Skip to main content

PS9594: Computational Text Analysis

Final Paper

The paper is the complete description of your project’s analysis and results. It should be 4,000 words max (single-spaced; 12pt font). The front page and references do not count toward the word limit. During Week 13, student will present the final project to the class. The final paper is due on April 16. The final paper is worth 40% of your grade.

Paper Outline

Below, I outline what you should aim to cover in each section. Note that your paper should read as a cohesive report (not as a set of disconnected answers to prompts):

  • Introduction
    • Summarize your motivation and briefly discuss prior work related to your question.
    • State your research question.
    • Provide a roadmap of the report.
  • Data and Methods
    • Where does the data come from?
    • What is the unit of observation?
    • What are the variables of interest?
    • What steps did you take to wrangle the data?
  • Analysis
    • Describe the methods/tools you explored in your project and how you implemented them.
  • Results
    • Provide a detailed summary of your results.
    • Present results clearly and concisely.
    • Use visualizations instead of tables whenever possible.
  • Discussion
    • Re-introduce your main results.
    • State your contributions.
    • Explain where you want to take this project next.

Important note (about literature reviews).
There is no standalone literature review section for this report. You should absolutely use the literature to motivate your work and situate your question, but you do not need a full section that tries to summarize an entire field. If you want examples of this style, read papers in general-interest journals: many excellent articles do not have long, self-contained literature review sections.

Submission of the final project

The end product should be a GitHub repository that contains:

  • The raw source data used for the project. If the data are too large for GitHub, talk to me and we will find a solution.
  • Your proposal.
  • A README that documents the repository. For each file, the README should clearly describe:
    • Inputs to the file (e.g., raw data; credentials needed to access an API),
    • What the file does (major transformations), and
    • Outputs produced by the file (e.g., a cleaned dataset; a figure).
  • The code files that transform the raw data into a form usable to answer your question.
  • Your final 4,000-word essay.

Of course, no commits after the deadline will be considered in the assessment.

Templates for writing

Below are a few templates you can use for writing your report. In my experience, writing data-heavy social science essays in LaTeX is a productivity boost in the long run: it makes formatting predictable, citations painless, and reproducibility easier.

If you want to experiment with LaTeX via Overleaf, you can start with the PNAS template (just remember to switch to a one-column format):

You can also use the Journal of Quantitative Description templates (Word or LaTeX):

Finally, you can use templates from Quarto and write your entire project using Quarto (or Markdown) files. This is a perfectly valid workflow for this course.