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
- /
- Reason for Higher Correlation weight Being Non-Sig...

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

01-20-2012 10:40 AM

Hi everyone,

I ran a multilevel model using proc mixed. I obtained the following results. I ran a single regression using two scales as predictors on a single measure I obtained the following:

Scale 1: r = -.32, p = .136

Scale 2: r = .29, p = .014

I'm a bit confused by how this might happen (since the absolute magnitude of the first scale is greater than the second scale). Specifically because their is no difference in missing data. I assume that there is some inflation of the standard error of the first scale, but what can cause this? My first guess was that the standard error of measurement was very high from the first scale, but I wasn't sure if that would also lower the regression coefficient as well.

I would appreciate any and all help!

Thanks!

Nick