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
- /
- How do I test balance between variables?

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

01-04-2017 11:14 AM

My question is more about statistics methods.

Say, I have variables A and B to predict variable C that is a count.

A balance between A and B is supposed to be positively associated with C, which has been tested and proved.

Now, my research takes a closer look at this equation. In particular, in the context of a balance between A and B, A may have four different dimensions. These dimensions are not mutually exclusive. In other words, two or more dimensions of A may happen at the same time. I want to test how each dimension of A impacts C, and how A's dimensions interactively impact C, contigent on a balance between A and B.

It seems that I'm stuck with too many interactions. Do you wise people have better ideas on how to test this, apparently in SAS? Thank you!

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

01-05-2017 02:09 PM - edited 01-05-2017 02:09 PM

Could you explain what you mean by "balance" and "dimensions"? Are A and B discrete (categorical) variables?

In the ANOVA setting, "balance" refers to the number of observations in each level of a categorical variable. For example, if a study has 50 males and 50 females, then the SEX variable is balanced.

When you say "dimension," do you mean the number of levels of a categorical variable? Please give an example of data where "two or more dimensions of A may happen at the same time."