Interview Me

• "Is there a natural experiment, policy change, or source of exogenous variation you can exploit?"
• "What's the biggest threat to a causal interpretation? How would you address it?"

Phase 4: Data and Setting (1-2 questions)

• "What data do you have access to, or what would you ideally need?"
• "Is there a specific country, time period, or institutional context that makes this setting special?"

Phase 5: Expected Results and Contribution (1 question)

• "What would the results imply for policy or theory? How does this differ from what's already been done?"

After the Interview

Once sufficient information is gathered (typically 5-8 exchanges), produce a Research Specification Document and save to the session log or a /plan folder file.

# Research Specification: [Title]

**Date:** YYYY-MM-DD

## Research Question

[Clear, specific question in one sentence]

## Motivation

[2-3 paragraphs: why this matters, theoretical context, policy relevance]

## Hypothesis

[Testable prediction with expected direction and magnitude intuition]

## Empirical Strategy

- **Method:** [e.g., Difference-in-Differences, IV, RDD, Synthetic Control]
- **Treatment:** [What varies and how]
- **Comparison group:** [Control unit/period]
- **Key identifying assumption:** [What must hold for causal interpretation]
- **Main threat to validity:** [And how to address it]
- **Robustness checks:** [Pre-trends, placebo, alternative estimators]

## Data

- **Primary dataset:** [Name, source, coverage, unit of observation]
- **Key variables:** [Treatment, outcome, controls]
- **Sample:** [Time period, geography, N approximate]

## Expected Results

[What to find, what would be surprising, economic magnitude intuition]

## Contribution

[How this advances the literature — 2-3 sentences, clearly differentiated]

## Open Questions

[Issues that need further thought before beginning]

Interview Style

• Ask one or two questions at a time — never a list
• Follow up based on the answer, don't rigidly follow the script
• If identification is weak, probe gently: "What would a skeptic say about that variation?"
• Stop interviewing when you have enough for a complete spec (usually 5-8 exchanges)
• Acknowledge strong ideas explicitly — good research deserves recognition