Conducting a Literature Review - jonathancolmer/lab-guide GitHub Wiki

Introduction

A literature review is the foundation for understanding where your research fits in the broader conversation.
It is more than a search for papers — it’s a process of mapping the existing evidence, identifying what’s missing, and sharpening the questions you ask.

Done well, it:

  • Gives you a clear picture of what is known (and unknown).
  • Highlights the strengths and weaknesses of existing work.
  • Helps you avoid duplicating effort.
  • Positions you to make a meaningful contribution.

This guide outlines a practical, structured approach to conducting a literature review for research in economics and related fields. Follow it to work systematically, think critically, and produce something that is useful both to you and to the wider research team.

1. Purpose of a Literature Review

A good literature review:

  • Maps the terrain — what has been done and what hasn’t.
  • Provides context for your research question.
  • Synthesizes existing work: identify common themes, differences, and debates.
  • Critically analyzes to find research gaps.
  • Builds credibility by demonstrating subject-matter expertise.
  • Is a journey of discovery — expect to learn along the way.

2. The Process

  • Treat it as a learning opportunity:
    • Understand the project.
    • Learn relevant methods.
    • See how findings vary across contexts and countries.
    • Train your ability to see the big picture.
  • Invest in your own skills — this is human capital development.
  • Don’t assume the literature is complete:
    • Look for older or obscure studies, other disciplines, books.
    • Even experts miss things.
  • Take pride in doing it well.

3. Types of Literature Reviews

  • Narrative review — summarizing broad themes.
  • Systematic review — comprehensive and methodical search.
  • Critical review — evaluating strengths, weaknesses, and gaps.

4. Using a Matrix Structure

Organize literature by multiple attributes:

Attribute Examples
Country studied High-, middle-, low-income
Methods Descriptive, RCT, RDD, IV, Diff-in-Diff, ML
Data type Survey, admin records
Unit of analysis Individual, district, country
Outcomes Short run, long run

Tips:

  • The taxonomy will evolve — start broad, refine as you go.
  • Get feedback from the PI.
  • Use consistent and comparable language.

5. Complement With a Narrative

The matrix isn’t enough — add a narrative that explains:

  • What’s new or different in each study.
  • Relative strengths and weaknesses.
  • Which gaps the work addresses.

Be prepared: a large, detailed matrix may translate to just a few sentences in your write-up.

6. Searching for Literature

Keywords: Identify clear, relevant search terms.
Databases:

  • Economics: EconLit, JSTOR, SSRN, NBER.
  • Broader: Google Scholar, Scopus, Web of Science.

Search techniques:

  • Boolean operators: AND, OR, NOT.
  • Citation chaining (follow references forward and backward).
  • Gray literature: working papers, conference proceedings.

Tools:

  • Citation managers: Zotero, Mendeley, EndNote.

7. Why Not Use AI to Search?

  • You don’t know what database it’s pulling from.
  • Risk of phantom citations.
  • At most, use AI to complement—not replace—your search.
  • Always do a final manual check.

8. Analyzing and Organizing Sources

  • How to read a journal article: ICPSR guide
  • Efficient reading: focus on abstract, intro, conclusion; skim tables and methods.
  • Note-taking: capture the question, method, and findings using a consistent format.
  • Thematic organization: group by method, geography, time, etc.

9. Writing the Review

Structure:

  1. Introduction — scope and significance.
  2. Main body — organized thematically, methodologically, or chronologically.
  3. Conclusion — summarize gaps, set up the contribution.

Best practices:

  • Synthesize, don’t just summarize.
  • Use transitions between studies.
  • Integrate citations naturally.

Pitfalls:

  • Over-reliance on high-profile journals.
  • Ignoring foundational work.
  • Lack of critical analysis.

10. Reviewing and Refining

Checklist:

  • Is the review logically structured?
  • Are sources critically analyzed?
  • Are citations complete and consistent?

Feedback:

  • Share drafts with peers or mentors.
  • Revise based on feedback.

Final Notes

A strong literature review is more than a list of studies — it’s a map, a critique, and a bridge to your own contribution.
Take the time to make it both comprehensive and insightful.