How to consider this dataset / which ... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Anonymous posted on Wednesday, August 10, 2005 - 12:44 pm
Hello,

My data is a longitudinal data.
I am new in Mplus. It is too difficult to understand and fix the dataset and interpretaion.

(1)Data format:

FamID Sub ID Wave1 Wave2 Wave3 Wave4 Wave5...
1 1 0 0 0 0
1 2 0 0 1 1
1 3 0 0 1 0
2 1 0 1 0 0
2 2 0 0 0 0
2 3 0 0 0 1
2 4 0 0 0 0
3 1 0 1 1 1
3 2 0 0 1 1
3 3 0 0 0 1
3 4 0 0 0 1
4 1 0 0 1 1
...............................

(2)Data describe:

(a)Family ID 1 have 3 Siblings.
(b)Each siblings have response.
About disorde present/absent in 4times.
(c)family and sub Siblings have a strong
correlate and sub ID and each wave have a
strong correlate.

(3)The aim:

To consider they have a trend between Y
(Disorder present or absent) and waves.

Now, I consider two kinds of model. Like a Growth Curve model or Multilevel model. But not sure.
Yes! I look the Ex9.13 (Version 3.12)But my dataset and this example dataset is different.

Hopefully you can understand this thing.
Could you give me your clear and good advice?

* One more Question.

I am really really new person in Mplus.
So, even if I see the output, it is too difficult to me to understand. What's mean ...etcs., could you recommand something?
I tried to read the Reference papers, but they still don't have enough explain about output.
Seriously, need to help.

Thank you so much.
 bmuthen posted on Wednesday, August 10, 2005 - 1:02 pm
You are right that ex9.13 is a good example to follow. Your data structure suits this example. Your cluster variable ("CLUS = ") is family. Each cluster has 3 members (the 3 siblings). You have 5 u variables (5 time points). You have no x or w variables. To study this topic you can read books such as Raudenbush-Bryk or Boskers-Snijder and for Mplus-oriented material you can watch my multilevel web movie, or come to our short courses in Virginia in Nov.
 Anonymous posted on Wednesday, August 10, 2005 - 10:15 pm
Thank you so much for your answer.
I have more Questions.

(1) You said, each cluster has 3 members.
But, each cluster has the different members.
But, is it fine??

(2) Yes! I don't have x and w variables.
But, I confuse the define x and w variables.
So, could you say to me about definition?
Because, I think I wil must consider about the subject and waves relations.

(3) If I still don't have a x and w variables, it is fine?

(4) I can't fine the Raudenbush-Bryk or Boskersisnijer books. So,if it is not bother you, could you tell me the book name?

(5) If I have a time, I will go this short courses.

Thank you.
 bmuthen posted on Thursday, August 11, 2005 - 10:58 am
See my answer in another thread.
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: