Saving regression results

[William Dudley WNDUDLEY]

In the past I have used the outfile command to save regression results to
a  .sav file
e.g.


REGRESSION
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT  Y
  /METHOD=ENTER  X
  /OUTFILE=COVB( 'reg1_a.sav') .

Although I am happy with how this works, I would also like to save the
Model Rsquared value in addition to the coefficients.

Can this be done?

I am using version 16 and the previous syntax was developed on version 12
or 13 (it has been a while).


Thanks
Bill

=====================
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

[Swank, Paul R]

If OMS in SPSS works the same as ODS in SAS, then that would be the way.

Paul R. Swank, Ph.D
Professor and Director of Research
Children's Learning Institute
University of Texas Health Science Center
Houston, TX 77038


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
William Dudley WNDUDLEY
Sent: Wednesday, October 01, 2008 2:22 PM
To: [hidden email]
Subject: Saving regression results

In the past I have used the outfile command to save regression results
to
a  .sav file
e.g.


REGRESSION
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT  Y
  /METHOD=ENTER  X
  /OUTFILE=COVB( 'reg1_a.sav') .

Although I am happy with how this works, I would also like to save the
Model Rsquared value in addition to the coefficients.

Can this be done?

I am using version 16 and the previous syntax was developed on version
12
or 13 (it has been a while).


Thanks
Bill

=====================
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

[Peck, Jon]

OMS will handle this, saving the appropriate table as a new dataset or in various other formats.
Here's a simple example.

dataset declare modelsummary.
oms select tables
/if subtype='Model Summary'
/destination outfile=modelsummary format=sav.
<your regression command>
omsend.

You can put multiple REGRESSION commands between oms and omsend, and they will each contribute to the same dataset.  The R squared will be in a variable named "RSquare".

HTH,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of William Dudley WNDUDLEY
Sent: Wednesday, October 01, 2008 1:22 PM
To: [hidden email]
Subject: [SPSSX-L] Saving regression results

In the past I have used the outfile command to save regression results to
a  .sav file
e.g.


REGRESSION
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT  Y
  /METHOD=ENTER  X
  /OUTFILE=COVB( 'reg1_a.sav') .

Although I am happy with how this works, I would also like to save the
Model Rsquared value in addition to the coefficients.

Can this be done?

I am using version 16 and the previous syntax was developed on version 12
or 13 (it has been a while).


Thanks
Bill

=====================
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

[bgreen]

Hello,

I have been asked about the analysis of data that was originally
scaled as a 5 point Likert scale (with a  mid point of neither
agree/disagree) but has been collapsed into a 3 point scale. The
question of interest is whether respondents from 9 hospitals differ
in their ratings on these Likert attitude scales.

The analysis I was shown was a series of chi-square analyses: 9
hospitals x 3 attitude categories (agree, disagree & neither).  The
midpoint ratings are not a problem in this analysis but I do have
some concern about the number of hospitals.  I was considering
suggesting that the data not be collapsed and the analyses be based
on 5 scale points. If the scale is to be treated as ordinal, then the
midpoint is a problem. I imagine that it would need to be re-coded to
the first category. How best to deal with the midpoint is my first question.

Secondly, is a Kruskal-wallis test the prefrred analysis? I was also
considering a logistic regression - though can see merit in the
chi-square but have the above concern  about the number of categories
in both the 3 point and 5 point scale options, given there are 9 hospitals.

Any assiatance is much appreciated,

regards

bob

=====================
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