SPSSX Discussion

Unconditional Linear Growth Model using SPSS

classic Classic list List threaded Threaded
♦
♦
 4 messages Options
sapen
Reply | Threaded
Open this post in threaded view
♦
♦
|

Unconditional Linear Growth Model using SPSS

sapen
Getting the following errors when running a null model to assessing ICC for
some DVs but not all.
For reference:
-3-time points
-8 leaders
-9 organizations

64 Mixed The final Hessian matrix is not positive definite although all
convergence criteria are satisfied. The MIXED procedure continues despite
this warning. Validity of subsequent results cannot be ascertained.

This covariance parameter is redundant. The test statistic and confidence
interval cannot be computed.

MIXED Support
  /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=ML
  /PRINT=G  SOLUTION TESTCOV
  /RANDOM=INTERCEPT | SUBJECT(Organization) COVTYPE(ID)
  /RANDOM=INTERCEPT | SUBJECT(LeaderWinOrg*Organization) COVTYPE(ID)
  /REPEATED=Time | SUBJECT(LeaderWinOrg*Organization) COVTYPE(DIAG).




--
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
Maguin, Eugene
Reply | Threaded
Open this post in threaded view
♦
♦
|

Re: Unconditional Linear Growth Model using SPSS

Maguin, Eugene
Would you describe your study. As i understand it, given what you've said, you have a two  level (organizations within leaders) model with a repeated measures factor (time). Seven of the eight of the leaders lead one organization and one leader leads two organizations. Is this accurate?
You are measuring something, let's call it Y. How are values of Y determined--that is, is Y an aggregate of the organization like gross income or is it the ratings of the organization by its employees or by an external entity?

Gene Maguin


________________________________________
From: SPSSX(r) Discussion <[hidden email]> on behalf of sapen <[hidden email]>
Sent: Sunday, June 30, 2019 2:02 PM
To: [hidden email]
Subject: Unconditional Linear Growth Model using SPSS

Getting the following errors when running a null model to assessing ICC for
some DVs but not all.
For reference:
-3-time points
-8 leaders
-9 organizations

64 Mixed The final Hessian matrix is not positive definite although all
convergence criteria are satisfied. The MIXED procedure continues despite
this warning. Validity of subsequent results cannot be ascertained.

This covariance parameter is redundant. The test statistic and confidence
interval cannot be computed.

MIXED Support
  /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=ML
  /PRINT=G  SOLUTION TESTCOV
  /RANDOM=INTERCEPT | SUBJECT(Organization) COVTYPE(ID)
  /RANDOM=INTERCEPT | SUBJECT(LeaderWinOrg*Organization) COVTYPE(ID)
  /REPEATED=Time | SUBJECT(LeaderWinOrg*Organization) COVTYPE(DIAG).




--
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

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

Re: Unconditional Linear Growth Model using SPSS

Maguin, Eugene

I added the listserv so that others can see your replies. 

Your subject line says "... liner growth model" but you have that repeated measures specification. I don't understand that.  Would you explain.


Also, 111 leaders within 9 organizations. What's the range in the number of persons rating each leader?


Gene Maguin


From: Samantha Penney <[hidden email]>
Sent: Monday, July 1, 2019 10:11 AM
To: Maguin, Eugene
Subject: Re: Unconditional Linear Growth Model using SPSS
 
Good morning, 

Thank you for inquiring. 
I apologize. That was a mistake on my part. 
There are 111 leaders across 9 organizations. 
Y (DV) are leaders individual ratings across 5 subscales. As I mentioned, I am able to run the syntax for some DVs but not others. The DVs are all very similar given that they are subscales. 

Thank you in advance. 

Samantha 


Samantha A. Penney, MSc
PhD. Industrial/Organizational Psychology Candidate
https://ca.linkedin.com/pub/samantha-penney/53/508/897



On Mon, Jul 1, 2019 at 10:06 AM Maguin, Eugene <[hidden email]> wrote:
Would you describe your study. As i understand it, given what you've said, you have a two  level (organizations within leaders) model with a repeated measures factor (time). Seven of the eight of the leaders lead one organization and one leader leads two organizations. Is this accurate?
You are measuring something, let's call it Y. How are values of Y determined--that is, is Y an aggregate of the organization like gross income or is it the ratings of the organization by its employees or by an external entity?

Gene Maguin


________________________________________
From: SPSSX(r) Discussion <[hidden email]> on behalf of sapen <[hidden email]>
Sent: Sunday, June 30, 2019 2:02 PM
To: [hidden email]
Subject: Unconditional Linear Growth Model using SPSS

Getting the following errors when running a null model to assessing ICC for
some DVs but not all.
For reference:
-3-time points
-8 leaders
-9 organizations

64 Mixed The final Hessian matrix is not positive definite although all
convergence criteria are satisfied. The MIXED procedure continues despite
this warning. Validity of subsequent results cannot be ascertained.

This covariance parameter is redundant. The test statistic and confidence
interval cannot be computed.

MIXED Support
  /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=ML
  /PRINT=G  SOLUTION TESTCOV
  /RANDOM=INTERCEPT | SUBJECT(Organization) COVTYPE(ID)
  /RANDOM=INTERCEPT | SUBJECT(LeaderWinOrg*Organization) COVTYPE(ID)
  /REPEATED=Time | SUBJECT(LeaderWinOrg*Organization) COVTYPE(DIAG).




--
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
===================== 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
Rich Ulrich
Reply | Threaded
Open this post in threaded view
♦
♦
|

Re: Unconditional Linear Growth Model using SPSS

Rich Ulrich
I think of two reasons why the errors might show up on
some DVs, various subscales, and not on others.

Maybe - The DVs have differing amounts of Missing - but you
probably would have mentioned that possibility of missing d.f.

Or - The covariance structure is a problem because of lack of
necessary variation/co-variation across particular sets of cells.
Are there scores that often are unchanged across three times?
Or the same for most people?

ML-computed solutions have potential problems with zeroes
in places that LS-solutions do not.  "Sub-scales" makes that
more likely than "scales" that total more items. 

--
Rich Ulrich

From: SPSSX(r) Discussion <[hidden email]> on behalf of Maguin, Eugene <[hidden email]>
Sent: Monday, July 1, 2019 3:33 PM
To: [hidden email]
Subject: Re: Unconditional Linear Growth Model using SPSS
 

I added the listserv so that others can see your replies. 

Your subject line says "... liner growth model" but you have that repeated measures specification. I don't understand that.  Would you explain.


Also, 111 leaders within 9 organizations. What's the range in the number of persons rating each leader?


Gene Maguin


From: Samantha Penney <[hidden email]>
Sent: Monday, July 1, 2019 10:11 AM
To: Maguin, Eugene
Subject: Re: Unconditional Linear Growth Model using SPSS
 
Good morning, 

Thank you for inquiring. 
I apologize. That was a mistake on my part. 
There are 111 leaders across 9 organizations. 
Y (DV) are leaders individual ratings across 5 subscales. As I mentioned, I am able to run the syntax for some DVs but not others. The DVs are all very similar given that they are subscales. 

Thank you in advance. 

Samantha 


Samantha A. Penney, MSc
PhD. Industrial/Organizational Psychology Candidate
https://ca.linkedin.com/pub/samantha-penney/53/508/897



On Mon, Jul 1, 2019 at 10:06 AM Maguin, Eugene <[hidden email]> wrote:
Would you describe your study. As i understand it, given what you've said, you have a two  level (organizations within leaders) model with a repeated measures factor (time). Seven of the eight of the leaders lead one organization and one leader leads two organizations. Is this accurate?
You are measuring something, let's call it Y. How are values of Y determined--that is, is Y an aggregate of the organization like gross income or is it the ratings of the organization by its employees or by an external entity?

Gene Maguin


________________________________________
From: SPSSX(r) Discussion <[hidden email]> on behalf of sapen <[hidden email]>
Sent: Sunday, June 30, 2019 2:02 PM
To: [hidden email]
Subject: Unconditional Linear Growth Model using SPSS

Getting the following errors when running a null model to assessing ICC for
some DVs but not all.
For reference:
-3-time points
-8 leaders
-9 organizations

64 Mixed The final Hessian matrix is not positive definite although all
convergence criteria are satisfied. The MIXED procedure continues despite
this warning. Validity of subsequent results cannot be ascertained.

This covariance parameter is redundant. The test statistic and confidence
interval cannot be computed.

MIXED Support
  /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=ML
  /PRINT=G  SOLUTION TESTCOV
  /RANDOM=INTERCEPT | SUBJECT(Organization) COVTYPE(ID)
  /RANDOM=INTERCEPT | SUBJECT(LeaderWinOrg*Organization) COVTYPE(ID)
  /REPEATED=Time | SUBJECT(LeaderWinOrg*Organization) COVTYPE(DIAG).




--
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
===================== 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
Free forum by Nabble Edit this page