Print

Print


***** To join INSNA, visit http://www.insna.org *****

Hi all,

 

Got it to work thanks to Martin Everett and Julie Birkholz.

It is essential to have the exact number of rows in both input files, and in the same order.

 

Best wishes,

 

Bas

 

 


Van: Social Networks Discussion Forum [[log in to unmask]] namens Reus, S.F. [[log in to unmask]]
Verzonden: woensdag 27 februari 2013 10:19
To: [log in to unmask]
Onderwerp: Re: [SOCNET] UCINET: Group Closeness Centrality

***** To join INSNA, visit http://www.insna.org *****

Hi,

 

Thanks for your reply.

Your example is working as expected.

 

I tried a similar approach with my data, so probably I'm doing something wrong somewhere.

 

Probably there is an actor in the input network that is not in the actor by group indicator matrix, because when I remove one row from your example, I don't get output.

 

Could that be the problem?

 

Kind regards,

Bas Reus

VU University Amsterdam

 

 


Van: Martin Everett [[log in to unmask]]
Verzonden: dinsdag 26 februari 2013 19:16
To: Reus, S.F.; [log in to unmask]
Onderwerp: RE: UCINET: Group Closeness Centrality

Hi

The groups can overlap so you cannot use a standard attribute file but have to use what is called an actor by group indicator matrix

 

This is described in the help file here is an example for the campnet data column3 is the women col 4 is the men columns 1 and 2 are random assignments.

 

 

               1 2 3 4

               1 2 3 4

               - - - -

    1   HOLLY  1 1 1 0

    2  BRAZEY  1 0 1 0

    3   CAROL  0 1 1 0

    4     PAM  1 0 1 0

    5     PAT  0 0 1 0

    6  JENNIE  1 0 1 0

    7 PAULINE  0 0 1 0

    8     ANN  1 0 1 0

    9 MICHAEL  0 1 0 1

   10    BILL  0 1 0 1

   11     LEE  0 0 0 1

   12     DON  1 1 0 1

   13    JOHN  0 1 0 1

   14   HARRY  0 1 0 1

   15    GERY  1 0 0 1

   16   STEVE  0 0 0 1

   17    BERT  1 1 0 1

   18    RUSS  0 0 0 1

 

 

Running this I get

 

 

Network is directed.

 

Centrality Measures

 

             1       2       3       4

       nOutDeg  nIndeg nOut2St nIn2Ste

       ------- ------- ------- -------

    1    0.700   1.000   0.667   0.833

    2    0.700   0.700   0.556   0.556

    3    0.400   0.500   0.667   0.667

    4    0.375   0.250   0.611   0.611

 

The groups are now in the rows and the centrality measures in the columns.

 

Martin

 

 

Mitchell Centre for Social Network Analysis 

School of Social Sciences

Arthur Lewis Building

University of Manchester

Bridgeford Street

Manchester M13 9PL

 

tel +44 (0)161 275 2515

 

 

From: Social Networks Discussion Forum [mailto:[log in to unmask]] On Behalf Of Reus, S.F.
Sent: 22 February 2013 21:18
To: [log in to unmask]
Subject: [SOCNET] UCINET: Group Closeness Centrality

 

***** To join INSNA, visit http://www.insna.org *****

Dear UCINET users,

 

First of all, UCINET is very useful for me working with network data.

Now I'm trying to work out how to use the Group Closeness Centrality measures for my Sunbelt paper.

 

Somehow it seems I can't get the input data in the right format. I'm not sure whether there is a problem in the input network or input group file. Also I don't get any output using the Freeman EIES dataset.

The log file doesn't give me any directions and couldn't find help in the Hanneman tutorial.

 

Any help is appreciated working with NETWORKS>CENTRALITY>GROUP CENTRALITY.

Can someone help me out?

 

Many thanks in advance,

 

Bas Reus

VU University Amsterdam

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

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