Advanced · Data and analysis · module 140 of 155
Project lesson: a data analysis you can justify
Carry out a small analysis from question to report, with checks that someone else can verify.
After this lesson
You make an analysis with a plan in advance, checked data, honest charts and a report with source, limitations and cautious conclusions.
In view · How it connects
An analysis you can account for
Question, definitions and source in advance; data checked and the original kept.
Formulas recalculated, a fair chart and no invented causes.
Answer, source, limitations, and which steps AI carried out.
Part 1 of 3 · Plan and data
Your preference for less movement is on. Choose each part yourself.
Plan and data
Choose a question from your work with a concrete decision behind it and write out the analysis plan: question, definitions, source, period and calculations (lesson 136). Check the data for empty fields, duplicate rows and outliers, and note what you changed. Work with fictional or anonymised data if you use AI, and only in a permitted tool (lesson 138). Keep the original export untouched, so that you can always go back to the source.
Analysis, chart and check
Let AI help with formulas, calculations or code, and recalculate at least one result by hand (lesson 109). Choose a chart that fits the question, with the axis, the period and the source included (lesson 137). Compare every number in your report with the calculation: a typo in a percentage goes unnoticed in fluent text. Do not let AI fill in causes you have not investigated; a link is not yet a cause.
Report and accountability
Write a short report: the question, the answer in one sentence, the figures, the chart, the source, the limitations and what you recommend. For a forecast, give a range (lesson 139). Record in an appendix which steps AI did and which you checked yourself. Have a colleague review the report with the question: can you find every number again? Only when that works is the analysis sound.
New practice example
From explanation to a usable result.
Someone else can follow a responsible analysis: every number can be traced, and the limitations are stated.
Fictional practice situationFictional: a library investigates whether the new Saturday opening hours attracted more borrowers. The first report says: “Thanks to the new hours, 25% more young people came.”
View a possible solution
This is how it can be done
After checking: the number of borrowers on Saturdays rose from 120 to 150 per Saturday, 25% more. The share of young people was not counted separately, so I delete “young people”. The cause remains unknown: a reading club also started in the same period. Report sentence: “On Saturdays, an average of 25% more borrowers came; whether that is due to the new hours cannot be concluded from these figures.”
Why this works
Every word in the conclusion now rests on a measured figure. The possible other cause is stated honestly.
Make it slightly more difficult
Think of a simple way to investigate the influence of the new hours more effectively.
Extra practice, at your own pace. You do not need to submit anything. This does not count towards an automatic assessment.
Your workplace · learning by doing
Make something. See what changes.
Checking the AI connection… You can already view the written example.
Present your assignment to Ende.
Read the answer, change one thing, and compare.
Previously held
New version
You remain in control
What do you want to improve in your next attempt?
On this device only. AI answers remain only in this page memory. This is self-review, not a grade.
Why this approach works
Every word in the conclusion now rests on a measured figure. The possible other cause is stated honestly.
A step further
Think of a simple way to investigate the influence of the new hours more effectively.
Try & discover · no obligation
Place source and concept side by side
Adjust the values in the concept. See what difference you would give back to the writer.
Fictional analysis report on a satisfaction survey, checked against the calculation: 212 respondents and 64% satisfied. The survey did not ask about causes. Compare with what is in the draft of the report.
| Part | Source | Concept | Check |
|---|---|---|---|
| Number of respondents | 212 | 212 | Same |
| Percentage satisfied | 64% | 68% | Difference |
| Cause of the satisfaction | Unknown | the new website | Difference |
Targeted feedback you could provide
- Percentage satisfied: This percentage does not match the calculation. Check whether the formula includes a total row. Ask to retain the source value "64%", unless a change is confirmed.
- Cause of the satisfaction: This value is not known in the source. Do not fill in the empty field yourself. Ask to leave this as unknown.
You are only changing three fixed fields of a fictitious concept. This is not a full document analysis or a check of a real upload.
This trial works locally with fictitious data and does not count towards your lesson completion. It starts anew with a new lesson or reload.
A prompt to get you started
Help me build a short analysis report for this fictional question: “Has the number of visitors to our Sunday market risen since the new car park?” Data: counts per Sunday from March to October, in 2025 and 2026. Give the structure of the report, the calculations I need to make, one suitable chart and the limitations I need to mention. Do not fill in any figures I did not give.
Apply it
An analysis report mentions a satisfaction rate of 68%. On rechecking, the calculation turns out to give 64%: the formula included a row with a total.
Which approach best fits this situation?
- Leave 68%; the difference is small.
- Round both to “about two thirds”, so nobody sees the difference.
- Correct the figure, fix the formula and have a colleague check the other numbers in the report.
With JavaScript, you can check your approach and receive immediate feedback.
Test your insight
Your report says that visitors rose by 13% thanks to the new car park. You only examined the numbers. What do you change?
- Nothing; the timing shows that the car park is the cause.
- You write that visitors rose by 13% and that the cause does not follow from these figures.
- You leave out the rise, because you cannot prove the cause.
Extra practice
Carry out a small analysis with fictional data, from plan to report. Have a colleague trace three numbers from the report back to your calculation.
Question: did the average number of visitors per Sunday rise from 2025 to 2026? Answer: from an average of 410 to 465 per Sunday, a rise of 13%. Chart: a line chart per Sunday, with the axis starting at 0 and the counts at the entrance as the source. Limitation: in 2026 there were two fewer market days because of rain, and whether the car park is the cause does not follow from these figures. My colleague found all three numbers again.
