Canonical Analysis Review Descriptive Statistics - Advanced Tab

Canonical Analysis

Select the Advanced tab of the Canonical Analysis Review Descriptive Statistics dialog box to access the options described here.

Means & standard deviations. Click the Means & standard deviations button to display a spreadsheet with the descriptive statistics for the variables in the analysis. If a raw data file is analyzed (i.e., raw data was specified on the Canonical Analysis Startup Panel - Quick tab) a particularly useful Graph of Input Data for that spreadsheet is the histogram with fitted normal distribution.

Correlations. Click the Correlations button to display a spreadsheet with the current correlation matrix. If a raw data file is analyzed (i.e., raw data was specified on the Canonical Analysis Startup Panel - Quick tab) a particularly useful Graph of Input Data for that spreadsheet is the scatterplot with the fitted regression line.

Covariances. Click the Covariances button to display a spreadsheet with the current covariance matrix. If a raw data file is analyzed (i.e., raw data was specified on the Canonical Analysis Startup Panel - Quick tab) a particularly useful Graph of Input Data for that spreadsheet is the scatterplot with the fitted regression line.

Box & whisker plot of vars. Click the Box & whisker plot of vars button to produce box and whisker plots for the variables in the current analysis. The box and whisker plot will summarize each variable by three components:

1. A central line to indicate central tendency or location;

2. A box to indicate variability around this central tendency;

3. Whiskers around the box to indicate the range of the variable.

After clicking the Box & whisker plot of vars button and selecting variables from the resulting Select Variables dialog box, the Box-Whisker Type dialog box will be displayed, in which you can choose to plot for each variable by selecting the appropriate option described here:

Median/Quart./Range.  Medians (central line), quartiles (box), and ranges (whiskers); note that the specific method that is used to compute these values can be configured via the Computation of percentiles box in the Options dialog box - Analyses/Graphs: Limits tab.

Means/SE/SD. Means, standard errors of the means, and standard deviations.

Means/SD/1.96*SD. Means, standard deviations, and 1.96 times the standard deviations (95% normal confidence interval for individual observations around the mean).

Means/SE/1.96*SE. Means standard errors of the means, and 1.96 times the standard errors of the means (95% normal confidence interval for means).

Matrix plot of correlations. Click the Matrix plot of correlations button to display a matrix scatterplot for the variables selected for the analysis.

SD=sqrt(Sums of Squares)/N. There are two ways in which standard deviations can be computed: as the sums of (deviation) squares divided by N (the valid number of cases) or N-1. If divided by N-1, the resulting standard deviations are estimates of the standard deviations in the population (Sigma); if divided by N, then the resulting standard deviations are descriptions of the sample only. Selecting the SD=Sums of Squares/N check box will not affect the subsequent computations for the canonical analysis.