turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Performing GLM/MANOVA, not getting all values

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

03-12-2013 02:17 AM

Trying to finish my Master's thesis and having trouble with the stats. I'm researching habitat preferences of turtles. So I have a number of sites where turtles were found, then several characteristics, such as how far away railroad tracks are, wetland types, size of the wetland, etc. I ran a sample of two, then three characteristics using a GLM, but I don't seem to be getting all the values I need. I get the partial correlation coefficients, that's it. No F value, for example. Shouldn't there be more? (My advisor, who uses SAS often seems to think there should be more.) Please help?

My code looks like this:

data TurtlePopDesignation;

input TurtlePopDesignation $ Proxtorail DisttoInterstate Area @@;

datalines;

(lines of data removed)

;

proc GLM;

class TurtlePopDesignation;

model Proxtorail DisttoInterstate Area = TurtlePopDesignation / nouni;

manova / printe;

run;

And yes, I've already been told my names are too long.

The turtles and I thank you in advance for your help!

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

03-12-2013 07:24 AM

Do you need H options?

H=effects | INTERCEPT | _ALL_ specifies effects in the preceding model to use as hypothesis matrices. For each matrix (the SSCP matrix associated with an effect), the H= specification displays the characteristic roots and vectors of (where is the matrix associated with the error effect), along with the Hotelling-Lawley trace, Pillai’s trace, Wilks’ lambda, and Roy’s greatest root. By default, these statistics are tested with approximations based on the distribution. To test them with exact (but computationally intensive) calculations, use the MSTAT=EXACT option. Use the keyword INTERCEPT to produce tests for the intercept. To produce tests for all effects listed in the MODEL statement, use the keyword _ALL_ in place of a list of effects. For background and further details, see the section Multivariate Analysis of Variance.