Research Ideation

Generate structured research questions, testable hypotheses, and empirical strategies from a topic, phenomenon, or dataset.

Input: $ARGUMENTS — a topic (e.g., "AI-assisted decision-making in organisations"), a phenomenon (e.g., "why do managers resist algorithmic recommendations?"), or a dataset description (e.g., "panel of UK firms with AI adoption and productivity, 2018–2024").

---

Before Starting

1. Read .context/profile.md to understand the researcher's areas and strengths.
2. Read .context/projects/_index.md to check for overlap with active projects.
3. If the topic relates to a specific project, read its context file.
4. Check existing bibliography files (.bib) in active projects for relevant references.

---

Steps

1. Understand the input. Read $ARGUMENTS and any referenced files. Identify the domain: human-AI collaboration, MCDM, multi-agent systems, organisational behaviour, or other.

2. Generate 3–5 research questions ordered from descriptive to causal:

   • Descriptive: What are the patterns? (e.g., "How has X evolved over time?")
   • Correlational: What factors are associated? (e.g., "Is X correlated with Y after controlling for Z?")
   • Causal: What is the effect? (e.g., "What is the causal effect of X on Y?")
   • Mechanism: Why does the effect exist? (e.g., "Through what channel does X affect Y?")
   • Policy/Design: What are the implications? (e.g., "How should system X be designed to improve outcome Y?")

3. For each research question, develop:

   • Hypothesis: A testable prediction with expected sign/magnitude
   • Identification/analytical strategy: How to establish causality or build the argument (DiD, experiment, simulation, design science, RDD, IV, etc.)
   • Data requirements: What data would be needed? Is it available?
   • Key assumptions: What must hold for the strategy to be valid?
   • Potential pitfalls: Common threats and mitigations
   • Related literature: 2–3 papers using similar approaches (verify these exist)

4. Rank the questions by feasibility and contribution.

5. Present the output to the user. Save only if requested.

---

Output Format

# Research Ideation: [Topic]

**Date:** [YYYY-MM-DD]
**Input:** [Original input]

## Overview

[1–2 paragraphs situating the topic and why it matters]

## Research Questions

### RQ1: [Question] (Feasibility: High/Medium/Low)

**Type:** Descriptive / Correlational / Causal / Mechanism / Policy

**Hypothesis:** [Testable prediction]

**Identification Strategy:**
- **Method:** [e.g., Difference-in-Differences, online experiment, agent-based simulation]
- **Treatment:** [What varies and when]
- **Control group:** [Comparison units or baseline]
- **Key assumption:** [e.g., parallel trends, SUTVA, rationality]

**Data Requirements:**
- [Dataset 1 — what it provides]
- [Dataset 2 — what it provides]

**Potential Pitfalls:**
1. [Threat 1 and possible mitigation]
2. [Threat 2 and possible mitigation]

**Related Work:** [Author (Year)], [Author (Year)]

---

[Repeat for RQ2–RQ5]

## Ranking

| RQ | Feasibility | Contribution | Priority |
|----|-------------|-------------|----------|
| 1  | High        | Medium      | ...      |
| 2  | Medium      | High        | ...      |

## Suggested Next Steps

1. [Most promising direction and immediate action]
2. [Data to obtain or experiment to design]
3. [Literature to review deeper]

---

Domain Adaptation

Adapt the identification strategies to the research area:

Domain	Typical strategies
Human-AI collaboration	Online experiments, field experiments, survey experiments, computational modelling
MCDM	Simulation, axiomatic analysis, case studies, experimental validation
Multi-agent systems	Agent-based models, game-theoretic analysis, computational experiments
Organisational behaviour	Field experiments, quasi-experiments, longitudinal surveys, qualitative
Environmental/carbon	DiD, RDD, IV, synthetic control, structural estimation

---

Principles

• Be creative but grounded. Push beyond obvious questions, but every suggestion must be empirically feasible.
• Think like a referee. For each causal question, immediately identify the identification challenge.
• Consider data availability. A brilliant question with no available data is not actionable.
• Suggest specific datasets where possible (WRDS, UK Data Service, FRED, Prolific, MTurk, OpenAlex, etc.).
• Use British English throughout.

---

Cross-References

Skill	When to use instead/alongside
$interview-me	When you have a specific idea and want to develop it through conversation
$devils-advocate	To stress-test a chosen research question
$literature	To verify and expand the related work for a chosen RQ
$literature	To find papers using similar methods or on similar topics (includes OpenAlex API)