Stephen
Did you do five separate analyses for each of the five historical points? Or is there a single analysis, within which you are examining order over five historical points?
If the former then you can legitimately reverse all the signs for the scores.
Perhaps I am not understanding the problem.
Richard
>
>Subject: Correspondence anaylsis inversion of ranking
> From: Stephen McTaggart <[log in to unmask]>
> Date: Thu, 4 Sep 2008 15:08:19 +1200
> To: [log in to unmask]
>
>Hi, I am using correspondence analysis to examine degrees of homogamy/social distance in society using the occupations of husbands and wives as markers of social position. I have done this over five historical points using New Zealand census data (1981-2001). I'm using the dimension scores (1 and 2) of the CA process to achieve a ranked scale of homogamy/social interaction . It is expected that the order of the ranking will be similar to that of the ranking of occupations in the issco model . This indeed is the case with three of the time periods. At two points in time however this ranking is inverted. Has anyone got tips on how to explain this/switch this around? I believe that the 'best fit model' in correspondence analysis can be a little nebulous. Greenacre talks about 'rotating the axis.' Will this work and how might I do this in SAS?
>Any help will be useful.
>Cheers, Stephen
>
>________________________________
>From: Classification, clustering, and phylogeny estimation [mailto:[log in to unmask]] On Behalf Of Liza Rovniak
>Sent: Thursday, 4 September 2008 10:40 a.m.
>To: [log in to unmask]
>Subject: cluster analysis validation technique
>
>Hi,
>
>I am hoping someone here can help me with a "how to" question on running McIntyre and Blashfield's (1980) nearest-centroid evaluation procedure to validate the stability of my cluster analysis solution. I am a newbie to cluster analysis, so this is my first time running this procedure.
>
>I have a sample of about 900 observations and have randomly split the sample in two (Sample A and Sample B). I conducted hierarchical cluster analysis and then calculated the centroid vectors for a 3-cluster solution on each of these two subsamples (i.e., steps 1 through 4 of McIntrye and Blashfield's evaluation technique).
>
>Step 5 of McIntrye and Blashfield's technique is to calculate "the squared Euclidean distance for each of Sample B's objects from each of the centroids of Sample A," and Step 6 is to assign "each object in Sample B to the closest centroid vector." At this point, I am not sure what buttons to press in SPSS to complete the analysis. One possibility I tried is to use K-means cluster analysis to achieve these two steps, but K-means uses simple Euclidean distance (not squared Euclidean distance as recommended by McIntyre and Blashfield) to assign the observations to clusters. Is this okay? (someone told me it was, but I just want to double-check). I would greatly appreciate any guidance on what buttons to press in SPSS/appropriate syntax to complete steps 5 and 6 of this analysis.
>
>Thank you.
>
>Liza Rovniak
>
>Liza S. Rovniak, PhD, MPH
>Adjunct Assistant Professor
>Center for Behavioral Epidemiology & Community Health
>Graduate School of Public Health, San Diego State University
>San Diego, CA 92123
>Phone: 858-505-4770, ext. 152; Fax: 858-505-8614
>Email: [log in to unmask]
>
>---------------------------------------------- CLASS-L list. Instructions: http://www.classification-society.org/csna/lists.html#class-l
>
>----------------------------------------------
>CLASS-L list.
>Instructions: http://www.classification-society.org/csna/lists.html#class-l
>
>--_000_6085BF643DA19E43B24174695D9A63F50BE7246C33artsmail4ARTS_
>Content-Type: text/html; charset="us-ascii"
>Content-Transfer-Encoding: quoted-printable
>
><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
><HTML xmlns=3D"http://www.w3.org/TR/REC-html40" xmlns:v =3D=20
>"urn:schemas-microsoft-com:vml" xmlns:o =3D=20
>"urn:schemas-microsoft-com:office:office" xmlns:w =3D=20
>"urn:schemas-microsoft-com:office:word" xmlns:m =3D=20
>"http://schemas.microsoft.com/office/2004/12/omml"><HEAD>
><META http-equiv=3DContent-Type content=3D"text/html; charset=3Dus-ascii">
><META content=3D"MSHTML 6.00.6000.16705" name=3DGENERATOR>
><STYLE>@font-face {
> font-family: Calibri;
>}
>@page Section1 {size: 8.5in 11.0in; margin: 1.0in 1.0in 1.0in 1.0in; }
>P.MsoNormal {
> FONT-SIZE: 11pt; MARGIN: 0in 0in 0pt; FONT-FAMILY: "Calibri","sans-serif"
>}
>LI.MsoNormal {
> FONT-SIZE: 11pt; MARGIN: 0in 0in 0pt; FONT-FAMILY: "Calibri","sans-serif"
>}
>DIV.MsoNormal {
> FONT-SIZE: 11pt; MARGIN: 0in 0in 0pt; FONT-FAMILY: "Calibri","sans-serif"
>}
>A:link {
> COLOR: blue; TEXT-DECORATION: underline; mso-style-priority: 99
>}
>SPAN.MsoHyperlink {
> COLOR: blue; TEXT-DECORATION: underline; mso-style-priority: 99
>}
>A:visited {
> COLOR: purple; TEXT-DECORATION: underline; mso-style-priority: 99
>}
>SPAN.MsoHyperlinkFollowed {
> COLOR: purple; TEXT-DECORATION: underline; mso-style-priority: 99
>}
>SPAN.EmailStyle17 {
> COLOR: windowtext; FONT-FAMILY: "Calibri","sans-serif"; mso-style-type: pe=
>rsonal-compose
>}
>..MsoChpDefault {
> mso-style-type: export-only
>}
>DIV.Section1 {
> page: Section1
>}
></STYLE>
><!--[if gte mso 9]><xml>
> <o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
></xml><![endif]--><!--[if gte mso 9]><xml>
> <o:shapelayout v:ext=3D"edit">
> <o:idmap v:ext=3D"edit" data=3D"1" />
> </o:shapelayout></xml><![endif]--></HEAD>
><BODY lang=3DEN-US vLink=3Dpurple link=3Dblue>
><DIV dir=3Dltr align=3Dleft><SPAN class=3D993351123-03092008><FONT face=3DA=
>rial=20
>color=3D#0000ff size=3D2>Hi, I am using correspondence analysis to examine =
>degrees=20
>of homogamy/social distance in society using the occupations of husba=
>nds=20
>and wives as markers of social position. I have done this over five=20
>historical points using New Zealand census data (1981-2001). I'm using the=
>=20
>dimension scores (1 and 2) of the CA process to achieve a ranked=
>=20
>scale of homogamy/social interaction . It is expected that the order of the=
>=20
> ranking will be similar to that of the ranking of occupations in the =
>issco=20
>model . This indeed is the case with three of the time periods. At two poin=
>ts in=20
>time however this ranking is inverted. Has anyone got tips on how to=20
> explain this/switch this around? I believe that the 'best fit model' =
>in=20
>correspondence analysis can be a little nebulous. Greenacre talk=
>s=20
>about 'rotating the axis.' Will this work and how might I do this in=20
>SAS?</FONT></SPAN></DIV>
><DIV dir=3Dltr align=3Dleft><SPAN class=3D993351123-03092008><FONT face=3DA=
>rial=20
>color=3D#0000ff size=3D2>Any help will be useful.</FONT></SPAN></DIV>
><DIV dir=3Dltr align=3Dleft><SPAN class=3D993351123-03092008><FONT face=3DA=
>rial=20
>color=3D#0000ff size=3D2>Cheers, Stephen</FONT> </SPAN></DIV><BR>
><DIV class=3DOutlookMessageHeader lang=3Den-us dir=3Dltr align=3Dleft>
><HR tabIndex=3D-1>
><FONT face=3DTahoma size=3D2><B>From:</B> Classification, clustering, and p=
>hylogeny=20
>estimation [mailto:[log in to unmask]] <B>On Behalf Of </B>Liza=20
>Rovniak<BR><B>Sent:</B> Thursday, 4 September 2008 10:40 a.m.<BR><B>To:</B>=
>=20
>[log in to unmask]<BR><B>Subject:</B> cluster analysis validation=20
>technique<BR></FONT><BR></DIV>
><DIV></DIV>
><DIV class=3DSection1>
><P class=3DMsoNormal>Hi,<o:p></o:p></P>
><P class=3DMsoNormal><o:p> </o:p></P>
><P class=3DMsoNormal>I am hoping someone here can help me with a “how=
> to” question=20
>on running McIntyre and Blashfield’s (1980) nearest-centroid evaluati=
>on=20
>procedure to validate the stability of my cluster analysis solution. I am a=
>=20
>newbie to cluster analysis, so this is my first time running this procedure=
>..=20
><o:p></o:p></P>
><P class=3DMsoNormal><o:p> </o:p></P>
><P class=3DMsoNormal>I have a sample of about 900 observations and ha=
>ve=20
>randomly split the sample in two (Sample A and Sample B). I conducted=20
>hierarchical cluster analysis and then calculated the centroid vectors for =
>a=20
>3-cluster solution on each of these two subsamples (i.e., steps 1 through 4=
> of=20
>McIntrye and Blashfield’s evaluation technique). <o:p></o:p></P>
><P class=3DMsoNormal><o:p> </o:p></P>
><P class=3DMsoNormal>Step 5 of McIntrye and Blashfield’s technique is=
> to calculate=20
>“the squared Euclidean distance for each of Sample B’s objects =
>from each of the=20
>centroids of Sample A,” and Step 6 is to assign “each object &n=
>bsp;in Sample B=20
>to the closest centroid vector.” At this point, I am not sure what bu=
>ttons to=20
>press in SPSS to complete the analysis. One possibility I tried is to use=20
>K-means cluster analysis to achieve these two steps, but K-means uses simpl=
>e=20
>Euclidean distance (not squared Euclidean distance as recommended by McInty=
>re=20
>and Blashfield) to assign the observations to clusters. Is this okay? (some=
>one=20
>told me it was, but I just want to double-check). I would greatly=20
>appreciate any guidance on what buttons to press in SPSS/appropriate syntax=
> to=20
>complete steps 5 and 6 of this analysis. <o:p></o:p></P>
><P class=3DMsoNormal><o:p> </o:p></P>
><P class=3DMsoNormal>Thank you.<o:p></o:p></P>
><P class=3DMsoNormal><o:p> </o:p></P>
><P class=3DMsoNormal>Liza Rovniak<o:p></o:p></P>
><P class=3DMsoNormal><o:p> </o:p></P>
><P class=3DMsoNormal>Liza S. Rovniak, PhD, MPH<o:p></o:p></P>
><P class=3DMsoNormal>Adjunct Assistant Professor<o:p></o:p></P>
><P class=3DMsoNormal>Center for Behavioral Epidemiology & Community=20
>Health<o:p></o:p></P>
><P class=3DMsoNormal>Graduate School of Public Health, San Diego State=20
>University<o:p></o:p></P>
><P class=3DMsoNormal>San Diego, CA 92123<o:p></o:p></P>
><P class=3DMsoNormal>Phone: 858-505-4770, ext. 152; Fax:=20
>858-505-8614<o:p></o:p></P>
><P class=3DMsoNormal>Email: [log in to unmask]<o:p></o:p></P>
><P=20
>class=3DMsoNormal><o:p> </o:p></P></DIV>------------------------------=
>----------------=20
>CLASS-L list. Instructions:=20
>http://www.classification-society.org/csna/lists.html#class-l </BODY></HTML=
>>
>----------------------------------------------
>CLASS-L list.
>Instructions: http://www.classification-society.org/csna/lists.html#class-l
>
>--_000_6085BF643DA19E43B24174695D9A63F50BE7246C33artsmail4ARTS_--
----------------------------------------------
CLASS-L list.
Instructions: http://www.classification-society.org/csna/lists.html#class-l