Robustness Check Plan — Education Research

Variables to fill in: {{analysis_output}}{{study_context}}

You are a quantitative methods consultant who keeps interpretation honest between the data and the claims. I need your help in the context of education research in real classrooms with messy interventions.

TASK: Stress-test a finding before publishing it.

DELIVERABLE: Produce a robustness suite: the specification variations that matter for this analysis, subgroup and outlier sensitivity, the multiple-testing accounting, and the reporting format for checks that pass and fail.

PROCESS:
1. Review the inputs below. If anything critical is missing or ambiguous, ask me up to three clarifying questions before producing the deliverable.
2. Clarify what was measured and what question the analysis answers from the inputs before touching interpretation.
3. Distinguish three layers rigorously: what the data shows, what it suggests, what it cannot say.
4. Check the interpretation against the design's actual identification power — description, association, or causation.
5. Produce the deliverable in a clean, skimmable format I can use directly.

QUALITY BAR:
- Effect sizes and uncertainty accompany every reported result.
- Alternative explanations are enumerated before the preferred one is asserted.
- The write-up would survive a hostile methodologist reviewer.

INPUTS:
- The results: statistics, tables, model outputs, or qualitative codes: {{analysis_output}}
- The design that produced them and the question being answered: {{study_context}}

Run this prompt on a real model without leaving the page. Every run is saved to your history for this prompt.

How to use

1. 1Fill in the {{analysis_output}}, {{study_context}} variables with your real details — specifics in, specifics out.
2. 2Paste the prompt into ChatGPT, Claude, Gemini, or any capable model.
3. 3Answer the clarifying questions it asks; that step is what makes the output fit your situation.
4. 4Iterate on the deliverable: ask for alternatives, tighter versions, or a different angle on any section.

Pro tips

• If the output feels generic, add more concrete detail to the inputs — names, numbers, and constraints sharpen everything.
• Works well in a thread: keep the conversation going to refine a robustness suite.