ANOVA mixed model

[francesca]

Hi,

I have three groups and three time points and I would therefore need to run a two way ANOVA to see how the differences between groups change over time.

My problem is that I have many missing points as not all subjects did the three visits and I can't use the regular test.

I know I should use the mixed model but I have found instructions only for the analysis on one Group. I can't find out how should I organize my groups to run a two way mixed model and I don't know if it's possible to use SPSS for that.

Can anybody help me with the organization of the data and the test?

Thanks a lot

Francesca

[Bruce Weaver]

Here's an example from textbook author David Howell's website (http://www.uvm.edu/~dhowell/StatPages/More_Stuff/Mixed-Models-Repeated/Mixed-Models-for-Repeated-Measures1.html) that is pretty similar to what you describe:

MIXED
  dv BY Group Time
   /CRITERIA = CIN(95) MXITER(100) MXSTEP(5) SCORING(1)
   SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0,
   ABSOLUTE)
   PCONVERGE(0.000001, ABSOLUTE)
   /FIXED = Group Time Group*Time | SSTYPE(3)
   /METHOD = REML
   /PRINT = DESCRIPTIVES SOLUTION
   /REPEATED = Time | SUBJECT(Subject) COVTYPE(CS)
   /EMMEANS = TABLES(Group)
   /EMMEANS = TABLES(Time)
   /EMMEANS = TABLES(Group*Time) .

IIRC, this approach assumes the data are MAR--i.e., missing at random.

HTH.

francesca wrote

Hi,

I have three groups and three time points and I would therefore need to run a two way ANOVA to see how the differences between groups change over time.

My problem is that I have many missing points as not all subjects did the three visits and I can't use the regular test.

I know I should use the mixed model but I have found instructions only for the analysis on one Group. I can't find out how should I organize my groups to run a two way mixed model and I don't know if it's possible to use SPSS for that.

Can anybody help me with the organization of the data and the test?

Thanks a lot

Francesca

[Mike]

The page that Bruce links to on Howell's website has a
minor error in the SPSS syntax that will cause it to fail.:

Howell has the following in the  /Repeated line

/Repeated = Time | SUBJECT(Subject) COVTYPE(CS)

But he should have the following instead if one uses
the names for the variables that he provides:

/Repeated = Time | SUBJECT(Subj) COVTYPE(CS)

That is, the variable name in the parentheses for
SUBJECT(...) should be the name one is using to
identify the subjects.  The syntax below has the
corrected syntax and works for me:

MIXED
  dv  BY Group Time
  /CRITERIA = CIN(95) MXITER(100) MXSTEP(5) SCORING(1)
  SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0,
    ABSOLUTE)
  PCONVERGE(0.000001, ABSOLUTE)
  /FIXED = Group Time Group*Time  | SSTYPE(3)
  /METHOD = REML
  /PRINT = DESCRIPTIVES SOLUTION
  /REPEATED = Time | SUBJECT(Subj) COVTYPE(CS)
  /EMMEANS = TABLES(Group)
  /EMMEANS = TABLES(Time)
  /EMMEANS = TABLES(Group*Time).

Modify as needed.

-Mike Palij
New York University
[hidden email]

P.S. Howell should really provide a CSV or Excel file for the
data (long and wide); he wouldn't have to warn about extra
spaces and such.

----- Original Message -----
From: "Bruce Weaver" <[hidden email]>
To: <[hidden email]>
Sent: Tuesday, June 02, 2015 6:11 PM
Subject: Re: ANOVA mixed model

> Here's an example from textbook author David Howell's website
> (http://www.uvm.edu/~dhowell/StatPages/More_Stuff/Mixed-Models-Repeated/Mixed-Models-for-Repeated-Measures1.html)
> that is pretty similar to what you describe:
>
>
> IIRC, this approach assumes the data are MAR--i.e., missing at random.
>
> HTH.
>
>
>
> francesca wrote
>> Hi,
>>
>> I have three groups and three time points and I would therefore need
>> to
>> run a two way ANOVA to see how the differences between groups change
>> over
>> time.
>>
>> My problem is that I have many missing points as not all subjects did
>> the
>> three visits and I can't use the regular test.
>>
>> I know I should use the mixed model but I have found instructions
>> only for
>> the analysis on one Group. I can't find out how should I organize my
>> groups to run a two way mixed model and I don't know if it's possible
>> to
>> use SPSS for that.
>>
>> Can anybody help me with the organization of the data and the test?
>>
>> Thanks a lot
>>
>> Francesca
>
>
>
>
>
> -----
> --
> 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.
>
> --
> View this message in context:
> http://spssx-discussion.1045642.n5.nabble.com/ANOVA-mixed-model-tp5729729p5729730.html
> Sent from the SPSSX Discussion mailing list archive at 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

=====================
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]

Well spotted, Mike!  

Mike Palij wrote

The page that Bruce links to on Howell's website has a
minor error in the SPSS syntax that will cause it to fail.:

Howell has the following in the  /Repeated line

/Repeated = Time | SUBJECT(Subject) COVTYPE(CS)

But he should have the following instead if one uses
the names for the variables that he provides:

/Repeated = Time | SUBJECT(Subj) COVTYPE(CS)

That is, the variable name in the parentheses for
SUBJECT(...) should be the name one is using to
identify the subjects.  The syntax below has the
corrected syntax and works for me:

MIXED
  dv  BY Group Time
  /CRITERIA = CIN(95) MXITER(100) MXSTEP(5) SCORING(1)
  SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0,
    ABSOLUTE)
  PCONVERGE(0.000001, ABSOLUTE)
  /FIXED = Group Time Group*Time  | SSTYPE(3)
  /METHOD = REML
  /PRINT = DESCRIPTIVES SOLUTION
  /REPEATED = Time | SUBJECT(Subj) COVTYPE(CS)
  /EMMEANS = TABLES(Group)
  /EMMEANS = TABLES(Time)
  /EMMEANS = TABLES(Group*Time).

Modify as needed.

-Mike Palij
New York University
[hidden email]

P.S. Howell should really provide a CSV or Excel file for the
data (long and wide); he wouldn't have to warn about extra
spaces and such.

----- Original Message -----
From: "Bruce Weaver" <[hidden email]>
To: <[hidden email]>
Sent: Tuesday, June 02, 2015 6:11 PM
Subject: Re: ANOVA mixed model


> Here's an example from textbook author David Howell's website
> (http://www.uvm.edu/~dhowell/StatPages/More_Stuff/Mixed-Models-Repeated/Mixed-Models-for-Repeated-Measures1.html)
> that is pretty similar to what you describe:
>
>
> IIRC, this approach assumes the data are MAR--i.e., missing at random.
>
> HTH.
>
>
>
> francesca wrote
>> Hi,
>>
>> I have three groups and three time points and I would therefore need
>> to
>> run a two way ANOVA to see how the differences between groups change
>> over
>> time.
>>
>> My problem is that I have many missing points as not all subjects did
>> the
>> three visits and I can't use the regular test.
>>
>> I know I should use the mixed model but I have found instructions
>> only for
>> the analysis on one Group. I can't find out how should I organize my
>> groups to run a two way mixed model and I don't know if it's possible
>> to
>> use SPSS for that.
>>
>> Can anybody help me with the organization of the data and the test?
>>
>> Thanks a lot
>>
>> Francesca
>
>
>
>
>
> -----
> --
> 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.
>
> --
> View this message in context:
> http://spssx-discussion.1045642.n5.nabble.com/ANOVA-mixed-model-tp5729729p5729730.html
> Sent from the SPSSX Discussion mailing list archive at 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

=====================
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