The following links lead to different types of decision trees to assist in selecting statistical tests/procedures depending on different perspectives.
1. THE DECISION TREE FOR STATISTICS
http://www.microsiris.com/Statistical%20Decision%20Tree/ - Starting Point: How many variables does the problem have?
- Based on "A Guide for Selecting Statistical Techniques for Analyzing Social Science Data" from the University of Michigan, this takes you through a series of yes/no questions to an appropriate statistical analysis.
- Please read the “cautionary comments” before using.
- The tab for “statistics programs” contains a table of statistics, along with what module in SAS, SPSS, and MicrOsiris produces it. MicrOsiris is a free Statistical and Data Management Software (see link in Decision Tree page for downloading program).
2. Links to two single-page decision trees are provide by Wikiversity
http://en.wikiversity.org/wiki/Inferential_statistics_decision-making_treeThese are based on two different
starting points:- The type of dependent variable (version 1 by Neill)
- The purpose of the statistical analysis (version 2 by Corston & Coleman)
3. Decision tree for tests of association (correlational measures) by Bill Frakes (Virginia Tech).
http://rmc.ncr.vt.edu/wp-content/uploads/file/statmethods.pdf - Starting point: How many variables are continuous?
- Single page tree with hot links to web pages or articles with information about many of the correlational measures.
4. Statistic Decision Aids by Alexis Grosofsky (Office of Teaching Resources in Psychology, Beloit College)
http://teachpsych.org/otrp/resources/grosofsky09.pdf- Introductory charts for selecting measures of central tendency, variability, and relationship as well as types of visual displays based on the scale of measurement.
- Inferential statistics decision tree for univariate techniques: (1) Nominal techniques (nonparametric), (2) Ordinal techniques (nonparametric), (3) Interval/ratio techniques (parametric).
