Mixed Model ANOVA

[Michael Kruger]

I have spent much of today trying to get a linear mixed-model repeated
measures ANOVA to run in SPSS 18. I know my data is structured properly
with a single row for each data point and the correct total number of
rows for my two within variables, 3 levels of one variable and 4 levels
of the other. But I get a processor crash and program terminating
message whenever I try to run it. Here is what the syntax looks
like that is in the output file after the crash. I don't know what the
_ASSERT line refers to. Any help with this vexing problerm would be
appreicated. I don't have a lot of hair to pull out anymore....

Michael Kruger

MIXED FQ1
  /CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1)
SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE)
PCONVERGE(0.000001, ABSOLUTE)
  /FIXED=| SSTYPE(3)
  /METHOD=REML
  /PRINT=CPS DESCRIPTIVES
  /REPEATED=Period*cycle | SUBJECT(id) COVTYPE(DIAG).
  _ASSERT(qfirst) failed in mxceff

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

[Maguin, Eugene]

Michael,

I don't normally do repeated measures analyses using mixed but I would setup
your problem like this. I am assuming you want both main and interaction
effects for period and cycle.

MIXED FQ1 by period cycle/CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1)
SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE)
PCONVERGE(0.000001, ABSOLUTE)
  /FIXED=period cysle | SSTYPE(3) /METHOD=REML
  /PRINT=CPS DESCRIPTIVES
  /REPEATED=Period cycle period*cycle | SUBJECT(id) COVTYPE(DIAG).

However, I am not confident that diag is the right covtype parameter. Others
on the list are much better with mixed than I am and you might want to wait
for their reply.

Gene Maguin



>>I have spent much of today trying to get a linear mixed-model repeated
measures ANOVA to run in SPSS 18. I know my data is structured properly
with a single row for each data point and the correct total number of
rows for my two within variables, 3 levels of one variable and 4 levels
of the other. But I get a processor crash and program terminating
message whenever I try to run it. Here is what the syntax looks
like that is in the output file after the crash. I don't know what the
_ASSERT line refers to. Any help with this vexing problerm would be
appreicated. I don't have a lot of hair to pull out anymore....

Michael Kruger

MIXED FQ1
  /CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1)
SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE)
PCONVERGE(0.000001, ABSOLUTE)
  /FIXED=| SSTYPE(3)
  /METHOD=REML
  /PRINT=CPS DESCRIPTIVES
  /REPEATED=Period*cycle | SUBJECT(id) COVTYPE(DIAG).
  _ASSERT(qfirst) failed in mxceff

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

[Ryan]

Michael,

It would really help if you provided more details regarding your
research question(s), research design, sample size, and variables.
I'll do my best with the limited information.

Am I correct in assuming that each subject is measured four times
during each period at equal intervals, and each time point during a
period represents a different cycle? Am I also correct in assuming
that each subject is measured in the same sequence or are you dealing
with a some sort of cross-over design?

Here's how I see your data set:

ID   Period    Cycle  FQ1
1         1         1      24
1         1         2      33
1         1         3      15
1         1         4      16
1         2         1      48
1         2         2      19
1         2         3      42
1         2         4      15
1         3         1      22
1         3         2      21
1         3         3      17
1         3         4      15
2         1         1      12
2         1         2      11
2         1         3        7
2         1         4        5
.
.
.

Without knowing anything about your data aside from the structure
(which may certainly be incorrect), it is very difficult for me to
make any suggestions. Having said that, one possibility might be:

MIXED FQ1 BY Period Cycle
 /FIXED= Period Cycle Period*Cycle | SSTYPE(3)
 /METHOD=REML
 /PRINT=DESCRIPTIVES SOLUTION
 /REPEATED=Cycle | SUBJECT(ID*Period) COVTYPE(AR1)
 /RANDOM = INTERCEPT | SUBJECT(ID*Period) COVTYPE(VC).

Ryan

On Fri, Aug 20, 2010 at 2:22 PM, Michael Kruger <[hidden email]> wrote:

> I have spent much of today trying to get a linear mixed-model repeated
> measures ANOVA to run in SPSS 18. I know my data is structured properly
> with a single row for each data point and the correct total number of
> rows for my two within variables, 3 levels of one variable and 4 levels
> of the other. But I get a processor crash and program terminating
> message whenever I try to run it. Here is what the syntax looks
> like that is in the output file after the crash. I don't know what the
> _ASSERT line refers to. Any help with this vexing problerm would be
> appreicated. I don't have a lot of hair to pull out anymore....
>
> Michael Kruger
>
> MIXED FQ1
>  /CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1)
> SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE)
> PCONVERGE(0.000001, ABSOLUTE)
>  /FIXED=| SSTYPE(3)
>  /METHOD=REML
>  /PRINT=CPS DESCRIPTIVES
>  /REPEATED=Period*cycle | SUBJECT(id) COVTYPE(DIAG).
>  _ASSERT(qfirst) failed in mxceff
>
> =====================
> 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

[Kelly Vander Ley]

I will be out of the office from August 23rd-August 30th, returning on the afternoon of August 31st. I will be checking e-mail on a limited basis. If you need immediate assistance please contact me at 503/539-5606. Thank you. Kelly