LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (July 2010)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Thu, 22 Jul 2010 06:47:04 -0600
Reply-To:     Jon K Peck <peck@us.ibm.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Jon K Peck <peck@us.ibm.com>
Subject:      Re: Can OMS catch HETCOR output?
Comments: To: Ruben van den Berg <ruben_van_den_berg@hotmail.com>
In-Reply-To:  <SNT110-W3BE67A9FA13F8BAE713A0FAA20@phx.gbl>
Content-Type: multipart/alternative;

See below. Jon Peck SPSS, an IBM Company peck@us.ibm.com 312-651-3435

From: Ruben van den Berg <ruben_van_den_berg@hotmail.com> To: SPSSX-L@LISTSERV.UGA.EDU Date: 07/22/2010 05:54 AM Subject: [SPSSX-L] Can OMS catch HETCOR output? Sent by: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>

Dear all,

I was trying to create a dataset consisting of polychoric correlations with the following syntax.

DATASET DECLARE @cor.

OMS /SELECT TABLES /IF COMMANDS=['Heterogeneous Correlation'] SUBTYPES=['Correlations'] /DESTINATION FORMAT=SAV NUMBERED=TableNumber_ OUTFILE='@cor' /TAG='@cor'.

SPSSINC HETCOR v2 v3 v4 v5 v6 v7 v8 /OPTIONS ESTIMATOR=twostep STDERR=TRUE MISSING=PAIRWISE N=TRUE TYPE=TRUE.

omsend tag=['@cor'].

However, I end up with an empty dataset. My guess (and I hope I'm wrong): everything OMS can catch has been built into OMS and since HETCOR is an extension, it hasn't been built into OMS.

>>>Not so. OMS is designed to catch everything specified by the selectors. You just need to specify the right OMS command identifier and table type. (In this case you could just omit the command identifier, since the subtype alone would pick out what you want.)

You can find these items by right clicking in the outline on the item of interest. In this case you will find that the command type is 'SPSSINC HETCOR', and the table type is 'HeterogeneousCorrelations'. You don't see these items in the OMS Identifiers dialog or the OMS Control Panel, because those lists are predefined.

If so, then could Python spssaux.GetValuesFromXMLWorkspace 'grab' the correlation matrix? Or does it use OMS 'underneath the surface' in order to get such values? If so, is there any other not-extremely-unelegant way (Output Export to EXCEL?) to get my correlations out of the viewer and into a .sav file? Eventually, I want to run a PCA on them. >>> Programmability uses the OMS mechanism to capture output items. OUTPUT EXPORT provides another way to capture tables from the Viewer, but that will be more roundabout and is slower.

HTH, Jon Peck

TIA!

Ruben van den Berg Consultant Models & Methods TNS NIPO Email: ruben.van.den.berg@tns-nipo.com Mobiel: +31 6 24641435 Telefoon: +31 20 522 5738 Internet: www.tns-nipo.com

Express yourself instantly with MSN Messenger! MSN Messenger


[text/html]


Back to: Top of message | Previous page | Main SPSSX-L page