SPSSX Discussion

Error 12494 when running MLmed Macro

Classic

List

Threaded

9 messages Options

Prof. Dr. Tanja Legenbauer

Error 12494 when running MLmed Macro

Dear All,

I need to perform a mediation analyses with a small set of multi-level
data. Initially, I used PROCESS and entered the level two variable as
Covariate. However, as I found a macro for a mediation analyses in
multilevel data sets, I used the macro of Rockwood & Hayes 2017 and run
the analyses with a basic model and a few modifications. The data
converged, but after displaying the indirect effect, the following error
occurs:

>Error # 12492
>An attempt has been made to use previously undefined matrix (or scalar).
>Execution of this command stops.
Matrix - 'TAB4' is undefined
>Error encountered in source line # 6142

>Error # 12332
>Undefined variable in PRINT.

I am totally lost what that means as I cannot see the original script
(this has to be started at the beginning of the SPSS session and
prepares the data for multilevel analyses, then one only has to modify
the macro according to one's needs).

Here is the macro I wrote:

* Encoding: UTF-8.
MLmed data = DataSet1
/x = Domaine5
/xW = 0
/L2cov1 = DIAGN
/m1 = EDCQ_RES
/m2 = EDCQ_EAT
/y = BINGE
/cluster = VP_NR
/folder = C:\Users\Tanja Legenbauer\Documents\Eigene
Datei\ARTIKEL\todo\FRontiers EMA special\BErechnungen
Paper\Revision_PROCESS Modell.

I would be more than grateful if anybody could be of help with this one.

Tanja

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

David Marso

Re: Error 12494 when running MLmed Macro

Administrator

Tanya,
It would be most helpful if you could attach a copy of the Rockwood & Hayes
2017 macro along with any relevant documentation.
David

Prof. Dr. Tanja Legenbauer wrote

> Dear All,
>
> I need to perform a mediation analyses with a small set of multi-level
> data. Initially, I used PROCESS and entered the level two variable as
> Covariate. However, as I found a macro for a mediation analyses in
> multilevel data sets, I used the macro of Rockwood & Hayes 2017 and run
> the analyses with a basic model and a few modifications. The data
> converged, but after displaying the indirect effect, the following error
> occurs:
>
> >Error # 12492
> >An attempt has been made to use previously undefined matrix (or scalar).
> >Execution of this command stops.
> Matrix - 'TAB4' is undefined
> >Error encountered in source line # 6142
>
> >Error # 12332
> >Undefined variable in PRINT.
>
> I am totally lost what that means as I cannot see the original script
> (this has to be started at the beginning of the SPSS session and
> prepares the data for multilevel analyses, then one only has to modify
> the macro according to one's needs).
>
> Here is the macro I wrote:
>
> * Encoding: UTF-8.
> MLmed data = DataSet1
> /x = Domaine5
> /xW = 0
> /L2cov1 = DIAGN
> /m1 = EDCQ_RES
> /m2 = EDCQ_EAT
> /y = BINGE
> /cluster = VP_NR
> /folder = C:\Users\Tanja Legenbauer\Documents\Eigene
> Datei\ARTIKEL\todo\FRontiers EMA special\BErechnungen
> Paper\Revision_PROCESS Modell.
>
> I would be more than grateful if anybody could be of help with this one.
>
> Tanja
>
> =====================
> 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

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
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

Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"

David Marso

Re: Error 12494 when running MLmed Macro

Administrator

The UGA portal does not allow attachments so use the following link.
http://spssx-discussion.1045642.n5.nabble.com/Error-12494-when-running-MLmed-Macro-td5736580.htmldoes

David Marso wrote

> Tanya,
> It would be most helpful if you could attach a copy of the Rockwood &
> Hayes
> 2017 macro along with any relevant documentation.
> David
>
>
> Prof. Dr. Tanja Legenbauer wrote
>> Dear All,
>>
>> I need to perform a mediation analyses with a small set of multi-level
>> data. Initially, I used PROCESS and entered the level two variable as
>> Covariate. However, as I found a macro for a mediation analyses in
>> multilevel data sets, I used the macro of Rockwood & Hayes 2017 and run
>> the analyses with a basic model and a few modifications. The data
>> converged, but after displaying the indirect effect, the following error
>> occurs:
>>
>> >Error # 12492
>> >An attempt has been made to use previously undefined matrix (or
>> scalar).
>> >Execution of this command stops.
>> Matrix - 'TAB4' is undefined
>> >Error encountered in source line # 6142
>>
>> >Error # 12332
>> >Undefined variable in PRINT.
>>
>> I am totally lost what that means as I cannot see the original script
>> (this has to be started at the beginning of the SPSS session and
>> prepares the data for multilevel analyses, then one only has to modify
>> the macro according to one's needs).
>>
>> Here is the macro I wrote:
>>
>> * Encoding: UTF-8.
>> MLmed data = DataSet1
>> /x = Domaine5
>> /xW = 0
>> /L2cov1 = DIAGN
>> /m1 = EDCQ_RES
>> /m2 = EDCQ_EAT
>> /y = BINGE
>> /cluster = VP_NR
>> /folder = C:\Users\Tanja Legenbauer\Documents\Eigene
>> Datei\ARTIKEL\todo\FRontiers EMA special\BErechnungen
>> Paper\Revision_PROCESS Modell.
>>
>> I would be more than grateful if anybody could be of help with this one.
>>
>> Tanja
>>
>> =====================
>> 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
>
>
>
>
>
> -----
> Please reply to the list and not to my personal email.
> Those desiring my consulting or training services please feel free to
> email me.
> ---
> "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos
> ne forte conculcent eas pedibus suis."
> Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in
> abyssum?"
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

David Marso

Re: Error 12494 when running MLmed Macro

Administrator

I located the appropriate webpage. I quote from that page.

"Please report any bugs found through email ([hidden email]) so I can
try to fix them ASAP. Include the SPSS Version number and Operating System
when reporting a bug. If an error is encountered while using the
point-and-click version, please try the same model using syntax to determine
if the issue is with the macro itself, or with the translation from the
point-and-click dialog box to the syntax. Also, do not hesitate to email for
help. An FAQ section will be constructed when time permits."

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
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

Kirill Orlov

Re: Error 12494 when running MLmed Macro

In reply to this post by Prof. Dr. Tanja Legenbauer

BTW, Tanja, are you sure the "/folder" subcommand allows not quoting the filepath, its argument?

03.09.2018 14:49, Prof. Dr. Tanja Legenbauer пишет:

Dear All,

I need to perform a mediation analyses with a small set of multi-level data. Initially, I used PROCESS and entered the level two variable as Covariate. However, as I found a macro for a mediation analyses in multilevel data sets, I used the macro of Rockwood & Hayes 2017 and run the analyses with a basic model and a few modifications. The data converged, but after displaying the indirect effect, the following error occurs:

>Error # 12492
>An attempt has been made to use previously undefined matrix (or scalar).
>Execution of this command stops.
Matrix - 'TAB4' is undefined
>Error encountered in source line # 6142

>Error # 12332
>Undefined variable in PRINT.

I am totally lost what that means as I cannot see the original script (this has to be started at the beginning of the SPSS session and prepares the data for multilevel analyses, then one only has to modify the macro according to one's needs).

Here is the macro I wrote:

* Encoding: UTF-8.
MLmed data = DataSet1
/x = Domaine5
/xW = 0
/L2cov1 = DIAGN
/m1 = EDCQ_RES
/m2 = EDCQ_EAT
/y = BINGE
/cluster = VP_NR
/folder = C:\Users\Tanja Legenbauer\Documents\Eigene Datei\ARTIKEL\todo\FRontiers EMA special\BErechnungen Paper\Revision_PROCESS Modell.

I would be more than grateful if anybody could be of help with this one.

Tanja

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

legentst-2

Re: Error 12494 when running MLmed Macro

In reply to this post by David Marso

I am not sure how do use the link. I emailed Dr. Rockwood, but he did not
reply to my request for several days.

--
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

legentst-2

Re: Error 12494 when running MLmed Macro

In reply to this post by Kirill Orlov

I think that is not the Problem as the analyses are performed. The
instruction is to open the SPSS.sav file with the data then open the macro,
then run the macro and after that, data is prepared for the multilevel
Mediation. Then one can run the modified macro which I posted in the Forum.
I am also not sure that the bug is in the MLmed Macro, it also may concern
my modification or even my SPSS data file. It would just be helpful to
undertstand what is meant be These error numbers, but Google does not help
in this case.
If anybody has an idea, it would be very helpful.
Thanks
Tanja

--
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

David Marso

Re: Error 12494 when running MLmed Macro

Administrator

Please quote whatever post to which you are replying for sake of continuity.
Most of the time error numbers don't mean squat.
Read the usually helpful accompanying text messages.
In your case it appears that TAB4 failed to be created.
TAB4 appears very few times in that huge sloppy macro and it is dependent on
a few other conditions.
Open the macro in a syntax window and search for TAB4.
Look at what it is dependent upon.
Search backwards for those dependencies.
Resolve them and see whether your macro invocation supports it.
My brief review suggests otherwise.
Check your parameters!!!!!!! Something =0 rather than 1????
Not sufficiently motivated to delve much further.

BTW: Good to use proper terminology. Hopefully you didn't modify the actual
macro. That would be truly foolhardy.
What you posted is NOT a modified macro. It is simply an invocation of the
macro. Also Kirell pointed out your filepath might be FUBAR. Try putting
that on one line and quote it. Not probably terribly relevant to the
current issue.
Perhaps create a short path for working on stuff and then move it later to
this monstrosity of a file structure.
FFS????
C:\Users\Tanja Legenbauer\Documents\Eigene Datei\ARTIKEL\todo\FRontiers EMA
special\BErechnungen Paper\Revision_PROCESS Modell.

HTH.

legentst-2 wrote

> I think that is not the Problem as the analyses are performed. The
> instruction is to open the SPSS.sav file with the data then open the
> macro,
> then run the macro and after that, data is prepared for the multilevel
> Mediation. Then one can run the modified macro which I posted in the
> Forum.
> I am also not sure that the bug is in the MLmed Macro, it also may concern
> my modification or even my SPSS data file. It would just be helpful to
> undertstand what is meant be These error numbers, but Google does not help
> in this case.
> If anybody has an idea, it would be very helpful.
> Thanks
> Tanja
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

legentst-2

Re: Error 12494 when running MLmed Macro

In reply to this post by Prof. Dr. Tanja Legenbauer

Please quote whatever post to which you are replying for sake of continuity.
Most of the time error numbers don't mean squat.
Read the usually helpful accompanying text messages.
In your case it appears that TAB4 failed to be created.
TAB4 appears very few times in that huge sloppy macro and it is dependent on
a few other conditions.
Open the macro in a syntax window and search for TAB4.
Look at what it is dependent upon.
Search backwards for those dependencies.
Resolve them and see whether your macro invocation supports it.
My brief review suggests otherwise.
Check your parameters!!!!!!! Something =0 rather than 1????
Not sufficiently motivated to delve much further.

Dear David,
I did as advised, but could not resolve the Problem. Nick Rockwood who wrote the macro answered my request inbetween and he told me that the error is a fault in his macro and he will fix it as soon as possible and Launch a new Version. He also stated that I could ignore the error at this Point. Thanks to all for their efforts!
Tánja

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