MDS map

[Faris Siraf]

Hi all,

I am working on a study in which, each respondent places 5 objects on a 2-dimensional space. I have (x,y) of each object in my data. My goal is to find out how those 5 objects are perceived on 2-dimensional space. I was thinking of using MDS map. In past, I have done MDS map in SPSS, but I had only one similarity matrix. Now, I think I have to deal with one similarity matrix for each respondents My question is that:

1. Is there a command in SPSS to created similarity/distance matrix using coordinates of objects?
2. How can I use similarity matrix of all respondents to come up with MDS map?


Thanks in advance for your help!


Homayoun

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

[Busing, Frank]

D ear Homayoun.
 
In SPSS you can use the PROXIMITIES subcommand to create distances
between the coordinates of each respondent.
Then, you can use PROXSCAL to analyse these data with an individual
difference model.
This will give you your MDS map.
 
Good luck, Frank.
 
 
Frank Busing
Leiden University
 
 
---------------------------------------------

Hi all,

I am working on a study in which, each respondent places 5 objects on a
2-dimensional space. I have (x,y) of each object in my data. My goal is
to find out how those 5 objects are perceived on 2-dimensional space. I
was thinking of using MDS map. In past, I have done MDS map in SPSS, but
I had only one similarity matrix. Now, I think I have to deal with one
similarity matrix for each respondents My question is that:

1. Is there a command in SPSS to created similarity/distance matrix
using coordinates of objects?
2. How can I use similarity matrix of all respondents to come up with
MDS map?


Thanks in advance for your help!


Homayoun

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com <http://mail.yahoo.com/>



**********************************************************************
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.
**********************************************************************

[Faris Siraf]

Dear Frank,

Thank you for your response! I have looked at PROXIMITIES in past. To my understanding, it just calculates distance when we have only one coordinate. In my case, I have (x,y) coordinate for each object. Can PROXIMITIES calculate EUCLIDEAN distance between object A (x1,y1), and object B (x2,y2)? If so, do you have the syntax for doing that?


Sincerely,

Homayoun


----- Original Message ----
From: "Busing, Frank" <[hidden email]>
To: [hidden email]
Sent: Sunday, January 21, 2007 6:58:39 AM
Subject: Re: MDS map


D ear Homayoun.

In SPSS you can use the PROXIMITIES subcommand to create distances
between the coordinates of each respondent.
Then, you can use PROXSCAL to analyse these data with an individual
difference model.
This will give you your MDS map.

Good luck, Frank.


Frank Busing
Leiden University


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

Hi all,

I am working on a study in which, each respondent places 5 objects on a
2-dimensional space. I have (x,y) of each object in my data. My goal is
to find out how those 5 objects are perceived on 2-dimensional space. I
was thinking of using MDS map. In past, I have done MDS map in SPSS, but
I had only one similarity matrix. Now, I think I have to deal with one
similarity matrix for each respondents My question is that:

1. Is there a command in SPSS to created similarity/distance matrix
using coordinates of objects?
2. How can I use similarity matrix of all respondents to come up with
MDS map?


Thanks in advance for your help!


Homayoun

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com <http://mail.yahoo.com/>



**********************************************************************
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.
**********************************************************************

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

[zstatman]

I know this is an old thread, couldn't find an update or one applicable so ........

Each respondent noted a Yes, if a particular attribute was favorable and left it blank if it wasn't - Thus a binary situation (recoded the blanks as a No).

Now would like to do a Perceptual Map of the results.

Ran MDS using the 12 attributes and do get a viable Map but when compare it to the % of Yes v No responses the map doesn't make sense so clearly missing something and really going down the wrong path

Thoughts/ideas clearly welcome