Obtaining analysis results via command syntax

[Barth Riley]

Hello



I have written a program that repeatedly generates a dataset and performs a
statistical analysis on that data for purposes of estimating power and
sample size. I would like to know if there is a way to find out whether the
result from an analysis was significant for a particular effect in the
model. I could then calculate the number of significant tests relative to
the total number of tests (iterations) conducted. Can anything like this be
done in command syntax?



Thanks



Barth Riley, Ph.D.

Res. Asst. Professor & Associate Program Director

Dept. of Disability and Human Development M/C 626

University of Illinois-Chicago

1640 W. Roosevelt Rd.

Chicago, IL 60608

Voice [phone]

Fax:   [phone]

Email:  <mailto:[hidden email]> [hidden email]

[Georg Maubach]

Hi Barth,

I am not sure what you really need. You can catch all output via the OMS command. This should contain all information you need to feed a macro that can iterate on a list of results. I can say more if I know more about your problem.

HTH

Best regards

Georg Maubach
Research Manager


-----Ursprüngliche Nachricht-----
Von: SPSSX(r) Discussion [mailto:[hidden email]] Im Auftrag von Barth Riley
Gesendet: Donnerstag, 30. August 2007 19:56
An: [hidden email]
Betreff: Obtaining analysis results via command syntax

Hello



I have written a program that repeatedly generates a dataset and performs a statistical analysis on that data for purposes of estimating power and sample size. I would like to know if there is a way to find out whether the result from an analysis was significant for a particular effect in the model. I could then calculate the number of significant tests relative to the total number of tests (iterations) conducted. Can anything like this be done in command syntax?



Thanks



Barth Riley, Ph.D.

Res. Asst. Professor & Associate Program Director

Dept. of Disability and Human Development M/C 626

University of Illinois-Chicago

1640 W. Roosevelt Rd.

Chicago, IL 60608

Voice [phone]

Fax:   [phone]

Email:  <mailto:[hidden email]> [hidden email]

[Georg Maubach]

Hi Barth,

Thanks for you off-list message. I will forward my answer to the list also so that everyone you needs the information can get it from the list.

In SPSS it is not as simple as this. You have to catch the output via OMS like

OMS
   /SELECT TABLES
   /IF COMMANDS=['Oneway']
       SUBTYPES=['ANOVA']
   /DESTINATION
    FORMAT = SAV
    OUTFILE = 'c:\test.sav'
   /TAG = "OMS1" .

ONEWAY
  f0801 BY brand
  /MISSING ANALYSIS .

OMSEND
    TAG = ['OMS1'] .

SAVE OUTFILE = 'c:\actual_dataset.sav' .

NEW FILE .
GET FILE = 'c:\test.sav' .

* Here follow all commands that process the output .
* It maybe that you have to save the different output tables in different OMS files
* to be able to process them separately.

NEW FILE .
GET FILE = 'c:\actual_dataset.sav' .

To use OMS you need the OMS command Id which you get if you look into the SPSS Command Syntax reference or by running the respective SPSS statistical procedure and right click on the table in the left navigation pane of the output. In the same way you get the OMS subtype Id.

HTH

Regards

Georg Maubach


-----Ursprüngliche Nachricht-----
Von: Barth Riley [mailto:[hidden email]]
Gesendet: Freitag, 31. August 2007 17:56
An: Maubach, Georg BC/MRSC
Betreff: RE: Obtaining analysis results via command syntax

Hi Georg

Thanks for your message. In Stata, one can obtain the significance of a test as follows:

anova y group time group*time
test group
if r(p) < .05 {
        ...
}

The "r(p)" returns the significance of the group effect as obtained through the test command. I would like to be able to do something similar to this in SPSS.  I am not familiar with OMS--if it can do something along these lines, I would like to know more about it.

Thanks

Barth


Barth Riley, Ph.D.
Res. Asst. Professor & Associate Program Director Dept. of Disability and Human Development M/C 626 University of Illinois-Chicago 1640 W. Roosevelt Rd.
Chicago, IL 60608
Voice [phone]
Fax:   [phone]
Email: [hidden email]

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Georg Maubach
Sent: Friday, August 31, 2007 2:33 AM
To: [hidden email]
Subject: AW: Obtaining analysis results via command syntax

Hi Barth,

I am not sure what you really need. You can catch all output via the OMS command. This should contain all information you need to feed a macro that can iterate on a list of results. I can say more if I know more about your problem.

HTH

Best regards

Georg Maubach
Research Manager


-----Ursprüngliche Nachricht-----
Von: SPSSX(r) Discussion [mailto:[hidden email]] Im Auftrag von Barth Riley
Gesendet: Donnerstag, 30. August 2007 19:56
An: [hidden email]
Betreff: Obtaining analysis results via command syntax

Hello



I have written a program that repeatedly generates a dataset and performs a statistical analysis on that data for purposes of estimating power and sample size. I would like to know if there is a way to find out whether the result from an analysis was significant for a particular effect in the model. I could then calculate the number of significant tests relative to the total number of tests (iterations) conducted. Can anything like this be done in command syntax?



Thanks



Barth Riley, Ph.D.

Res. Asst. Professor & Associate Program Director

Dept. of Disability and Human Development M/C 626

University of Illinois-Chicago

1640 W. Roosevelt Rd.

Chicago, IL 60608

Voice [phone]

Fax:   [phone]

Email:  <mailto:[hidden email]> [hidden email]