Prompt that checks whether a test is analysed on the same unit it was assigned on. Paste it into an AI tool of your choice, which then checks the unit of analysis in tests.
Results depend on the tool, the model and what you enter.
Result preview
Example: written by our editors, not real AI output.
Assignment: person
Metric: click rate per page view
Risk: 3 % of people = 41 % of views
Correction: click rate per person, then mean
Recomputed: lift +2.1 % → +0.6 %, interval includes 0
Fill in
What you type stays in this browser window. We store and send nothing.
What is assigned?
person
device
session
account
mixed
What do you measure on?
page view
click
order
session
mixed
What shape is it in?
event table
daily aggregate
export
unclear
mixed
Your prompt
Help me check the unit of analysis in my test.
Assignment: [Assignment]
Metric: [Metric]
Data: [Data]
Steps: 1. Name the unit of assignment and the unit of the metric. 2. Flag cases where one person produces many rows. 3. Explain why many dependent rows understate the spread. 4. Propose: aggregate to the assignment unit, or use robust error estimates. 5. Recompute the result where possible. 6. Write the rule for future tests.
Output: unit comparison, risk, correction and rule.
Take particular care to analyse on the same unit that was assigned.
Before you answer, check: Are all instructions met? If key information is missing or a placeholder in square brackets is still in the text, ask instead of guessing. Mark assumptions as assumptions.
You have edited the prompt. Changes in the fields above no longer overwrite it.
This is an instruction for an AI, not the result itself.
Prompt that explains a cohort table, compares cohorts at the same age and separates chance differences from real trends.
Prompt · Intermediate · about 8 min
You may use our prompts freely, including at work. Please do not republish them as your own collection. We give no warranty for AI results; check them yourself. Terms of use
