Use lessons 01–05 or equivalent skills before this exercise. You can revisit them from the lesson list.
A spreadsheet or calculator; optional local Python. Text-only AI can explain a plan but cannot establish that calculations ran.
Start with the idea
Deliver calculations another person can repeat and inspect.
Keep original rows, validation decisions, formulas and results together. Explain missing records and unequal tasks. A polished chart cannot replace that evidence.
Reproducible analysis: An analysis whose inputs, choices and calculations can be followed and repeated.
See the example and check your understanding
A made-up teaching example
The worked package traces source rows through validation to rates and a decision note. Missing duration stays missing. Unresolved records prevent a general comparison claim.
Ask yourself
Does a smaller known subtotal prove overall savings when work is still missing?
Show the explanation
No. State the incomplete coverage and unmatched workload. Keep exclusions and formulas inspectable.
Start with these checks
- Preserve source rows and cleaning decisions.
- Check formulas independently.
- Limit conclusions to comparable known work.
Use the different practice case below and check every requirement in its full instructions.
Full learning objectives
- Integrate cleaning, rates and quality checks.
- Preserve a reproducible method.
- Write a recommendation proportionate to evidence.
Captions appear here during playback.
Capstone: produce a reproducible decision analysis
Deliver calculations another person can repeat and inspect.
Video playback is unavailable. Read the descriptive transcript below, go directly to practice, or open the video file.
Source-based teaching example · Fictional data · Reviewed 2026-10-03. Microsoft Mark · locally synthesized. On a small screen, use fullscreen for diagram details or read the transcript.
Read the descriptive transcript
0:00 · What you will learn
Deliver calculations another person can repeat and inspect.
What appears on screen
Deliver calculations another person can repeat and inspect.
0:06 · The idea
Keep original rows, validation decisions, formulas and results together. Explain missing records and unequal tasks. A polished chart cannot replace that evidence.
What appears on screen
Keep original rows, validation decisions, formulas and results together. Explain missing records and unequal tasks. A polished chart cannot replace that evidence.
0:21 · One useful term
Reproducible analysis. An analysis whose inputs, choices and calculations can be followed and repeated.
What appears on screen
An analysis whose inputs, choices and calculations can be followed and repeated.
0:32 · A worked example
The worked package traces source rows through validation to rates and a decision note. Missing duration stays missing. Unresolved records prevent a general comparison claim.
What appears on screen
The worked package traces source rows through validation to rates and a decision note. Missing duration stays missing. Unresolved records prevent a general comparison claim.
0:45 · Your turn to think
Does a smaller known subtotal prove overall savings when work is still missing? Pause and decide before continuing.
What appears on screen
Does a smaller known subtotal prove overall savings when work is still missing? Pause the video if you need more time.
1:04 · Why this answer works
No. State the incomplete coverage and unmatched workload. Keep exclusions and formulas inspectable.
What appears on screen
No. State the incomplete coverage and unmatched workload. Keep exclusions and formulas inspectable.
1:14 · Check 1
Preserve source rows and cleaning decisions.
What appears on screen
Preserve source rows and cleaning decisions.
1:19 · Check 2
Check formulas independently.
What appears on screen
Check formulas independently.
1:24 · Check 3
Limit conclusions to comparable known work.
What appears on screen
Limit conclusions to comparable known work.
1:30 · Now try a different case
Open the lesson and try its different practice case. Use every check in the full task. Watching alone does not mark the lesson complete.
What appears on screen
Open the lesson Try the new case Check and revise The full task and its requirements are in the linked lesson.
Watching explains the example. Complete the lesson practice to check what you can do. Explore all visual lessons →
The method
Your package should contain the original fixture, data contract, cleaning log, formulas or code, results, and a short decision note. A reviewer should be able to recompute the values without asking a model to recreate hidden steps. Record optional AI assistance and the execution environment honestly.
Use separate gates for valid input, correct calculation, interpretable output and justified recommendation. Explain exclusions and missing repair costs before proposing an improvement. Keep the fixture fictional and avoid converting classroom numbers into a claim about VetsLearn or any provider. Peer review can help reveal errors but does not confer professional certification.
Choose your model, harness and offline route
Text-only assistant
Use supplied text. Ask for a plan or draft; do not claim browsing, code execution or file access that was unavailable.
Tool-enabled application or harness
Record which tools actually run, allowed files and actions, approval boundaries and observable results. Tools are optional unless the task needs them.
Local or cloud deployment
Inspect inference, retrieval, connectors, logs and storage separately. Use only authorized environments for real data; training fixtures are fictional.
Higher-capability or frontier model
Evaluate the exact task and configuration. Capability does not supply missing evidence or grant permission. No provider ranking or guarantee is implied.
Offline alternative
Use the supplied inputs and inspect the worked example manually. Builder exercises use local deterministic Python. No paid account or live model call is required.
Record the exact product/model label when available, tools actually used, inputs, permitted actions, observed output and checks. Live accounts or APIs may have external costs; they are optional and are not used by this page.
Read the general workflow guide →Worked example
Supplied input
Worked package outline: original rows → validation log → accepted rows → checked rates → decision note. A missing duration stays missing; no chart substitutes for the log.
A checked result
Result note: “These supplied cases support a limited comparison. Unresolved records and unmatched workload prevent a general claim.”
Why this works
Reproducibility includes both the math and the choices that determined which rows entered it.
Your independent capstone
Independent CSV-style fixture: id,method,minutes,pass. M1,manual,10,yes; M2,manual,12,yes; A1,assisted,6,yes; A1,assisted,6,yes; A2,assisted,8,no; A3,assisted,,yes. IDs are unique cases; duplicate A1 is accidental. A2 repair took 7 additional minutes. Workloads are not matched.
Download this task and checklist →Interactive practice / Fictional fixture
Reconcile the incomplete capstone dataset
Use the supplied independent case above. Check the bounded decisions, then review your written artifact against the full acceptance criteria.
Everything runs in this browser. No AI model is called and no work is uploaded. Bounded checks cover specific facts and rules. Your writing is not graded: you review it yourself. Practice never marks a lesson complete automatically.
Preserve original rows, exclusions, missing values and a matched follow-up plan.
Try the task, run the checks, then revise. Checks are teaching feedback, not an instructor grade.
Observed browser simulation
This trace comes from the bounded local checker. It is not a Python run, model result or production tool execution.
Self-review
Locked. Write every written part, then run the checks. A few words is not an attempt.
This is self-review, not grading. Read our answers, then tick what your own work does. Ticks are optional. They stay on this page and are not saved or sent anywhere. A different answer can be just as good if it meets the lesson criteria.
Cleaning log, calculations and decision note
A strong answer
Cleaning log: removed the second A1 row as an accidental duplicate; original rows kept. A3 has no minutes: left blank, not zero. Manual: 10 + 12 = 22 minutes, both pass. Assisted, known: A1 6, A2 8 plus 7 repair = 21 minutes. A3 unknown, so the assisted total is at least 21 and not final. First-pass rate: 2 of 3 distinct assisted cases, 66.7%. Decision: no savings claim. The workloads are not matched and one effort value is missing. Follow-up: run the same tasks both ways, record repair time and fill in A3.
A flawed answer
Assisted took 20 minutes against 22 manual and every case passed, so assisted saves about 9%. A3 was set to 0 because no time was logged.
What is wrong with it: 20 counts A1 twice and leaves out the 7-minute repair. “Every case passed” hides the first failure on A2. Zero for A3 invents a number. The workloads were never matched.
Keep or restore your work
Drafts are not saved automatically. Save only fictional practice data. A browser draft may be lost if storage is cleared; download a copy to keep it.
Your steps
- Preserve all original rows and log duplicate/missing entries.
- Compute known effort and pass rates over distinct cases.
- Include A2 rework; keep A3 effort unknown.
- Write a 200-word recommendation and a matched follow-up test plan.
Check your work
- Manual known effort 22 minutes; assisted known corrected effort 21 plus unknown A3.
- Initial assisted pass rate 2/3 after deduplication; manual 2/2.
- Do not infer overall savings from incomplete unmatched data.
- Keep formulas, exclusions and uncertainty inspectable.
After your attempt: review notes and a common mistake
Dropping A2 inflates quality, counting A1 twice alters the denominator, and setting A3 to zero invents effort. The strongest recommendation is a bounded matched trial or a justified pause, not a proven productivity claim.
Different wording or a different justified decision can pass. Check your work against every requirement; these notes are not an automated grade.
Explain your approach
Which missing value most affects your decision, and how would you collect it?
Keep your artifact, one correction or decision, its evidence and one remaining limitation. Nothing on this page sends or uploads your work.
Check your understanding
Choose an answer to see feedback. This is a knowledge check, not instructor assessment.
Review your independent practice
Complete the supplied task before recording progress. Use the acceptance criteria and reveal the review notes after your attempt. Keep your artifact yourself; nothing is uploaded.
Not yet self-reviewed. Completion stays in this browser and does not issue a badge.
Optional deeper study
Explore the related general-course lesson → This opens the existing canonical lesson; it is optional depth, not another copy of this focused task.
Primary references
Revision 2026-10-03. Fictional examples, not observed provider results. Knowledge checks and completion records are self-study aids.