SPSSX Discussion

V19 error? retraction

classic Classic list List threaded Threaded
♦
♦
 3 messages Options
Maguin, Eugene
Reply | Threaded
Open this post in threaded view
♦
♦
|

V19 error? retraction

Maguin, Eugene
Apologies. I got tripped up by the unanticipated result of a previous operation. But that result seems quite odd to me as well. Briefly. Consider this command sequence.

OMS SELECT TABLES/IF COMMANDS=['CROSSTABS'] SUBTYPES=['CROSSTABULATION']/
   DESTINATION FORMAT=SAV OUTFILE='H:\filename1'.
CROSSTABS PROGRAM BY pos edu maj/CELLS=COUNT ROW.
OMSEND.

OMS SELECT TABLES/IF COMMANDS=['MEANS'] SUBTYPES=['REPORT']/
   DESTINATION FORMAT=SAV OUTFILE='H:\filename2'.
MEANS yrsexp yrspres age FEMALE BY PROGRAM/CELLS=COUNT MEAN MIN MAX MEDIAN STDDEV.
OMSEND.

Here's the really, really odd thing, I think.

The data part of the first OMS file, that would be the cell counts and cell row percents for each value of program and repeated for each of the three listed variables, are numeric variables. As I would expect and I think that is a reasonable expectation.
BUT
The data part of the second OMS file, that would be the contents of the report table for specified variable set for each value of program are STRING variables. I would not expect this and I don't think it should function this way. As I see it, there is absolutely no reason for those values to coded as strings.

Gene Maguin

=====================
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
Salbod
Reply | Threaded
Open this post in threaded view
♦
♦
|

Re: V19 error? retraction

Salbod
Hi Eugene, I played around with it using Version 20, and with a different data set. It seem to work.  Is it possible you have strings in the data? --Steve

GET
  FILE='C:\Program Files\IBM\SPSS\Statistics\20\Samples\English\customer_dbase.sav'.
DATASET NAME Cutomer WINDOW=FRONT.

DATASET DECLARE OMS_stuff1.
DATASET DECLARE OMS_stuff2.
OMS SELECT TABLES
    /IF COMMANDS=['CROSSTABS'] SUBTYPES=['CROSSTABULATION']
    /DESTINATION FORMAT=SAV OUTFILE='OMS_stuff1'.
CROSSTABS gender BY agecat
    /CELLS=COUNT ROW.
OMSEND.

OMS SELECT TABLES
    /IF COMMANDS=['MEANS'] SUBTYPES=['REPORT']
   /DESTINATION FORMAT=SAV OUTFILE='OMS_stuff2'.
MEANS ed age income BY gender
   /CELLS=COUNT MEAN MIN MAX MEDIAN STDDEV.
OMSEND.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Maguin, Eugene
Sent: Friday, April 20, 2012 12:02 PM
To: [hidden email]
Subject: V19 error? retraction

Apologies. I got tripped up by the unanticipated result of a previous operation. But that result seems quite odd to me as well. Briefly. Consider this command sequence.

OMS SELECT TABLES/IF COMMANDS=['CROSSTABS'] SUBTYPES=['CROSSTABULATION']/
   DESTINATION FORMAT=SAV OUTFILE='H:\filename1'.
CROSSTABS PROGRAM BY pos edu maj/CELLS=COUNT ROW.
OMSEND.

OMS SELECT TABLES/IF COMMANDS=['MEANS'] SUBTYPES=['REPORT']/
   DESTINATION FORMAT=SAV OUTFILE='H:\filename2'.
MEANS yrsexp yrspres age FEMALE BY PROGRAM/CELLS=COUNT MEAN MIN MAX MEDIAN STDDEV.
OMSEND.

Here's the really, really odd thing, I think.

The data part of the first OMS file, that would be the cell counts and cell row percents for each value of program and repeated for each of the three listed variables, are numeric variables. As I would expect and I think that is a reasonable expectation.
BUT
The data part of the second OMS file, that would be the contents of the report table for specified variable set for each value of program are STRING variables. I would not expect this and I don't think it should function this way. As I see it, there is absolutely no reason for those values to coded as strings.

Gene Maguin

=====================
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
Maguin, Eugene
Reply | Threaded
Open this post in threaded view
♦
♦
|

Re: V19 error? retraction

Maguin, Eugene
Stephen,
No, I don't. The variables that go into the Means command are all numeric and I get valid-appearing out from the means command. If I were putting strings in, I think I'd get an error because there couldn't be a computation on strings.

Well, who knows. It sounds like it's fixed in 20, which is good.
Thanks, Gene Maguin

-----Original Message-----
From: Salbod, Mr. Stephen [mailto:[hidden email]]
Sent: Friday, April 20, 2012 3:32 PM
To: Maguin, Eugene; [hidden email]
Subject: RE: V19 error? retraction

Hi Eugene, I played around with it using Version 20, and with a different data set. It seem to work.  Is it possible you have strings in the data? --Steve

GET
  FILE='C:\Program Files\IBM\SPSS\Statistics\20\Samples\English\customer_dbase.sav'.
DATASET NAME Cutomer WINDOW=FRONT.

DATASET DECLARE OMS_stuff1.
DATASET DECLARE OMS_stuff2.
OMS SELECT TABLES
    /IF COMMANDS=['CROSSTABS'] SUBTYPES=['CROSSTABULATION']
    /DESTINATION FORMAT=SAV OUTFILE='OMS_stuff1'.
CROSSTABS gender BY agecat
    /CELLS=COUNT ROW.
OMSEND.

OMS SELECT TABLES
    /IF COMMANDS=['MEANS'] SUBTYPES=['REPORT']
   /DESTINATION FORMAT=SAV OUTFILE='OMS_stuff2'.
MEANS ed age income BY gender
   /CELLS=COUNT MEAN MIN MAX MEDIAN STDDEV.
OMSEND.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Maguin, Eugene
Sent: Friday, April 20, 2012 12:02 PM
To: [hidden email]
Subject: V19 error? retraction

Apologies. I got tripped up by the unanticipated result of a previous operation. But that result seems quite odd to me as well. Briefly. Consider this command sequence.

OMS SELECT TABLES/IF COMMANDS=['CROSSTABS'] SUBTYPES=['CROSSTABULATION']/
   DESTINATION FORMAT=SAV OUTFILE='H:\filename1'.
CROSSTABS PROGRAM BY pos edu maj/CELLS=COUNT ROW.
OMSEND.

OMS SELECT TABLES/IF COMMANDS=['MEANS'] SUBTYPES=['REPORT']/
   DESTINATION FORMAT=SAV OUTFILE='H:\filename2'.
MEANS yrsexp yrspres age FEMALE BY PROGRAM/CELLS=COUNT MEAN MIN MAX MEDIAN STDDEV.
OMSEND.

Here's the really, really odd thing, I think.

The data part of the first OMS file, that would be the cell counts and cell row percents for each value of program and repeated for each of the three listed variables, are numeric variables. As I would expect and I think that is a reasonable expectation.
BUT
The data part of the second OMS file, that would be the contents of the report table for specified variable set for each value of program are STRING variables. I would not expect this and I don't think it should function this way. As I see it, there is absolutely no reason for those values to coded as strings.

Gene Maguin

=====================
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
Free forum by Nabble Edit this page