Table of Contents

Introduction

Tool Use

CADStat: Statistical Tools for Causal Analysis

Factor Analysis/Principle Components Analysis

Introduction

Factor analysis and principle components analysis provide a multivariate approaches for examining how groups of variables are correlated.

Tool Use

Select Analysis Tools -> Factor Analysis from the menus. A dialog box will open. Select the data set of interest from the pull-down menu, or browse for a tab-delimited text file (the example provided uses envdata.or.txt, found in CADStat/extdata). The Data Subsetting tab can be used to select a subset of the data file by choosing a variable from the pull down menu and then selecting the levels of that variable to include. You can hold down the <CTRL> key to add several different levels.

Select the No. of Factors you with to compute. Select 0 to compute a principle component analysis. Select one or more factors to compute a Factor Analysis.

Select all Variables that you wish to include in the analysis. You may hold down the <CTRL> key to several variables.

Further options are avalable depending on whether a principle components analysis or factor analysis is selected.

  1. Principle Components Analysis

    • Scree Plot: Output a scree plot of eigenvalues, showing the variance associated with each component. The sum of these variances is equivalent to the total variance in the selected variables.
    • Loading/Score Plot: Output a biplot of the first two principle components.
    • Rotated Variables: Return the rotated variables, appended to the original data frame.
  2. Factor Analysis

    • Varimax/Promax: Select the rotation you with to apply to your factors
    • None/Thompson/Bartlett: Select the type of scores that you would like returned. See R-help pages on factanal for more information. Scores will be returned appended to the original data frame.
graphics1