***** To join INSNA, visit http://www.insna.org *****
I guess you have your data already clustered by CONCOR, and you would like
to present the reduced blockmatrix by MDS?
When I run MDS directly in NetMiner it draws me a map with only a few visible actors. It seems like 90% of nodes are exactly in the centre and one-two are at the end of each axis. It's almost imposible to interpret such picture.† But when I check MDS as an output option in CONCOR, it shows me nice scatter plot with 3 distinctive axis and some dispersion. This picture corresponds with my substantive assumptions and very interesting for interpretation.
Anyhow, you will be much more exposed to criticism for adopting CONCOR, than
for using MDS...
Why? Unfortunatelly, I really doubt it =( Here in Russia there is only one thesis on SNA and only one book about it. And the author of both of them is my tutor. My search for SNA in local sociological journals revealed only 4 articles. So, this listserve is the only place for me to search an advice. That's why I'm asking such a noob questions.
Alexander.


2010/3/31 Balazs Vedres <[log in to unmask]>
***** †To join INSNA, visit http://www.insna.org †*****

"rarely used" - I think it is very often used.
I guess you have your data already clustered by CONCOR, and you would like
to present the reduced blockmatrix by MDS? That is fine.
You can also use MDS to represent the original structural similarity measure
matrix, and possibly include the centroids of CONCOR blocks.
Anyhow, you will be much more exposed to criticism for adopting CONCOR, than
for using MDS...

-----Original Message-----
From: Social Networks Discussion Forum [mailto:[log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of the message. ========================================================================Received: by LISTS.UFL.EDU (LISTSERV-TCP/IP release 15.5) with spool id 10875050 for [log in to unmask]; Wed, 31 Mar 2010 12:11:09 -0400 Received: from smtp.ufl.edu (smtp03.osg.ufl.edu [128.227.74.70]) by listserv-prod01.osg.ufl.edu (8.13.8/8.13.8/2.3.0) with ESMTP id o2VGB8O9027316 for <[log in to unmask]>; Wed, 31 Mar 2010 12:11:09 -0400 Received: from ch-smtp01.sth.basefarm.net (ch-smtp01.sth.basefarm.net [80.76.149.212]) by smtp.ufl.edu (8.14.0/8.14.0/3.0.0) with ESMTP id o2VGB65Z011815 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits%6 verify=NOT) for <[log in to unmask]>; Wed, 31 Mar 2010 12:11:08 -0400 Received: from c83-249-96-198.bredband.comhem.se ([83.249.96.198]:2463 helo=[192.168.0.2]) by ch-smtp01.sth.basefarm.net with esmtp (Exim 4.68) (envelope-from <[log in to unmask]>) id 1Nx0U6-0000nk-40; Wed, 31 Mar 2010 18:09:40 +0200 User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 References: <[log in to unmask]> <[log in to unmask]> <000d01cad0bb$55110140$ff3303c0$@hu> <[log in to unmask]> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: 83.249.96.198 X-Scan-Result: No virus found in message 1Nx0U6-0000nk-40. X-Scan-Signature: ch-smtp01.sth.basefarm.net 1Nx0U6-0000nk-40 a47383c1053f08f1b9eafa2ceec016ea X-Proofpoint-Virus-Version: vendor=fsecure engine=1.12.8161:2.4.5,1.2.40,4.0.166 definitions 10-03-31_15:2010-02-06,2010-03-31,2010-03-31 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policyřfault score=0 spamscore=0 ipscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=5.0.0-0908210000 definitions=main-1003310130 X-Spam-Level: * X-UFL-Spam-Level: * Message-ID: <[log in to unmask]> Date: Wed, 31 Mar 2010 18:09:37 +0200 Reply-To: Carl Nordlund <[log in to unmask]> Sender: Social Networks Discussion Forum <[log in to unmask]> From: Carl Nordlund <[log in to unmask]> Subject: Re: MDS visualization of CONCOR output Comments: To: Semenov Alexander <[log in to unmask]> In-Reply-To: <[log in to unmask]> ***** To join INSNA, visit http://www.insna.org ***** Hi, Even though the Concor algorithm can provide results that are intuitive, there has been quite a lot of critique raised against it (e.g. Sim and Schwartz 1979; Doreian 1988; Faust 1988). One of its founders has explicitly said that there is "[no] justification for advocating the iteration of [correlance coefficients] matrices as a method for analysis of data" (Schwartz 1977:266ff). Tha algorithm always produced 2 splits, and makes subsequent partitions quite arbitrary when it comes to which of these splits to split further. If there are strong arguments to use structural equivalence, perhaps it would be better to simply use the correlance matrix (distance-based or Pearson) and use MDS to plot these in two dimensions, subsequently interpreting this in combination with a dendrogram. Depending on the data, and what you are looking for, you should also look into regular equivalence (if that specific sociological notion of "roles" is applicable to your particular study). I might be wrong, but it was quite some time since I last saw a solid argument on the advantages of using the Concor algorithm. Yours, Carl Semenov Alexander wrote: > ***** To join INSNA, visit http://www.insna.org ***** > > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? > > When I run MDS directly in NetMiner it draws me a map with only a few > visible actors. It seems like 90% of nodes are exactly in the centre > and one-two are at the end of each axis. It's almost imposible to > interpret such picture. But when I check MDS as an output option in > CONCOR, it shows me nice scatter plot with 3 distinctive axis and some > dispersion. This picture corresponds with my substantive assumptions > and very interesting for interpretation. > > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > Why? Unfortunatelly, I really doubt it =( Here in Russia there is only > one thesis on SNA and only one book about it. And the author of both > of them is my tutor. My search for SNA in local sociological journals > revealed only 4 articles. So, this listserve is the only place for me > to search an advice. That's why I'm asking such a noob questions. > Alexander. > > > 2010/3/31 Balazs Vedres <[log in to unmask] > > > ***** To join INSNA, visit http://www.insna.org ***** > > "rarely used" - I think it is very often used. > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? That is fine. > You can also use MDS to represent the original structural > similarity measure > matrix, and possibly include the centroids of CONCOR blocks. > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > -----Original Message----- > From: Social Networks Discussion Forum [mailto: > -- Carl Nordlund, BA, PhD student carl.nordlund(at)hek.lu.se Human Ecology Division, Lund university www.hek.lu.se _____________________________________________________________________ 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. ========================================================================Received: by LISTS.UFL.EDU (LISTSERV-TCP/IP release 15.5) with spool id 10875570 for [log in to unmask]; Wed, 31 Mar 2010 12:26:41 -0400 Received: from smtp.ufl.edu (smtp04.osg.ufl.edu [128.227.74.71]) by listserv-prod01.osg.ufl.edu (8.13.8/8.13.8/2.3.0) with ESMTP id o2VGQece031119 for <[log in to unmask]>; Wed, 31 Mar 2010 12:26:41 -0400 Received: from ch-smtp01.sth.basefarm.net (ch-smtp01.sth.basefarm.net [80.76.149.212]) by smtp.ufl.edu (8.14.0/8.14.0/3.0.0) with ESMTP id o2VGQcdU013410 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits%6 verify=NOT) for <[log in to unmask]>; Wed, 31 Mar 2010 12:26:40 -0400 Received: from c83-249-96-198.bredband.comhem.se ([83.249.96.198]:2500 helo=[192.168.0.2]) by ch-smtp01.sth.basefarm.net with esmtp (Exim 4.68) (envelope-from <[log in to unmask]>) id 1Nx0kK-0008JL-3m; Wed, 31 Mar 2010 18:26:27 +0200 User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 References: <[log in to unmask]> <[log in to unmask]> <000d01cad0bb$55110140$ff3303c0$@hu> <[log in to unmask]> <[log in to unmask]> <[log in to unmask]> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: 83.249.96.198 X-Scan-Result: No virus found in message 1Nx0kK-0008JL-3m. X-Scan-Signature: ch-smtp01.sth.basefarm.net 1Nx0kK-0008JL-3m 6a2664c718585a11b9128aeda2dc183b X-Proofpoint-Virus-Version: vendor=fsecure engine=1.12.8161:2.4.5,1.2.40,4.0.166 definitions 10-03-31_15:2010-02-06,2010-03-31,2010-03-31 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policyřfault score=0 spamscore=0 ipscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=5.0.0-0908210000 definitions=main-1003310134 X-Spam-Level: * X-UFL-Spam-Level: * Message-ID: <[log in to unmask]> Date: Wed, 31 Mar 2010 18:26:23 +0200 Reply-To: Carl Nordlund <[log in to unmask]> Sender: Social Networks Discussion Forum <[log in to unmask]> From: Carl Nordlund <[log in to unmask]> Subject: Re: MDS visualization of CONCOR output Comments: To: Semenov Alexander <[log in to unmask]> In-Reply-To: <[log in to unmask]> ***** To join INSNA, visit http://www.insna.org ***** References: Sim, F. M. and Schwartz, M. R. 1979. /Does Concor Find Positions?/ Paper presented at the Annual Meeting of the Eastern Sociological Society, New York. Doreian, P. 1988. Using Multiple Network Analytic Tools for a Single Social Network. /Social Networks/, 10, pp. 287-312. Faust, K. 1988. Comparison of Methods for Positional Analysis: Structural and General Equivalences. /Social Networks/, 10(4), pp. 313-341. Schwartz, J. E. 1977. An Examination of CONCOR and Related Methods for Blocking Sociometric Data. /Sociological Methodology/, 8, pp. 255-282. Semenov Alexander wrote: > Thanks! > I have one more question - can you give me full citations of the > mentioned articles? I know only Doreian's article and can't identify > others. > Alexander. > > 2010/3/31 Carl Nordlund <[log in to unmask] > > > > Hi, > Even though the Concor algorithm can provide results that are > intuitive, there has been quite a lot of critique raised against > it (e.g. Sim and Schwartz 1979; Doreian 1988; Faust 1988). One of > its founders has explicitly said that there is "[no] justification > for advocating the iteration of [correlance coefficients] matrices > as a method for analysis of data" (Schwartz 1977:266ff). Tha > algorithm always produced 2 splits, and makes subsequent > partitions quite arbitrary when it comes to which of these splits > to split further. > > If there are strong arguments to use structural equivalence, > perhaps it would be better to simply use the correlance matrix > (distance-based or Pearson) and use MDS to plot these in two > dimensions, subsequently interpreting this in combination with a > dendrogram. Depending on the data, and what you are looking for, > you should also look into regular equivalence (if that specific > sociological notion of "roles" is applicable to your particular > study). > > I might be wrong, but it was quite some time since I last saw a > solid argument on the advantages of using the Concor algorithm. > > Yours, > Carl > > Semenov Alexander wrote: > > ***** To join INSNA, visit http://www.insna.org ***** > > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? > > When I run MDS directly in NetMiner it draws me a map with > only a few visible actors. It seems like 90% of nodes are > exactly in the centre and one-two are at the end of each axis. > It's almost imposible to interpret such picture. But when I > check MDS as an output option in CONCOR, it shows me nice > scatter plot with 3 distinctive axis and some dispersion. This > picture corresponds with my substantive assumptions and very > interesting for interpretation. > > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > Why? Unfortunatelly, I really doubt it =( Here in Russia there > is only one thesis on SNA and only one book about it. And the > author of both of them is my tutor. My search for SNA in local > sociological journals revealed only 4 articles. So, this > listserve is the only place for me to search an advice. That's > why I'm asking such a noob questions. > Alexander. > > > 2010/3/31 Balazs Vedres <[log in to unmask] > >> > > > ***** To join INSNA, visit http://www.insna.org ***** > > "rarely used" - I think it is very often used. > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? That is fine. > You can also use MDS to represent the original structural > similarity measure > matrix, and possibly include the centroids of CONCOR blocks. > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > -----Original Message----- > From: Social Networks Discussion Forum [mailto: > > > > -- > Carl Nordlund, BA, PhD student > carl.nordlund(at)hek.lu.se > Human Ecology Division, Lund university > www.hek.lu.se > > > > > -- > Alexander Semenov. > MA student > Faculty of Sociology > Moscow School of Social and Economic Sciences (MSSES) > http://www.msses.ru/English/index.html > > Graduate Student in Sociology at > State University - Higher School of Economics > http://www.hse.ru/eng -- Carl Nordlund, BA, PhD student carl.nordlund(at)hek.lu.se Human Ecology Division, Lund university www.hek.lu.se _____________________________________________________________________ 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. ========================================================================Received: by LISTS.UFL.EDU (LISTSERV-TCP/IP release 15.5) with spool id 10878332 for [log in to unmask]; Wed, 31 Mar 2010 13:53:00 -0400 Received: from smtp.ufl.edu (smtp01.osg.ufl.edu [128.227.74.149]) by listserv-prod01.osg.ufl.edu (8.13.8/8.13.8/2.3.0) with ESMTP id o2VHqxLW017719 for <[log in to unmask]>; Wed, 31 Mar 2010 13:52:59 -0400 Received: from sys08.mail.msu.edu (sys08.mail.msu.edu [35.9.75.108]) by smtp.ufl.edu (8.14.0/8.14.0/3.0.0) with ESMTP id o2VHquXl028394 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits%6 verify=NOT) for <[log in to unmask]>; Wed, 31 Mar 2010 13:52:57 -0400 Received: from 76-242-144-11.lightspeed.lnngmi.sbcglobal.net ([76.242.144.11] heloś500TAG1047) by sys08.mail.msu.edu with esmtpsa (Exim 4.69 #1) (TLSv1:RC4-MD5:128) id 1Nx263-0007sD-I3; Wed, 31 Mar 2010 13:52:56 -0400 References: <[log in to unmask]> <[log in to unmask]> <000d01cad0bb$55110140$ff3303c0$@hu> <[log in to unmask]> <[log in to unmask]> <[log in to unmask]> <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_001C_01CAD0D9.78996580" X-Mailer: Microsoft Office Outlook 11 thread-index: AcrQ71SvrZHTLGgbThKkKqK1I3z6dAAC1LoA X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.5579 X-Virus: None found by Clam AV X-Proofpoint-Virus-Version: vendor=fsecure engine=1.12.8161:2.4.5,1.2.40,4.0.166 definitions 10-03-31_15:2010-02-06,2010-03-31,2010-03-31 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policyřfault score=0 spamscore=0 ipscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=5.0.0-0908210000 definitions=main-1003310159 X-Spam-Level: * X-UFL-Spam-Level: * Message-ID: Date: Wed, 31 Mar 2010 13:52:57 -0400 Reply-To: Ken Frank <[log in to unmask]> Sender: Social Networks Discussion Forum <[log in to unmask]> From: Ken Frank <[log in to unmask]> Subject: Re: MDS visualization of CONCOR output Comments: To: Carl Nordlund <[log in to unmask]> In-Reply-To: <[log in to unmask]> This is a multi-part message in MIME format. ------=_NextPart_000_001C_01CAD0D9.78996580 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit ***** To join INSNA, visit http://www.insna.org ***** See the below as well. They are for cohesive subgroups, but the first piece has some for structural similarity as well Frank, K. 1996. Mapping interactions within and between cohesive subgroups. Social Networks 18: 93-119. Frank, K. A. and Yasumoto, J. 1996. "Embedding Subgroups in the Sociogram: Linking Theory and Image". Connections 19 (1): 43-57 . Frank, K.A. & Yasumoto, J. (1998). "Linking Action to Social Structure within a System: Social Capital Within and Between Subgroups." American Journal of Sociology, Volume 104, No 3, pages 642-686 Krause, A., Frank, K.A., Mason, D.M., Ulanowicz, R.E. and Taylor, W.M. (2003). "Compartments exposed in food-web structure." Nature 426:282-285 See https://www.msu.edu/~kenfrank/research.htm#representation -----Original Message----- From: Social Networks Discussion Forum [mailto:[log in to unmask]] On Behalf Of Carl Nordlund Sent: Wednesday, March 31, 2010 12:26 PM To: [log in to unmask] Subject: Re: MDS visualization of CONCOR output ***** To join INSNA, visit http://www.insna.org ***** References: Sim, F. M. and Schwartz, M. R. 1979. /Does Concor Find Positions?/ Paper presented at the Annual Meeting of the Eastern Sociological Society, New York. Doreian, P. 1988. Using Multiple Network Analytic Tools for a Single Social Network. /Social Networks/, 10, pp. 287-312. Faust, K. 1988. Comparison of Methods for Positional Analysis: Structural and General Equivalences. /Social Networks/, 10(4), pp. 313-341. Schwartz, J. E. 1977. An Examination of CONCOR and Related Methods for Blocking Sociometric Data. /Sociological Methodology/, 8, pp. 255-282. Semenov Alexander wrote: > Thanks! > I have one more question - can you give me full citations of the > mentioned articles? I know only Doreian's article and can't identify > others. > Alexander. > > 2010/3/31 Carl Nordlund <[log in to unmask] > > > > Hi, > Even though the Concor algorithm can provide results that are > intuitive, there has been quite a lot of critique raised against > it (e.g. Sim and Schwartz 1979; Doreian 1988; Faust 1988). One of > its founders has explicitly said that there is "[no] justification > for advocating the iteration of [correlance coefficients] matrices > as a method for analysis of data" (Schwartz 1977:266ff). Tha > algorithm always produced 2 splits, and makes subsequent > partitions quite arbitrary when it comes to which of these splits > to split further. > > If there are strong arguments to use structural equivalence, > perhaps it would be better to simply use the correlance matrix > (distance-based or Pearson) and use MDS to plot these in two > dimensions, subsequently interpreting this in combination with a > dendrogram. Depending on the data, and what you are looking for, > you should also look into regular equivalence (if that specific > sociological notion of "roles" is applicable to your particular > study). > > I might be wrong, but it was quite some time since I last saw a > solid argument on the advantages of using the Concor algorithm. > > Yours, > Carl > > Semenov Alexander wrote: > > ***** To join INSNA, visit http://www.insna.org ***** > > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? > > When I run MDS directly in NetMiner it draws me a map with > only a few visible actors. It seems like 90% of nodes are > exactly in the centre and one-two are at the end of each axis. > It's almost imposible to interpret such picture. But when I > check MDS as an output option in CONCOR, it shows me nice > scatter plot with 3 distinctive axis and some dispersion. This > picture corresponds with my substantive assumptions and very > interesting for interpretation. > > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > Why? Unfortunatelly, I really doubt it =( Here in Russia there > is only one thesis on SNA and only one book about it. And the > author of both of them is my tutor. My search for SNA in local > sociological journals revealed only 4 articles. So, this > listserve is the only place for me to search an advice. That's > why I'm asking such a noob questions. > Alexander. > > > 2010/3/31 Balazs Vedres <[log in to unmask] > >> > > > ***** To join INSNA, visit http://www.insna.org ***** > > "rarely used" - I think it is very often used. > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? That is fine. > You can also use MDS to represent the original structural > similarity measure > matrix, and possibly include the centroids of CONCOR blocks. > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > -----Original Message----- > From: Social Networks Discussion Forum [mailto: > > > > -- > Carl Nordlund, BA, PhD student > carl.nordlund(at)hek.lu.se > Human Ecology Division, Lund university > www.hek.lu.se > > > > > -- > Alexander Semenov. > MA student > Faculty of Sociology > Moscow School of Social and Economic Sciences (MSSES) > http://www.msses.ru/English/index.html > > Graduate Student in Sociology at > State University - Higher School of Economics > http://www.hse.ru/eng -- Carl Nordlund, BA, PhD student carl.nordlund(at)hek.lu.se Human Ecology Division, Lund university www.hek.lu.se _____________________________________________________________________ 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. ------=_NextPart_000_001C_01CAD0D9.78996580 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable ***** To join INSNA, visit http://www.insna.org *****

See the below as well.  They are for cohesive subgroups, but the first piece has some for structural similarity as well

Frank, K. 1996. Mapping interactions within and between cohesive subgroups. Social Networks 18: 93-119.

Frank, K. A. and Yasumoto, J. 1996. "Embedding Subgroups in the Sociogram: Linking Theory and Image". Connections 19 (1): 43-57 .

Frank, K.A. & Yasumoto, J. (1998). "Linking Action to Social Structure within a System: Social Capital Within and Between Subgroups." American Journal of Sociology, Volume 104, No 3, pages 642-686

Krause, A., Frank, K.A., Mason, D.M., Ulanowicz, R.E. and Taylor, W.M. (2003). "Compartments exposed in food-web structure." Nature 426:282-285

See https://www.msu.edu/~kenfrank/research.htm#representation

 

 

-----Original Message-----
From: Social Networks Discussion Forum [mailto:[log in to unmask]] On Behalf Of Carl Nordlund
Sent: Wednesday, March 31, 2010 12:26 PM
To: [log in to unmask]
Subject: Re: MDS visualization of CONCOR output

 

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

 

References:

 

 

Sim, F. M. and Schwartz, M. R. 1979. /Does Concor Find Positions?/ Paper

presented at the Annual Meeting of the Eastern Sociological Society, New

York.

 

Doreian, P. 1988. Using Multiple Network Analytic Tools for a Single

Social Network. /Social Networks/, 10, pp. 287-312.

 

Faust, K. 1988. Comparison of Methods for Positional Analysis:

Structural and General Equivalences. /Social Networks/, 10(4), pp. 313-341.

 

Schwartz, J. E. 1977. An Examination of CONCOR and Related Methods for

Blocking Sociometric Data. /Sociological Methodology/, 8, pp. 255-282.

 

 

 

 

Semenov Alexander wrote:

> Thanks!

> I have one more question - can you give me full citations of the

> mentioned articles? I know only Doreian's article and can't identify

> others.

> Alexander.

> 

> 2010/3/31 Carl Nordlund <[log in to unmask]

> <mailto:[log in to unmask]>>

> 

>     Hi,

>     Even though the Concor algorithm can provide results that are

>     intuitive, there has been quite a lot of critique raised against

>     it (e.g. Sim and Schwartz 1979; Doreian 1988; Faust 1988). One of

>     its founders has explicitly said that there is "[no] justification

>     for advocating the iteration of [correlance coefficients] matrices

>     as a method for analysis of data" (Schwartz 1977:266ff). Tha

>     algorithm always produced 2 splits, and makes subsequent

>     partitions quite arbitrary when it comes to which of these splits

>     to split further.

> 

>     If there are strong arguments to use structural equivalence,

>     perhaps it would be better to simply use the correlance matrix

>     (distance-based or Pearson) and use MDS to plot these in two

>     dimensions, subsequently interpreting this in combination with a

>     dendrogram. Depending on the data, and what you are looking for,

>     you should also look into regular equivalence (if that specific

>     sociological notion of "roles" is applicable to your particular

>     study).

> 

>     I might be wrong, but it was quite some time since I last saw a

>     solid argument on the advantages of using the Concor algorithm.

> 

>     Yours,

>     Carl

> 

>     Semenov Alexander wrote:

> 

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

> 

>            I guess you have your data already clustered by CONCOR, and you

>            would like

>            to present the reduced blockmatrix by MDS?

> 

>         When I run MDS directly in NetMiner it draws me a map with

>         only a few visible actors. It seems like 90% of nodes are

>         exactly in the centre and one-two are at the end of each axis.

>         It's almost imposible to interpret such picture.  But when I

>         check MDS as an output option in CONCOR, it shows me nice

>         scatter plot with 3 distinctive axis and some dispersion. This

>         picture corresponds with my substantive assumptions and very

>         interesting for interpretation.

> 

>            Anyhow, you will be much more exposed to criticism for adopting

>            CONCOR, than

>            for using MDS...

> 

>         Why? Unfortunatelly, I really doubt it =( Here in Russia there

>         is only one thesis on SNA and only one book about it. And the

>         author of both of them is my tutor. My search for SNA in local

>         sociological journals revealed only 4 articles. So, this

>         listserve is the only place for me to search an advice. That's

>         why I'm asking such a noob questions.

>         Alexander.

> 

> 

>         2010/3/31 Balazs Vedres <[log in to unmask]

>         <mailto:[log in to unmask]> <mailto:[log in to unmask]

>         <mailto:[log in to unmask]>>>

> 

> 

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

> 

>            "rarely used" - I think it is very often used.

>            I guess you have your data already clustered by CONCOR, and you

>            would like

>            to present the reduced blockmatrix by MDS? That is fine.

>            You can also use MDS to represent the original structural

>            similarity measure

>            matrix, and possibly include the centroids of CONCOR blocks.

>            Anyhow, you will be much more exposed to criticism for adopting

>            CONCOR, than

>            for using MDS...

> 

>            -----Original Message-----

>            From: Social Networks Discussion Forum [mailto:

> 

> 

> 

>     --

>     Carl Nordlund, BA, PhD student

>     carl.nordlund(at)hek.lu.se <http://hek.lu.se>

>     Human Ecology Division, Lund university

>     www.hek.lu.se <http://www.hek.lu.se>

> 

> 

> 

> 

> --

> Alexander Semenov.

> MA student

> Faculty of Sociology

> Moscow School of Social and Economic Sciences (MSSES)

> http://www.msses.ru/English/index.html

> 

> Graduate Student in Sociology at

> State University - Higher School of Economics

> http://www.hse.ru/eng

 

 

--

Carl Nordlund, BA, PhD student

carl.nordlund(at)hek.lu.se

Human Ecology Division, Lund university

www.hek.lu.se

 

_____________________________________________________________________

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. ------=_NextPart_000_001C_01CAD0D9.78996580-- ========================================================================Received: by LISTS.UFL.EDU (LISTSERV-TCP/IP release 15.5) with spool id 10880107 for [log in to unmask]; Wed, 31 Mar 2010 14:56:22 -0400 Received: from smtp.ufl.edu (smtp04.osg.ufl.edu [128.227.74.71]) by listserv-prod01.osg.ufl.edu (8.13.8/8.13.8/2.3.0) with ESMTP id o2VIuMtk031092 for <[log in to unmask]>; Wed, 31 Mar 2010 14:56:22 -0400 Received: from mail00a.mail.t-online.hu (mail00a.mail.t-online.hu [84.2.40.5]) by smtp.ufl.edu (8.14.0/8.14.0/3.0.0) with ESMTP id o2VIuJ29025710 for <[log in to unmask]>; Wed, 31 Mar 2010 14:56:20 -0400 Received: from Balazs (dsl5401C1F4.pool.t-online.hu [84.1.193.244]) by mail00a.mail.t-online.hu (Postfix) with ESMTPA id 36D6425C240 for <[log in to unmask]>; Wed, 31 Mar 2010 20:56:18 +0200 (CEST) References: <[log in to unmask]> <[log in to unmask]> <000d01cad0bb$55110140$ff3303c0$@hu> <[log in to unmask]> <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 12.0 thread-index: AcrQ7Nbq7lAM5hmlQ06/nKZ1bg68qAAFvsog Content-Language: en-us X-DCC-mail.t-online.hu-Metrics: mail00a.mail.t-online.hu 32710; Body=2 Fuz1=2 Fuz2=2 X-Proofpoint-Virus-Version: vendor=fsecure engine=1.12.8161:2.4.5,1.2.40,4.0.166 definitions 10-03-31_15:2010-02-06,2010-03-31,2010-03-31 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policyřfault score=0 spamscore=0 ipscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=5.0.0-0908210000 definitions=main-1003310174 X-Spam-Level: * X-UFL-Spam-Level: * Message-ID: <003101cad103$d5a4e0e0$80eea2a0$@hu> Date: Wed, 31 Mar 2010 20:56:13 +0200 Reply-To: [log in to unmask] Sender: Social Networks Discussion Forum <[log in to unmask]> From: Balazs Vedres <[log in to unmask]> Subject: Re: MDS visualization of CONCOR output In-Reply-To: <[log in to unmask]> ***** To join INSNA, visit http://www.insna.org ***** Yes, those are valid criticisms - nevertheless, CONCOR usually comes out as the best fitting clustering solution for structural equivalence (by comparing R-squared of various methods with the same data for example) -----Original Message----- From: Social Networks Discussion Forum [mailto:[log in to unmask]] On Behalf Of Carl Nordlund Sent: Wednesday, March 31, 2010 6:10 PM To: [log in to unmask] Subject: Re: MDS visualization of CONCOR output ***** To join INSNA, visit http://www.insna.org ***** Hi, Even though the Concor algorithm can provide results that are intuitive, there has been quite a lot of critique raised against it (e.g. Sim and Schwartz 1979; Doreian 1988; Faust 1988). One of its founders has explicitly said that there is "[no] justification for advocating the iteration of [correlance coefficients] matrices as a method for analysis of data" (Schwartz 1977:266ff). Tha algorithm always produced 2 splits, and makes subsequent partitions quite arbitrary when it comes to which of these splits to split further. If there are strong arguments to use structural equivalence, perhaps it would be better to simply use the correlance matrix (distance-based or Pearson) and use MDS to plot these in two dimensions, subsequently interpreting this in combination with a dendrogram. Depending on the data, and what you are looking for, you should also look into regular equivalence (if that specific sociological notion of "roles" is applicable to your particular study). I might be wrong, but it was quite some time since I last saw a solid argument on the advantages of using the Concor algorithm. Yours, Carl Semenov Alexander wrote: > ***** To join INSNA, visit http://www.insna.org ***** > > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? > > When I run MDS directly in NetMiner it draws me a map with only a few > visible actors. It seems like 90% of nodes are exactly in the centre > and one-two are at the end of each axis. It's almost imposible to > interpret such picture. But when I check MDS as an output option in > CONCOR, it shows me nice scatter plot with 3 distinctive axis and some > dispersion. This picture corresponds with my substantive assumptions > and very interesting for interpretation. > > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > Why? Unfortunatelly, I really doubt it =( Here in Russia there is only > one thesis on SNA and only one book about it. And the author of both > of them is my tutor. My search for SNA in local sociological journals > revealed only 4 articles. So, this listserve is the only place for me > to search an advice. That's why I'm asking such a noob questions. > Alexander. > > > 2010/3/31 Balazs Vedres <[log in to unmask] > > > ***** To join INSNA, visit http://www.insna.org ***** > > "rarely used" - I think it is very often used. > I guess you have your data already clustered by CONCOR, and you > would like > to present the reduced blockmatrix by MDS? That is fine. > You can also use MDS to represent the original structural > similarity measure > matrix, and possibly include the centroids of CONCOR blocks. > Anyhow, you will be much more exposed to criticism for adopting > CONCOR, than > for using MDS... > > -----Original Message----- > From: Social Networks Discussion Forum [mailto: > -- Carl Nordlund, BA, PhD student carl.nordlund(at)hek.lu.se Human Ecology Division, Lund university www.hek.lu.se _____________________________________________________________________ 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.