LISTSERV mailing list manager LISTSERV 16.0

Help for SOCNET Archives


SOCNET Archives

SOCNET Archives


SOCNET@LISTS.UFL.EDU


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Monospaced Font

LISTSERV Archives

LISTSERV Archives

SOCNET Home

SOCNET Home

SOCNET  November 2002

SOCNET November 2002

Subject:

Help w/ 2-Mode Sets

From:

Giovanni Roberto Ruffini <[log in to unmask]>

Reply-To:

[log in to unmask]

Date:

Sat, 23 Nov 2002 14:42:33 -0500

Content-Type:

TEXT/PLAIN

Parts/Attachments:

Parts/Attachments

TEXT/PLAIN (42 lines)

***** To join INSNA, visit http://www.sfu.ca/~insna/ *****

Hello -- I would much appreciate a note from someone willing to trade a
few emails offering a rudimentary tutorial on how to work with two-mode
data-sets in ucinet. I've been using DL nodelist2 to data-enter people as
they occur in texts, e.g.:

Person1 Text1, Text3, Text5
Person2 Text2, Text3
Person3 Text1, Text4

and so on. I've been told that the way to make this into a one-node list
is through the affiliations option. After a few rounds of plunging ahead
assuming I knew what this meant, I've started to doubt my results. I would
have expected this to create a matrix in which (with the above example)
the columns would be labeled Persons 1-3, as would the rows, and there
would be ones and zeros accordingly at each intersection. Rather, the
affiliations option generates in each row a pair of people, joined through
I cannot tell what criterion, and then generates columns labeled in a way
I cannot understand, and then puts ones and zeros at each intersection,
giving results which I am incapable of interpreting.

I have a data set of nearly 2000 names and the texts in which they appear
in common. One analysis I want to do is to discover which of these 2000
people are the most connected, have the lowest average distances to the
rest of the set, etc. When I thought I was doing that, I was getting
results that listed pairs instead of individuals, and I don't know what
these means. I simply cannot face the sheer horror of trying to create a
2000x2000 one-node matrix linking all the people directly. Two-mode,
linking via the texts they appear in together, seems such a wiser
approach. But I'm obviously missing the middle step, or not understanding
the significance of the affiliations option.

Any advice much appreciated,

Giovanni Ruffini

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.

Top of Message | Previous Page | Permalink

Advanced Options