CONTRAST Subcommand in GLM

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

CONTRAST Subcommand in GLM

R C Intrieri

I am trying to do the following:

 

I have two grouping factors (Sex (1, 2) and Attachment (1, 2, 3, 4)

 

I want to test a contrast between sex on Attachment (level 2 only)

 

I have set these contrast up before using one grouping variable but seem to have problems setting it up with two grouping factors.

 

Any suggestions would be appreciated.

 

Bob I

===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: CONTRAST Subcommand in GLM

Bruce Weaver
Administrator
Hello Bob.  If you are using UNIANOVA (or GLM), you could mess around with the LMATRIX sub-command to get your contrast.  But I think it is much easier to get it via EMMEANS with COMPARE.  Here is an example using one of the data sets that comes with SPSS.

* Modify path below to indicate the folder where you
* store the sample datasets that come with SPSS.
GET FILE='C:\SPSSdata\survey_sample.sav'.

UNIANOVA age BY sex race
  /METHOD=SSTYPE(3)
  /EMMEANS=TABLES(sex)
  /EMMEANS=TABLES(race)
  /EMMEANS=TABLES(sex*race) COMPARE(sex) ADJ(LSD)
  /CRITERIA=ALPHA(.05)
  /DESIGN=sex race sex*race.

* Suppose I want the M vs F contrast for
* the Other race category only.  I can simply
* ignore the results for White and Black races.


In your case, the key EMMEANS sub-command would look like this:

  /EMMEANS=TABLES(Sex*Attachment) COMPARE(Sex) ADJ(LSD)

And you would ignore the results for all Attachment levels except 2.

HTH.


R C Intrieri wrote
I am trying to do the following:



I have two grouping factors (Sex (1, 2) and Attachment (1, 2, 3, 4)



I want to test a contrast between sex on Attachment (level 2 only)



I have set these contrast up before using one grouping variable but seem to
have problems setting it up with two grouping factors.



Any suggestions would be appreciated.



Bob I

=====================
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
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

PLEASE NOTE THE FOLLOWING: 
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/).