Month 11 · Test exceptions and compare evidence
Four relative self-study weeks and one activity with fictional manual tests. No real performance claims, external actions or guaranteed study time.
This is the developed Month 11 learning resource, not the entire twelve-month course. The four exercises form one monthly activity M11. Reading, personal marks and self-scores do not verify academic completion.
Complete fictional F11 source
Separate timing sample: before five comparable completed tasks, total 30 minutes,2 corrections; after five comparable completed tasks, total 25 minutes,1 correction. Individual tasks/times are not supplied. Manual exception inputs below are a separate test set;N01 reference priceCOP 8000, stock/delivery unknown.
| Row | Test ID | Input | Expected boundary |
|---|---|---|---|
| 1 | T01 | 2 N01 | Reference subtotal only |
| 2 | T02 | N01 quantity absent | Clarification |
| 3 | T03 | 1 X99 unknown | Clarification |
| 4 | T01 | 2 N01 exact repeat | No new request |
| Additional | T05 | Incomplete; no reference/quantity | Pending, not completed |
No actual durations are measured by this page. No tool caused the sample difference by established evidence. T05 must never be added as a completed sixth task.
Week 1 · Define tests before running
Goal: Specify inputs and expected results.
A test needs identified input, a rule, expected result and observed evidence. Write expectations first: changing criteria after seeing outputs hides errors. F11 continues a fictional manual demonstration, not connected software.
N01 costsCOP 8000 in the exercise; stock and delivery are unknown. T01 requests 2 N01;T02 requests N01 without quantity;T03 requests 1 X99, unknown product; the fourth row exactly repeats T01. No names, contacts or payments. Additional T05 is incomplete and remains pending.
Separate calculation from confirmation. A normal case may produce a reference subtotal, but reserves no stock and sends no reply. Missing data or unknown reference requires clarification. Exact repetition must not become a new order.
Case for review
A plan calls every numeric output successful despite missing data.
Your activity
Define normal, missing-data, unknown-product and duplicate tests with expected/observed/pending fields. Add a separate T05 check.
Worked feedback — compare after trying
T01 allows 2×8000=COP 16000 for products only. T02 cannot calculate without quantity;T03 without known product; repeated T01 is no new order. T05 stays pending. No stock, delivery or sending is confirmed.
Supporting artifact 1: prior-expectation matrix.
Week 2 · Trace failures and evidence
Goal: Record outcomes without hiding exceptions.
Walk through rows in order and compare against expectations. An unauthorized promise fails even with correct arithmetic. Manual evidence is useful when source, steps and results are preserved; invent no server call.
Do not remove exceptions to clean the report. Incomplete T05 stays visible and outside completions. Pending does not mean corrected or a commercial response sent. State what is known rather than what would look impressive.
Keep before/after drafts and explain the rule behind corrections. When a duplicate was counted twice, identify both source rows and a deduplication note. Delete no actual records or accounts in this exercise.
Case for review
A report removes T05 and counts the duplicate as new work.
Your activity
Document two fictional failures: invented T02 subtotal and T01 confirmation. Record correction, rule and repeated manual check. Preserve T05 pending.
Worked feedback — compare after trying
T02 needs quantity, not assumed 1. T01 may showCOP 16000 with limits, not confirmed order. Exact repetition adds no request. T05 remains incomplete. Written corrections improve the exercise artifact, not prove a real system improved.
Supporting artifact 2: failure/correction/recheck log.
Week 3 · Compare time and corrections
Goal: Calculate differences within a comparable sample.
F11 supplies two fictional sets of five comparable tasks: before 30 minutes/2 corrections; after 25 minutes/1 correction. These are sample totals, not your measurements or actual APEX results. The additional incomplete case is not one of the five completed tasks in either set.
Time difference 30−25=5 minutes. Descriptive reduction relative to before:5/30×100≈16.7%. Average per task:30/5=6 minutes before,25/5=5 after. Totals do not tell us each individual duration.
Corrections decline 2 to 1: one fewer,50% relative to 2. You may describe 0.4 and 0.2 corrections per task, but not 40% and 20% of tasks with errors: one task could have multiple corrections. Small aggregate comparisons prove neither causation nor future improvement.
Case for review
A chart promises guaranteed 16.7% productivity improvement and zero errors.
Your activity
Calculate absolute difference, relative percentage with before base, and averages. Write a correction observation and two limits. Explain why six tasks are not counted.
Worked feedback — compare after trying
Five fewer minutes,16.7% descriptive reduction,6 versus 5 minutes/task. One fewer correction,2 to 1, not zero. Incomplete T05 is excluded. Attribute no difference to AI or future outcomes; these are fictional aggregate conditions.
Supporting artifact 3: base-labeled calculations and bounded conclusion.
Week 4 · Present and assemble M11
Goal: Deliver reproducible evidence with pending work.
Preserve source, expectation, result and correction. Keep exception tests separate from timing data: their rows are not identified as the five measured tasks. Sharing a month does not establish a row-level link. Invent no such correspondence.
Include limits: no live tool, no connected account, no personal data, no external actions and T05 pending. A future measurement proposal should define comparable aggregate records and authorized review before collection; do not measure real workers or launch campaigns.
Combine four artifacts into M11. Rubric v 1: facts/sources, artifact/process, data/human review, reasoning/limits; each 0–2. Zero absent/wrong, one partial, two sufficiently justified. Minimum 6/8 without zeros requires authorized human review, not page approval.
Case for review
The conclusion says all tests completed and the business optimized despite pending T05.
Your activity
Rewrite it and attach matrix, failure log, calculations and limits. Score using evidence and a concrete improvement per partial criterion.
Worked feedback — compare after trying
Valid:“A fictional manual walkthrough documented exceptions and one pending case; a separate five-task comparable sample changed 30 to 25 minutes and 2 to 1 corrections.” This guarantees no operational optimization. A 6/8 without zeros may be reviewed;6/8 with zero fails. No statements or benefits.
Single M11 submission: tests, corrections, descriptive comparison and self-review.
Assemble your single monthly package
Download M11 package guide (TXT)
Keep your work locally. Formal review requires actual artifacts and an authorized human reviewer. No certificate, enrollment or benefit is issued by this page. The other eleven months and final portfolio remain separate parts of the annual program.
Optional AI tutor — read consent and limits first · My personal study marks, not academic verification