J818: Computational Approaches to Communication Research

Course Staff

Course Objectives

• •Articulate the fundamental challenges of making statistical inferences from observational data in communication research, particularly the problems of confounding and self-selection bias.
• •Understand the conceptual basis and key assumptions underlying a range of statistical methods designed for observational data, including Interrupted Time Series, Difference-in-Differences logic, Synthetic Controls, Matching, Weighting, and Machine Learning approaches (DML, Causal Forests).
• •Implement these statistical methods using the R programming language and relevant packages, applying them to communication-related datasets.
• •Critically evaluate the appropriateness of different methods for specific research questions and data structures common in communication research.
• •Design and execute appropriate robustness checks and sensitivity analyses to assess the credibility and stability of statistical findings derived from observational data.
• •Interpret and communicate the results of these analyses clearly and cautiously, acknowledging underlying assumptions and limitations, in a manner suitable for academic publication.

Course Schedule

Required Textbooks