I've attached two docs.
One is a handout I used in a class to help students convert from stata
to ucinet using your nodelist approach.
The second is an ado program to create an adjacency matrix in stata,
which you could then export. There isn't a help file, but if you open
it in a text editor, you'll see some notes in the heading. It isn't the
most elegant coding, having been done quickly for a specific reason, but
it may work in your case. It is based on a data setup that is similar
to what you describe.
>>> Jenny Godley <[log in to unmask]> 04/18/02 12:25PM >>>
Dear Socnet List:
Can anyone help me get some STATA files into UCINET?
My understanding is that I write out the files (from STATA) as ASCII
files, and then include a DL header, then import them into UCINET as
I seem to be doing something wrong, though. I'm not sure at which
stage I'm making my mistakes, but hopefully someone can help! Here's
what I'm doing...
We have two types of files:
One is a file containing 6 variables: respondents id, and then five
variables with friends' ids (so the values for these variables are id
numbers). think it is possible to read this into UCINET as a 'nodelist1'
dl file with 'labels embedded', but I can't seem to get the syntax
right. Can anyone help with the DL file header for this type of file?
The second is a file containing 398 variables: respondents id, and then
397 clubs (so the values for these variables are just 0's and 1's). I
tried to read this into UCINET as a 'fullmatrix' file with 'row' and
'column labels embedded', but that syntax is not working either. Can
anyone help with the DL file header for this type of file?
If I have the DL file headers correct, there must be something else
that UCINET does not like - could it be the id numbers (3 digits)? Or
the variable names (some of the club names are 8 characters long, and
contain both letters and numbers)?
Or the number of rows or columns?
Thanks so much for your help!
- Jenny Godley
Dept. of Sociology
[log in to unmask]