy: reduces careless, reduces time.

Stage 4 — Check

Mandatory check stack (see §5).

Stage 5 — Decide

If stuck > time budget, skip and return.

4) Pace Budget System (E-Math)

Pace Budget Rule:

  • “easy marks” must be fast
  • “medium marks” must be within budget
  • “hard marks” get controlled time or deferred

Operator policy (simple):

  • If no progress after 60–90 seconds: move.
  • Return after collecting available marks.

End-of-paper protection:

  • last 10 minutes reserved for audit + unfinished items.

5) Checking Protocol (E-Math Anti-Careless)

Check Stack (E-Math):

  1. arithmetic sanity / estimation
  2. sign scan
  3. substitution check for equations (when possible)
  4. units/labels check
  5. “answer the asked thing” re-read
  6. for graphs: confirm intercepts/gradient logically

Rule: B5 (careless) is treated as a system leak requiring a lab, not scolding.

6) Mixed-Topic Stability (Switching Control)

O-Level E-Math requires a “mixed-set engine”:

Mixed Set Rule

Train in mixed sets early (not only topic drills), but only after:

  • Foundations stable (Fluency Sensor OK)
  • Method signals installed (Method Selection Sensor OK)

Switching Policy

  • quick classification line reduces switching cost
  • prevents “try random method” behaviour

7) Transfer Ladder (O-Level)

Transfer levels:

  • T0 routine form
  • T1 rewritten wording / surface change
  • T2 multi-step with same idea
  • T3 mixed-topic chain
  • T4 unfamiliar exam variant

Wrapper rule:
A topic is “O-Level ready” only after it survives T3 at reasonable pace.

8) O-Level E-Math Error Buckets (B1–B6)

Use the same classifier, but typical O-Level patterns:

  • B2 Manipulation: algebra slips, fraction handling, ratio errors
  • B3 Method Selection: wrong approach wastes time
  • B4 Representation: word → equation/table/graph errors
  • B5 Careless: sign/units/reading wrong values
  • B6 Pace: paper not finished; late collapse

9) Wrapper Labs (E-Math Add-ons)

LAB-OE-1 — Method Signal Lab (Mixed)

  • 15–25 questions
  • choose method only
  • goal: fast classification

LAB-OE-2 — Mixed Mini-Paper Lab

  • 25–40 minutes
  • strict pace budget
  • builds switching stamina

LAB-OE-3 — Graph/Representation Lab

  • encode + interpret
  • gradient/intercept reading, scale traps

LAB-OE-4 — Careless Trap + Audit Lab

  • short set designed to trigger typical slips
  • trains final 10-minute audit routine

LAB-OE-5 — Transfer Mix Ladder

  • T0→T4 progression across weeks

10) O-Level E-Math Gauge (Wrapper Scorecard)

Weekly score:

  • Precision (mixed set accuracy)
  • Bandwidth (classification speed + structure recognition)
  • Speed (pace stability)
  • DriftRisk (careless spikes + time collapse)

Readiness targets (default):

  • Precision ≥ 3/5
  • Bandwidth ≥ 3/5
  • Speed ≥ 3/5
  • DriftRisk not High
  • Method Selection Sensor stable

11) Failure Mode Trace (E-Math wrapper)

Trace ID: EDK-TRACE-OLEVEL-EMTH-METHODPANIC-v0.2

mixed paper
→ slow classification
→ wrong method attempt
→ time wasted
→ panic switch
→ careless spike
→ unfinished paper

Repair

install method signals
→ mixed mini-papers
→ pace budget + audit routine
→ transfer ladder
→ stability returns

Start here if you want the full sequence:

Vocabulary OS Series Index:
https://edukatesg.com/vocabulary-os-series-index/

Fence English Learning System: 

eduKateSG Learning Systems: 

Recommended Internal Links (Spine)

Start Here for Lattice Infrastructure Connectors