1.11 Workshop: Classifications = structure? (2)

  • Distinctions are only so helpful…
    • e.g., any graph can be used for exploration and explanation; Description = explanation etc.
  • What do we need in every-day work?
    • Descriptive questions:: How are observations distributed across values of Y (and X and Z)?
      • Visualize variation/distribution (Uni-dimensional or multi-dimensional)5
    • Causal questions:: Is there a causal link between the distribution across values of Y and values of D?
      • Visualize results of statistical models (t-tests, regressions etc.)
  • Structure of this workshop: See outline in syllabus.

  1. Multi-dimensional = joint distribution.