43

ECOM II

menu. See 

Export Active Grid

.

7.14

Binary Environments

While quantitative environmental variables can best be represented on a biplot as a

vector (see 

Correlation of Env. Vars

) some environmental variables such light or dark

are best coded as a 0 or 1. These are termed binary variables (see

Dummy Environmental variables

 ). Such variables can be represented on a biplot as a

vector but are best displayed as a single point   a centroid. This grid gives the co 

ordinates of the binary variables in the data set. N/A (for not applicable) is marked

beside quantitative environmental variables.

These results can be printed or exported from the File menu and copied from the Edit

menu. See 

Export Active Grid

7.15

Biplot Scores

This tab gives the co ordinates of the environmental variables which are used in the

biplots.

When a canonical Correspondence Analysis is undertaken you will be offered three

buttons above the grid that can be used to show the correlations between the

environmental variables and the site (sample) scores for each axis. These correlations

will help to identify which environmental variables are most influential for each axis.

For each axis, the environmental variable that has the greatest positive or negative

correlation will be the most important determinant. These correlations should not be

used in isolation as they can be misleading, you should also consider the size of the

Eigenvalue (see 

Variance & Eigenvalues

) for the axis to assess if the environmental

variables explain an appreciable proportion of the total variability.

Use the buttons at the top of the form to show three different parts of the output .

Copyright 2004  PISCES Conservation Ltd