From Exploration to Inference

(Moore BPS4e p. 186)

Exploratory and Confirmatory statistics, differences and interactions

Exploratory Data Analysis Confirmatory Analysis =Statistical Inference
Purpose:  Unrestricted exploration of the data, searching for interesting patterns.
Purpose:  To answer specific questions, posed before the data were produced.
Conclusions apply:  Only to the individuals and circumstances for which we have data in hand.
Conclusions apply to a larger group of individuals (population) or a broader class of circumstances.
Conclusions:  Informal:  Based on what we see in the data.
Conclusions:  Formal: Backed by a statement of our confidence in them (confidence level, p-value)
Use Exploratory Analysis to generate hypotheses and formal questions.
Design instrument(s) to measure, design experiment(s) to test hypotheses
Use methods of data collection which guard against bias.
   Analyze using Confirmatory Analysis---
  ("Publish"; let others (or self) critique, analyze, replicate, and
      extend, possibly using Exploratory Analysis.)



Sievers home  Math151-Sp07/ExplorevsConf.htm  2pm 2/13/07
This page belongs to Sally Sievers who is solely responsible for its content. Please see our statement of responsibility.