2nd Year Workshop: From idea to paper

Ian McCarthy, Emory University and NBER

Economics PhD Professionalism Workshop

From idea to “proposal”

Key steps

  1. What is the research question?
  2. What is the hook?
  3. What is your economic model or framework?
  4. What is your data and identification strategy?

1. The question

From Jesse Shapiro’s Four Steps to an Applied Micro Paper:

  • The question has an answer
  • The answer is not obvious
  • The answer is actionable/relevant

1. The question

A quick test: can you state the question in one sentence that a non-economist would understand?

  • Vague: “What is the effect of insurance on health?”
  • Better: “Does gaining Medicaid coverage reduce emergency department visits?”

If you can’t write the sentence, you don’t have a question yet. That’s OK — keep narrowing.

2. The hook

Why should anyone care? This should be aimed at a general audience, not someone specific to your field and perhaps not even to an economist.

Same question, two pitches:

  • Without a hook: “I study the effect of Medicaid on ED visits.”
  • With a hook: “Emergency departments are the most expensive way to deliver basic care, and 30 million people just gained insurance. Did that change how they use the ER?”

The hook is the reason your question matters to someone who doesn’t already study your topic.

3. The model

  • Question should be specific enough so that you are able to write down a model, an equation, or a series of equations that would facilitate answering this question
  • Not simply a mathematical representation of your research question…it should help inform the empirical methods and variation you will be using

4. Data and identification strategy

  • What data do you plan to use?
  • Where is the variation in the data that will identify the effect of interest? How does such variation lend a credible identification strategy?
  • How does your data and identification address your original question?
  • Need to know that the data exists and how to access it (within a reasonable time period/cost)

Repeat

  • Repeat this process a handful of times
  • Discuss the ideas with classmates and faculty
  • Identify your best proposals and move to the next stage

The research plan

  • Plan consists of the proposal plus…
  • The “value-added”

The value-added

  • What do we know in the literature about this problem?
  • What are the relevant papers when it comes to the topic, methods, and data?
  • How exactly do you contribute to the literature, above and beyond what was already done?

Doing the same thing for a different population or using different methods or different data is not exciting unless the hook is related to the specific population, data, method, etc.

Starting the plan

The aspirational introduction

Write the introduction to the paper you want to write, before you’ve done the analysis. This forces you to:

  • State the question clearly
  • Articulate why it matters
  • Describe what you plan to do and what you expect to find

It will be wrong. That’s the point. It gives you something concrete to work toward and revise as you learn more.

Then what?

  • Data management
  • Preliminary analysis
  • Revise that aspirational introduction as results come in
  • Draft paper
  • See 3rd year workshop materials for workflow details

Final plea

Please, please, please start thinking of research early and often. Brainstorm topics, read papers, talk to classmates and faculty. Iterate on ideas until you find something you can do. And DO NOT shy away from a challenge.