Error Log Engine

The practice test measures; the error log is what makes you better — review is the work, the test is just the diagnostic. Top scorers improve by converting every wrong answer into a labelled, re-practised pattern. Skipping this loop is the single most common reason capable students plateau.

Sources: learning science (retrieval, spacing, self-explanation — Dunlosky et al.; Roediger & Karpicke); College Board My Practice review workflow; MySatCoach, Galvanize. Examples MERIDIAN-original.

---

#THE LOOP (run daily on every miss + every lucky guess / shaky correct)

1. LOG the row (template below).
2. TAG the error type — ask "with unlimited time, would I get this right?"
     NO  → CONTENT gap (relearn the concept)
     YES → CARELESS (slip) · MISREAD/TRAP (wrong thing / distractor) · TIMING (ran out)
3. SELF-EXPLAIN in writing: why the right answer is right AND why each wrong one is wrong.
     Verify against the official explanation (Bluebook/Khan). An unchecked wrong explanation reinforces the error.
4. RE-ATTEMPT the question from scratch the same day.
5. SPACED RE-TEST: revisit the item a few days later (Anki for recurring ones).
6. WEEKLY PATTERN SCAN: sort by domain + error type; any 3+ cluster = next week's deliberate-practice target.

Why (mechanism): logging forces retrieval; self-explanation builds transfer; tagging tells you what kind of fix is needed (a concept gap and a careless slip need opposite responses); the weekly scan turns scattered misses into a targeted plan. The score report only shows domain bars — the log is the layer that says which sub-skill, and whether it's knowledge or execution. [This loop is the ~80%-of-gains engine.]

---

#THE TEMPLATE (copy into a spreadsheet or notebook)

| Date | Source/Q | Section | Domain | Skill | My ans | Correct | Error type | Why I missed it (1 sentence) | Fix / re-practice | Confidence 1–5 | Time | Revisit date |

• Confidence + Time columns are the tell: high confidence + wrong = dangerous misconception (priority); right but slow = a timing problem, not knowledge.
• Keep it ruthless and short — one sentence per "why." The point is the pattern, not a diary.

#TRIAGE — REVIEW IN THIS ORDER (highest leverage first)

1. Missed easy/medium items (preventable points — biggest, fastest gains).
2. Repeated misses in the same domain/skill (your structural leak).
3. Questions that took too long (timing leak).
4. Hard misses that still matter for the goal.

#SAME APPROACH, TWO LEVELS (MERIDIAN-original)

• ~700 habit: takes 4 practice tests in a week, glances at the score, moves on. Result: repeats the same Inference and conditional-probability misses every test — no pattern ever surfaces. [plateau]
• 800 habit: takes 1 test, spends two days logging + self-explaining every miss, scans for clusters, drills the 2 clusters that appear, then re-tests. Result: those clusters stop recurring. Move: treated review as the work.

#DRILLS / USAGE

Drill 1 — Self-explanation (every miss): write the "why right / why each wrong." PASS: you can articulate the trap, not just the answer. Drill 2 — Weekly scan (every Sunday): sort the log; name your top 2 clusters; set next week's worst-first targets. PASS: next week's plan is set from the data. Repetition: the log is permanent; clusters should shrink week over week (track the count).

#TRANSFER

Abstract rule: every error is data; label it, explain it, re-practise it, re-test it. Works for both sections, all topics, and (frankly) any exam — including your IAL papers. Constant: the 6-step loop + the 4 error types. Changes: what's in your log.

#REFERENCE CARD — ERROR LOG

LOG every miss → TAG (content / careless / misread / timing) → SELF-EXPLAIN (why right + why each wrong, verify)
→ RE-ATTEMPT same day → SPACED RE-TEST → WEEKLY SCAN (3+ cluster = next target)
Triage: easy/medium misses → repeats → too-slow → hard. Confidence+Time columns separate gaps from slips.
Review > volume. The test measures; the log improves.

---

MERIDIAN™ · © 2026 · For Neo's personal study use. Not affiliated with or endorsed by the College Board. SAT® is a trademark of the College Board.

MERIDIAN · FORGE · ERROR LOG ENGINE | v1.0