Using Mplus: the Multilevel model Que... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Anonymous posted on Friday, August 26, 2005 - 2:50 pm
Dear Muthen,

Now, I use the Mplus and I have questions.

(1) If I have missing outcomes in Dataset, the Mplus couldn't accept this points.or it's O.k?? the outcome type is a binary. It is like an Example 9.13 in User's guide.

(2) Do you have a limitation about time point?
Example, time1~time 8, so 8 times..is it ok?
Because, it is not sure Mpuls have this
limitation. Could you tell me?

Thank you so much.
 Linda K. Muthen posted on Friday, August 26, 2005 - 3:05 pm
1. If you have missing data on outcomes and do not estimate the model under TYPE=MISSING; then there will be a listwise deletion of any observation with a missing value on one or more analysis variables.

2. This should be fine.

Please send questions that are about Mplus specifically to support@statmodel.com and include your license number. Mplus Discussion is for the types of questions that are not answered as part of Mplus Support. See the following link for clarification:

http://www.statmodel.com/contract.html
 Anonymous posted on Friday, August 26, 2005 - 5:30 pm
Ok, Thank you so much.
But, I confuse Q1 answer.
Because, I try to do this input.(outcome have a Missing value)

*Input:
-------------------------------------------------
DATA: FILE IS sample1.dat;
VARIABLE: NAMES ARE FID SID u1-u5 C Sex;
USEVARIABLES ARE FID u1-u5 C;
CATEGORICAL = u1-u5;
CLUSTER = FID;
Between C;
MISSING ARE ALL (9); <- missing data(outcome)
ANALYSIS: TYPE = TWOLEVEL ;
INTEGRATION = 7;
MODEL: %WITHIN%
iw sw | u1@0 u2@1 u3@2 u4@3 u5@4;
%BETWEEN%
ib sb | u1@0 u2@1 u3@2 u4@3 u5@4;
ib sb on C;
OUTPUT: TECH1 TECH8;
-------------------------------------------------
After running, I got a result from this input. But, it is not correct input??
 Linda K. Muthen posted on Friday, August 26, 2005 - 9:22 pm
Please send your input, data, output, and license number to support@statmodel.com.
 Patchara Popaitoon posted on Monday, March 21, 2011 - 2:49 pm
Hi,

I got an error message: The labels %BETWEEN% or %WITHIN% must be specified.

I suspect that it came from the specification of latent AOC variable. I am not sure how to specify this since it is used in both within and between groups. Below is my model specification.

Within = MHRJS1-MHRJS4 AHRTR1-AHRTR4 OHRPA1-OHRPA4;
Between = Sales09;
Cluster = branch;
Centering = grandmean (AOC);

Analysis: Type = Twolevel;


Model:

AOC by AOC1-AOC6;
%within%
M by MHRJS1-MHRJS4;
A by AHRTR1-AHRTR4;
O by OHRPA1-OHRPA4;
AOC on A M O;
%between%
Sales09 on AOC;

Plesae advise. Thanks.
 Linda K. Muthen posted on Monday, March 21, 2011 - 3:45 pm
All model specifications must follow either %WITHIN% or %BETWEEN%. You need to move to AOC statement.
 Patchara Popaitoon posted on Monday, March 21, 2011 - 6:28 pm
Hi Linda,

I am aware of the issue but how could I specify this since AOC needs to be in both the within and between group. Here is what I got when I specified it in the %within% list. Please advise.

Model:
%within%
AOC by AOC1-AOC6;
M by MHRJS1-MHRJS4;
A by AHRTR1-AHRTR4;
O by OHRPA1-OHRPA4;
AOC on A M O;
%between%
Sales09 on AOC;


*** ERROR in MODEL command
Within-level variables cannot be used on the between level.
Within-level variable used: AOC
*** ERROR
The following MODEL statements are ignored:
* Statements in the BETWEEN level:
SALES09 ON AOC
 Linda K. Muthen posted on Monday, March 21, 2011 - 7:09 pm
If you want to define a factor with the same factor indicators, take the variables off of the WITHIN list. Then define the factor on both levels using a different name. See Example 9.6.
 Patchara Popaitoon posted on Tuesday, March 22, 2011 - 11:20 am
I got this error message when specifying a multilevel model:
*** ERROR
One or more between-level variables have variation within a cluster for
one or more clusters. Check your data and format statement.
Between Cluster ID with variation in this variable
Variable (only one cluster ID will be listed)

LY1 300
LY2 300
LY3 300
LY4 94
LY5 351

I am not sure what this error means. These variables were specified in the list of between group variables. They are continuous variables measured at the individual level but used at the group (branch) level. Please advise. Thanks.
 Linda K. Muthen posted on Tuesday, March 22, 2011 - 11:40 am
This means that variables you are declaring as between don't have the same value within clusters. Any variable placed on the BETWEEN list must have the same value for members of a cluster. You can use the CLUSTER_MEAN option of the DEFINE command to turn them into between-level variables.
 Patchara Popaitoon posted on Tuesday, March 22, 2011 - 12:11 pm
Thanks. In fact, I have tried aggregating these variables to the group following the user's manual guideline but it seems that the syntax was incorrect. Maybe you can help.

!define:
MLY1 = cluster_mean (LY1);
*** ERROR in VARIABLE command
Unknown option:
MLY1
 Linda K. Muthen posted on Tuesday, March 22, 2011 - 12:38 pm
You have DEFINE commented out. If you have further problems, please send your output and license number to support@statmodel.com. Mplus Discussion is not meant for these types of questions.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: