How to use the Causal Claim Reviewer.
A causal claim requires more than two things changing together. Identify alternative explanations and what the design can actually establish.
Make the workflow fit your task.
Separate the observed association from the claimed causal effect. Identify timing, comparison groups, concurrent changes and plausible alternative explanations. Describe what the design supports and what additional evidence would help isolate the proposed cause.
- What you provide
- Claim and study description.
- What you get
- Causation assumptions, confounders and alternative explanations.
See the input and the result.
Illustrative input and output · a teaching example, not a live WebAct run
Example input
Signups increased after a homepage edit while an advertising campaign also launched.
Completed example
Treat the increase as an observation; both changes and other factors may contribute, so the edit's causal effect remains unresolved.
Load this input into the prompt, then copy it to WebAct to try the task. Your result may differ from the illustration.
Decisions and troubleshooting.
Does an increase immediately after a website change prove the change caused it?
Timing alone is insufficient. Campaigns, seasonality, measurement changes and other factors may explain some or all of the increase.
Why can a controlled test still leave causal uncertainty?
Check assignment, adherence, missing data and whether the measured outcome matches the claim. Study design details affect interpretation.
Try it with your own source.
Replace the example with your material in the task prompt. Keep the requirements you need, then copy the task into WebAct.
Customize and copy the task ↑