correspondence analysis with multiple response set

[kathrin-2]

Hello,

I would like to do a correspondence analysis in SPSS with a multiple
response set. As you perhaps know this is not provided in SPSS. So I would
like to know if anyone knows a kind of workaround for this!
I have several variables which are coded dichotomous in my dataset. I want
to create some "bigger" variable out of these dichotomous variables and
use that kind of variables for my correspondence analysis.
Many thanks in advance for your help!

[Anthony Babinec]

In the menus, specify

Data Reduction -> Optimal Scaling

and work with All Variables Multiple Nominal
and One Set of variables.

This specifies HOMALS, also called
homogeneity analysis or multiple correspondence analysis.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
kathrin
Sent: Wednesday, January 10, 2007 3:23 AM
To: [hidden email]
Subject: correspondence analysis with multiple response set

Hello,

I would like to do a correspondence analysis in SPSS with a multiple
response set. As you perhaps know this is not provided in SPSS. So I would
like to know if anyone knows a kind of workaround for this!
I have several variables which are coded dichotomous in my dataset. I want
to create some "bigger" variable out of these dichotomous variables and
use that kind of variables for my correspondence analysis.
Many thanks in advance for your help!

[Peck, Jon]

A little clarification: HOMALS has been superseded by MULTIPLE CORRESPONDENCE in newer versions of SPSS (introduced in 14, if memory serves), and the Optimal Scaling menu will take you to that procedure now, although HOMALS is still in the system.

-Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Babinec, Tony
Sent: Wednesday, January 10, 2007 9:35 AM
To: [hidden email]
Subject: Re: [SPSSX-L] correspondence analysis with multiple response set

In the menus, specify

Data Reduction -> Optimal Scaling

and work with All Variables Multiple Nominal
and One Set of variables.

This specifies HOMALS, also called
homogeneity analysis or multiple correspondence analysis.

[Kooij, A.J. van der]

Jon is right (but MCA was introduced in 13). Some more clarification: HOMALS performs  MULTIPLE CORRESPONDENCE, the algorithm is the same in both procedures; MCA offers more analysis options and more output than HOMALS.
 
Anita van der Kooij
Data Theory Group
Leiden University
________________________________

From: SPSSX(r) Discussion on behalf of Peck, Jon
Sent: Wed 10/01/2007 16:49
To: [hidden email]
Subject: Re: correspondence analysis with multiple response set



A little clarification: HOMALS has been superseded by MULTIPLE CORRESPONDENCE in newer versions of SPSS (introduced in 14, if memory serves), and the Optimal Scaling menu will take you to that procedure now, although HOMALS is still in the system.

-Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Babinec, Tony
Sent: Wednesday, January 10, 2007 9:35 AM
To: [hidden email]
Subject: Re: [SPSSX-L] correspondence analysis with multiple response set

In the menus, specify

Data Reduction -> Optimal Scaling

and work with All Variables Multiple Nominal
and One Set of variables.

This specifies HOMALS, also called
homogeneity analysis or multiple correspondence analysis.



**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************

[Zdaniuk, Bozena]

Could someone be so kind and comment on the connection between beta
coefficient in multiple regression and the effect size. I am familiar
with using Rsq as an estimate of the effect size but I think it relates
to the effect size of the overall model. But if I want to estimate the
effect size of one particular predictor, I should use the beta, right?
But how does beta translate into the effect size? Thanks for your help
and patience in advance :)
Bozena

Bozena Zdaniuk, Ph.D.

University of Pittsburgh

UCSUR, 6th Fl.

121 University Place

Pittsburgh, PA 15260

Ph.: 412-624-5736

Fax: [phone]

email: [hidden email]

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Kooij, A.J. van der
Sent: Wednesday, January 10, 2007 12:22 PM
To: [hidden email]
Subject: Re: correspondence analysis with multiple response set

Jon is right (but MCA was introduced in 13). Some more clarification:
HOMALS performs  MULTIPLE CORRESPONDENCE, the algorithm is the same in
both procedures; MCA offers more analysis options and more output than
HOMALS.

Anita van der Kooij
Data Theory Group
Leiden University
________________________________

From: SPSSX(r) Discussion on behalf of Peck, Jon
Sent: Wed 10/01/2007 16:49
To: [hidden email]
Subject: Re: correspondence analysis with multiple response set



A little clarification: HOMALS has been superseded by MULTIPLE
CORRESPONDENCE in newer versions of SPSS (introduced in 14, if memory
serves), and the Optimal Scaling menu will take you to that procedure
now, although HOMALS is still in the system.

-Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Babinec, Tony
Sent: Wednesday, January 10, 2007 9:35 AM
To: [hidden email]
Subject: Re: [SPSSX-L] correspondence analysis with multiple response
set

In the menus, specify

Data Reduction -> Optimal Scaling

and work with All Variables Multiple Nominal
and One Set of variables.

This specifies HOMALS, also called
homogeneity analysis or multiple correspondence analysis.



**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************

[statisticsdoc]

Bozena,

By way of a general commentary, the beta weight will give you the unique
effect of the single variable, in the context of all other variables in the
model.  The meaning of beta always has to be interpreted in terms of the
unique contribution of the variable to a model containing all of the other
variables.  Sometimes investigators suggest that if the standardized beta
for X1 is twice that for X2, then X1 accounts for twice as much variance as
X2 does, but this practice has been questioned, as the values of the betas
will vary depending on what other variables are in the model and how closely
they relate to one another.

There is a general relationship between beta and the increment in r-squared.
The square of the t-value for associated with the beta weight for a given
predictor is related to the increment in r-squared see
http://www.people.vcu.edu/~nhenry/Rsq.htm ).  As a general rule, when the a
particular variable produces a significant increment in r squared when it is
added to a regression equation, then the beta for that variable is also
significant.


HTH,

Stephen Brand

For personalized and professional consultation in statistics and research
design, visit
www.statisticsdoc.com


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
Zdaniuk, Bozena
Sent: Wednesday, January 10, 2007 12:44 PM
To: [hidden email]
Subject: beta coeff and effect size


Could someone be so kind and comment on the connection between beta
coefficient in multiple regression and the effect size. I am familiar
with using Rsq as an estimate of the effect size but I think it relates
to the effect size of the overall model. But if I want to estimate the
effect size of one particular predictor, I should use the beta, right?
But how does beta translate into the effect size? Thanks for your help
and patience in advance :)
Bozena

Bozena Zdaniuk, Ph.D.

University of Pittsburgh

UCSUR, 6th Fl.

121 University Place

Pittsburgh, PA 15260

Ph.: 412-624-5736

Fax: [phone]

email: [hidden email]

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Kooij, A.J. van der
Sent: Wednesday, January 10, 2007 12:22 PM
To: [hidden email]
Subject: Re: correspondence analysis with multiple response set

Jon is right (but MCA was introduced in 13). Some more clarification:
HOMALS performs  MULTIPLE CORRESPONDENCE, the algorithm is the same in
both procedures; MCA offers more analysis options and more output than
HOMALS.

Anita van der Kooij
Data Theory Group
Leiden University
________________________________

From: SPSSX(r) Discussion on behalf of Peck, Jon
Sent: Wed 10/01/2007 16:49
To: [hidden email]
Subject: Re: correspondence analysis with multiple response set



A little clarification: HOMALS has been superseded by MULTIPLE
CORRESPONDENCE in newer versions of SPSS (introduced in 14, if memory
serves), and the Optimal Scaling menu will take you to that procedure
now, although HOMALS is still in the system.

-Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Babinec, Tony
Sent: Wednesday, January 10, 2007 9:35 AM
To: [hidden email]
Subject: Re: [SPSSX-L] correspondence analysis with multiple response
set

In the menus, specify

Data Reduction -> Optimal Scaling

and work with All Variables Multiple Nominal
and One Set of variables.

This specifies HOMALS, also called
homogeneity analysis or multiple correspondence analysis.



**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************