Curriculum
Research Methods Paper Guidelines
Each HSPA student must demonstrate the ability to use appropriate doctoral-level research methods by submitting an empirical research paper. The paper must be approved by the student’s advisor and the HSPA research methods coordinator (currently Professor Will Dow). This requirement must be met before the student can take the oral qualifying examination.
There is considerable flexibility in the research paper requirement, depending on each student’s area of interest and expertise. The paper could be an independent project exploring a potential thesis topic, could build on a class term paper, could be part of a research project conducted with a faculty member or as part of a job, or anything else acceptable to the adviser and methods coordinator. There are no bounds set on types of methods or data used, however it is recommended that the student consult the advisor and methods coordinator early in the process.
The key expectations are:
- Use actual empirical data to explore an interesting question in the field of HSPA.
- Use methods appropriate for a researcher trained at the Ph.D. level.
- Write a paper of potentially publishable quality.
Typical Structure
The following is an outline of the structure of a typical methods paper, although individual papers may vary considerably from this structure. Text is 10-25 double-spaced pages, 1" margins, 10-12 point font.
1. Title page
- Title, name, date.
- Short abstract (structured or paragraph).
- In a footnote acknowledge anyone who has made important contributions to the paper.
2. Introduction (1-2 pages)
- Explain the broad question and motivate why it is interesting.
- Outline the hypotheses that you will be testing, and discuss how they will inform the broad question.
- Relate your paper to any important previous studies that you have built on (complete literature review not required).
Conceptual framework
- Provide a conceptual framework for considering your question, and state the exact hypotheses to be tests.
3. Data
- State the exact source of data. Describe the data’s population and sampling design, particularly complex survey features such as clustering and weights.
- If using a sub-sample of the data, describe the exact selection rules so that another researcher could replicate your sample in the future.
- Define each dependent and independent variable, and discuss how it is measured and transformed for analysis.
- Provide a table of summary statistics, and highlight important features of it in the text.
- Discuss missing data issues. If relevant, discuss how they will be addressed and provide a table comparing included and excluded observations (comparing samples with appropriate statistical tests).
4. Methods
- Describe the statistical methods that you will be using.
- If using advanced estimation techniques (beyond ordinary regression), motivate in words your alternative estimation approach, and state the strengths and limitations of each estimator.
- Clearly present equations illustrating the models that you will estimate (independent variables, error structure with appropriate subscripts if appropriate).
- Discuss each test that you will use to choose between models (there is no required number of tests; use the tests that are appropriate to your application).
5. Results
- Provide tables of results for each key model that you estimate (not just the preferred model), and in the text highlight the main results.
- Interpret the results of the specification tests, and argue which model is preferred and why.
6. Limitations
- Discuss each of the important limitations of your analysis, and how they might affect your results. You are not expected to have solved all potential problems, but you must thoughtfully discuss them.
7. Discussion
- Briefly summarize your main findings, and relate them to the hypotheses and broad questions in the introduction.
8. References
9. Tables and Figures
- All tables (followed by figures) should be attached to the end of the paper, not included in the middle of the text.
- Consult journals for sample formats (appropriate labels, avoid vertical lines in tables, etc.).
10. Appendix
- Include a concise program file (i.e., .do or .sas) used to generate your results. Document it with comments so that a year from now you would be able to follow it.
- If relevant, include IRB approval letter in the appendix.
Response to comments
- If this is a revised submission of the paper, include a point-by-point response to specific comments raised by the advisor or methods coordinator.