"The Seven Steps Toward a Thesis"

by Eduard Hovy
USC Information Sciences Institute

1. The idea
   • Is there a claim?
   • Is the claim clear?
   • Is the idea large or small?
   • Is the idea as large as it can be? Can you generalize it, or apply it elsewhere?
2. Motivation, Use, or Application
   • Why should we care?
   • How can the claim be used? Are there other applications?
3. Details of the Idea
   • What are the basic items/elements /representation units of the idea?
   • What are the rules or types of interrelationships between them?
   • How elaborated are these items and rules/relationships?
   • How much of the phenomena do they cover?
4. Data
   • Is there enough data in the study?
   • Is it representative? trustworthy? applicable?
5. Discovery Methodology
   • Is the method of investigation clear?
   • Is it appropriate? Does it ignore phenomena that look relevant?
   • Is it well-reasoned? no biases or mistakes?
6. History
   • Is prior work recognized? used?
7. Proof
   • Is there an evaluation?
     • If so, is it adequate? Complete enough?
       • Does it speak to the claim?
       • Does it actually prove the claim?
     • If not, why not?
       • Is there a discussion of how one might try to test or prove the claim?
       • Can one make predictions and (easily) test them?