different results with mean centered moderator in MIXED

[Zdaniuk, Bozena-3]

Hello, everyone. I have a hard time wrapping my head around that one.
I am running a MIXED analysis of 4 time points data testing potential moderation effect of a baseline moderator. So, the moderator has the same baseline values for each of the four time rows for each subject. I decided to re-run the analysis on the mean-centered moderator so I subtracted the mean from each score. My understanding from regression was that mean-centering the predictor does not change the significance levels. However, in my MIXED analysis, in the ANOVA table and in the ESTIMATES table, the significance levels for the models with uncentered and mean-centered moderator are completely different. Shouldn't they be the same? What am I not getting here?
thanks so much,
bozena

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

Hello Bozena.  It would help if you provided your code and a small
reproducible example.  But if all you've done is mean-center the variables
involved in the interaction, you should see that only the coefficients for
the first order terms that are different in the two models, whereas the
coefficient (and SE) for the interaction are the same.  Here are some notes
that might help.

https://www.theanalysisfactor.com/interpreting-lower-order-coefficients-when-the-model-contains-an-interaction/

The classic book by Aiken & West is also an excellent resource.  

  https://us.sagepub.com/en-us/nam/multiple-regression/book3045

HTH.



Zdaniuk, Bozena-3 wrote

> Hello, everyone. I have a hard time wrapping my head around that one.
> I am running a MIXED analysis of 4 time points data testing potential
> moderation effect of a baseline moderator. So, the moderator has the same
> baseline values for each of the four time rows for each subject. I decided
> to re-run the analysis on the mean-centered moderator so I subtracted the
> mean from each score. My understanding from regression was that
> mean-centering the predictor does not change the significance levels.
> However, in my MIXED analysis, in the ANOVA table and in the ESTIMATES
> table, the significance levels for the models with uncentered and
> mean-centered moderator are completely different. Shouldn't they be the
> same? What am I not getting here?
> thanks so much,
> bozena
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD