Graphical representation of a principal component factor analysis

[farwan]

Hello,

I can easily compute a factorial analysis with SPSS, but after i would like to plot a graph showing the factors as axes (eventually before and after rotation, but at least after rotation), and variables and subjects as dots. this done with at least two, hopefully three factors. For example, if i run a factor analysis and get two factors out of it, if i then want to picture the males and females subjects on each of those two dimensions how do i do that? I think i ll have to make the graph with excell if i want something presentable, but which datas should i take into account to get that kind of results?

Can anyone help?

Have a nice day everyone :-)

[Brian Moore-7]

Hi all-
Is it possible to export all of the variable view to a spreadsheet?

Barring that the problem I'm trying to solve is a set of evaluations with different variables for each speaker at an event where I have a structured set of 4 questions per speaker, but the variable names are NOT consistent but order /placement within a group is consistent.  So with a spreadsheet I could:
a- sort by labels column and get all of the variables that go with each question in a list [to get at overall average for example]
b- cast speaker name onto the full set of 4 in original sort order with an excel formula

Anyhow - any suggestions much appreciated.
Regards,
Brian


Brian Moore
Market Research Manager
WorldatWork
Washington, D.C. Office & Conference Center
1100 13th St., NW
Suite 800
Washington, D.C. 20005 USA
Tel: [phone] Fax: [phone]
E-mail: [hidden email]
www.worldatwork.org

UPCOMING CONFERENCES:
Total Rewards 2009 Asia Pacific
Oct. 7-9 | Hong Kong

Total Rewards 2010
May 16-19  |  Dallas / Ft.Worth, TX

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Peck, Jon]

You can sort Variable View by any column.  Right click on the header and choose from the context menu.  You will be offered the option of saving the original sort order as a custom attribute, so you can restore the original order later if desired.
I'm not sure what b means.

Regards,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Brian Moore
Sent: Tuesday, June 09, 2009 3:20 PM
To: [hidden email]
Subject: [SPSSX-L] Export Variable View to a Spreadsheet

Hi all-
Is it possible to export all of the variable view to a spreadsheet?

Barring that the problem I'm trying to solve is a set of evaluations with different variables for each speaker at an event where I have a structured set of 4 questions per speaker, but the variable names are NOT consistent but order /placement within a group is consistent.  So with a spreadsheet I could:
a- sort by labels column and get all of the variables that go with each question in a list [to get at overall average for example]
b- cast speaker name onto the full set of 4 in original sort order with an excel formula

Anyhow - any suggestions much appreciated.
Regards,
Brian


Brian Moore
Market Research Manager
WorldatWork
Washington, D.C. Office & Conference Center
1100 13th St., NW
Suite 800
Washington, D.C. 20005 USA
Tel: [phone] Fax: [phone]
E-mail: [hidden email]
www.worldatwork.org

UPCOMING CONFERENCES:
Total Rewards 2009 Asia Pacific
Oct. 7-9 | Hong Kong

Total Rewards 2010
May 16-19  |  Dallas / Ft.Worth, TX

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Brian Moore-7]

Sorry, I should have mentioned I'm using version 15 where right click
nor double click appear to do anything.

& in b I will extract the name of the presenter from the variable label
text in the first question to use in each automated report page.

Example (current):
Q1- This speaker (Joe Johnson) was knowledgeable.
Q2- This speaker conducted himself/herself professionally.
Q3- This speaker tried to sell something.
Q4- I would attend another session by this speaker.


Goal:
Report for                                                      var
extracted from Q1->     Joe Johnson                     Compared to
overall
Q1- This speaker  was knowledgeable.
Q2- This speaker conducted himself/herself professionally.
Q3- This speaker tried to sell something.
Q4- I would attend another session by this speaker.
-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Peck, Jon
Sent: Tuesday, June 09, 2009 6:07 PM
To: [hidden email]
Subject: Re: Export Variable View to a Spreadsheet

You can sort Variable View by any column.  Right click on the header and
choose from the context menu.  You will be offered the option of saving
the original sort order as a custom attribute, so you can restore the
original order later if desired.
I'm not sure what b means.

Regards,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Brian Moore
Sent: Tuesday, June 09, 2009 3:20 PM
To: [hidden email]
Subject: [SPSSX-L] Export Variable View to a Spreadsheet

Hi all-
Is it possible to export all of the variable view to a spreadsheet?

Barring that the problem I'm trying to solve is a set of evaluations
with different variables for each speaker at an event where I have a
structured set of 4 questions per speaker, but the variable names are
NOT consistent but order /placement within a group is consistent.  So
with a spreadsheet I could:
a- sort by labels column and get all of the variables that go with each
question in a list [to get at overall average for example]
b- cast speaker name onto the full set of 4 in original sort order with
an excel formula

Anyhow - any suggestions much appreciated.
Regards,
Brian


Brian Moore
Market Research Manager
WorldatWork
Washington, D.C. Office & Conference Center
1100 13th St., NW
Suite 800
Washington, D.C. 20005 USA
Tel: [phone] Fax: [phone]
E-mail: [hidden email]
www.worldatwork.org

UPCOMING CONFERENCES:
Total Rewards 2009 Asia Pacific
Oct. 7-9 | Hong Kong

Total Rewards 2010
May 16-19  |  Dallas / Ft.Worth, TX

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[William Dudley WNDUDLEY]

I am using mixed to estimate individual growth models for changes in hip flexion upon landing from a jump.
I have hip flexion data at the millisecond level.  Subjects vary in  the length of their landing epoch form about 200 msec to 500 msec.
I have the data set reformatted to that I have columns for id,  msec, and flexion.
I created a quadratic term (time squared and cubic term Time raised to the 3rd.


my syntax look likes this

mixed  HipFLX_mean_1 with time timesq timecu
  /print=solution
  /method=ml
  /fixed=intercept time timesq timecu
  /random=intercept time timesq timecu| subject( Subject) covtype(un).


I may eventually focus the analyses on the 25th percentile of the epoch (for some this would be the first 50 msec and for others it could be 200 msecs)
Thus the design is unbalanced.

In the past I have run models somewhat like this in SAS proc mixed and was provided with stats on the number of missing observations
 (subsets with only 300 msec are "missing" 200 observations because the slowest lander has 500 msecs worth of landing).

Two questions for now...

Do I have to indicate missinging so that I have a balanced data set?
How can I capture each individual growth parameters?


Thanks in advance


William N. Dudley, PhD
Associate Dean for Research
The School of Health and Human Performance Office of Research
The University of North Carolina at Greensboro
126 HHP Building, PO Box 26170
Greensboro, NC 27402-6170
VOICE 336.2562475
FAX [phone]