Spot selection bias in samples

Prompt that checks whether a sample excludes groups and narrows the claim. Paste it into an AI tool of your choice, which then spots selection bias in samples.

Topic: Spreadsheets & data
Type: Prompt
Level: Intermediate
Time: about 15 min

Results depend on the tool, the model and what you enter.

Result preview

Example: written by our editors, not real AI output.

Claim: 82 % satisfied Sample: users who finished survey (12 %) Missing: churned + silent (88 %) Included vs excluded: tenure shorter in sample Narrower claim: among survey finishers Limit text under chart required

Fill in

What you type stays in this browser window. We store and send nothing.

QuestionWhat do you claim?

SampleHow was it built?

MissingWho is out?

Your prompt

Help me spot selection bias in a sample.
Question: [Question]
Sample: [Sample]
Missing: [Missing]
Steps: 1. State the population the claim is about. 2. Describe who enters the sample and who is filtered out. 3. Flag self-selection, convenience samples and survivors only. 4. Compare key traits of included vs excluded where possible. 5. Propose a fairer sample or a narrower claim. 6. Write limits that must appear next to the number.
Output: bias map, comparison, narrower claim and limits text.

Take particular care to name who is missing from the sample before you trust the percentage.

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.

This is an instruction for an AI, not the result itself.

Where this prompt works

These AI tools are a good fit for this prompt:

How to use it

Fill in the fields or tap a suggestion. Leave out names, addresses and other personal data.
Copy the finished prompt into an AI tool of your choice.
Paste inclusion rules, counts and any exclusion logs, without personal raw answers.

Why does this work?

Measuring only people who finished the form ignores everyone who bounced. The rate then looks better than reality.

Tips

Write the population in one sentence above every chart that uses the sample.

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

Spot selection bias in samples

Result preview

Fill in

Your prompt

Where this prompt works

How to use it

Why does this work?

Tips

Also a good fit

Find cardinality explosions in metrics New

Estimate customer lifetime value simply New

Read a retention cohort table New