Recode and lose values

[Leslie-26]

Hello, and thanks in advance if you can help:

I am recoding three environment variables (civilian, military, blend
military/civilian) into two age groups (1-young, 2-old) and find that on the
recode the total number between the groups is not equal to the original
number, in addition, one environment appears to have no value.  Below, I
include the descriptives before the recode and then after the recode, and
below that I've embedded the syntax.

If you can help me understand where I've gone wrong I greatly appreciate
your insight and help.

Leslie
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

DESCRIPTIVES  BEFORE RECODE
AGE BEFORE
AgeG2r
    Frequency Percent Valid Percent Cumulative Percent
Valid 1.00 440 44.5 44.6 44.6
 2.00 546 55.2 55.4 100.0
 Total 986 99.7 100.0
Missing System 3 .3
Total  989 100.0

CIVILIAN BEFORE RECODE
ETHN3_1c
    Frequency Percent Valid Percent Cumulative Percent
Valid 1.00 28 2.8 71.8 71.8
 2.00 8 .8 20.5 92.3
 3.00 3 .3 7.7 100.0
 Total 39 3.9 100.0
Missing System 950 96.1
Total  989 100.0

DESCRIPTIVES AFTER RECODE
CIVILIAN AFTER
ETHN3_1c1
    Frequency Percent Valid Percent Cumulative Percent
Valid 2.00 3 .3 60.0 60.0
 3.00 2 .2 40.0 100.0
 Total 5 .5 100.0
Missing System 984 99.5
Total  989 100.0



SYNTAX
COMMENT break each ethnicity/environment into two age groups.

COMMENT civilian young
IF (AgeG2r = 1 & ETHN3_1c  = 1) ETHN3_1c1 = 1 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1c  = 2) ETHN3_1c1 = 2 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1c  = 3) ETHN3_1c1 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'young white Citizen Only'
                       2 'young afam citizen only'
  3 'young other citizen only'
EXECUTE.
FREQUENCIES
  VARIABLES= AgeG2r ETHN3_1c   ETHN3_1c1
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .

COMMENT civilian old
IF (AgeG2r = 2 & ETHN3_1c  = 1) ETHN3_1c2 = 1 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1c  = 2) ETHN3_1c2 = 2 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1c  = 3) ETHN3_1c2 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'old white Citizen Only'
                       2 'old afam citizen only'
  3 'old other citizen only'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1c2
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .


COMMENT military young
IF (AgeG2r = 1 & ETHN3_1m  = 1) ETHN3_1m1 = 1 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1m  = 2) ETHN3_1m1 = 2 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1m  = 3) ETHN3_1m1 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'young white military Only'
                       2 'young afam military only'
  3 'young other military only'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1m1
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .

COMMENT military old
IF (AgeG2r = 2 & ETHN3_1m  = 1) ETHN3_1m2 = 1 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1m  = 2) ETHN3_1m2 = 2 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1m  = 3) ETHN3_1m2 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'old white military Only'
                       2 'old afam military only'
  3 'old other military only'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1m2
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .


COMMENT both young
IF (AgeG2r = 1 & ETHN3_1cm  = 1) ETHN3_1cm1 = 1 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1cm  = 2) ETHN3_1cm1 = 2 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1cm  = 3) ETHN3_1cm1 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'young white both'
                       2 'young afam both'
  3 'young other both'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1cm1
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .

COMMENT both old
IF (AgeG2r = 2 & ETHN3_1cm  = 1) ETHN3_1cm2 = 1 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1cm  = 2) ETHN3_1cm2 = 2 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1cm  = 3) ETHN3_1cm2 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'old white both'
                       2 'old afam both'
  3 'old other both'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1cm2
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .

[Spousta Jan]

Hello Leslie,

The first try: Make full stops (".") after _all_ COMMENTs and try again. E.g. Instead of

COMMENT civilian young
IF (AgeG2r = 1 & ETHN3_1c  = 1) ETHN3_1c1 = 1 .
EXECUTE .

write

COMMENT civilian young .
IF (AgeG2r = 1 & ETHN3_1c  = 1) ETHN3_1c1 = 1 .
EXECUTE .

(Because otherwise SPSS does not know where the comment ends.)

Then let us know whether it works.

Good luck,

Jan

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Leslie
Sent: Tuesday, September 11, 2007 7:42 AM
To: [hidden email]
Subject: Recode and lose values

Hello, and thanks in advance if you can help:

I am recoding three environment variables (civilian, military, blend
military/civilian) into two age groups (1-young, 2-old) and find that on the recode the total number between the groups is not equal to the original number, in addition, one environment appears to have no value.  Below, I include the descriptives before the recode and then after the recode, and below that I've embedded the syntax.

If you can help me understand where I've gone wrong I greatly appreciate your insight and help.

Leslie
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

DESCRIPTIVES  BEFORE RECODE
AGE BEFORE
AgeG2r
    Frequency Percent Valid Percent Cumulative Percent Valid 1.00 440 44.5 44.6 44.6  2.00 546 55.2 55.4 100.0  Total 986 99.7 100.0 Missing System 3 .3 Total  989 100.0

CIVILIAN BEFORE RECODE
ETHN3_1c
    Frequency Percent Valid Percent Cumulative Percent Valid 1.00 28 2.8 71.8 71.8  2.00 8 .8 20.5 92.3  3.00 3 .3 7.7 100.0  Total 39 3.9 100.0 Missing System 950 96.1 Total  989 100.0

DESCRIPTIVES AFTER RECODE
CIVILIAN AFTER
ETHN3_1c1
    Frequency Percent Valid Percent Cumulative Percent Valid 2.00 3 .3 60.0 60.0  3.00 2 .2 40.0 100.0  Total 5 .5 100.0 Missing System 984 99.5 Total  989 100.0



SYNTAX
COMMENT break each ethnicity/environment into two age groups.

COMMENT civilian young
IF (AgeG2r = 1 & ETHN3_1c  = 1) ETHN3_1c1 = 1 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1c  = 2) ETHN3_1c1 = 2 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1c  = 3) ETHN3_1c1 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'young white Citizen Only'
                       2 'young afam citizen only'
  3 'young other citizen only'
EXECUTE.
FREQUENCIES
  VARIABLES= AgeG2r ETHN3_1c   ETHN3_1c1
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .

COMMENT civilian old
IF (AgeG2r = 2 & ETHN3_1c  = 1) ETHN3_1c2 = 1 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1c  = 2) ETHN3_1c2 = 2 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1c  = 3) ETHN3_1c2 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'old white Citizen Only'
                       2 'old afam citizen only'
  3 'old other citizen only'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1c2
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .


COMMENT military young
IF (AgeG2r = 1 & ETHN3_1m  = 1) ETHN3_1m1 = 1 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1m  = 2) ETHN3_1m1 = 2 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1m  = 3) ETHN3_1m1 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'young white military Only'
                       2 'young afam military only'
  3 'young other military only'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1m1
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .

COMMENT military old
IF (AgeG2r = 2 & ETHN3_1m  = 1) ETHN3_1m2 = 1 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1m  = 2) ETHN3_1m2 = 2 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1m  = 3) ETHN3_1m2 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'old white military Only'
                       2 'old afam military only'
  3 'old other military only'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1m2
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .


COMMENT both young
IF (AgeG2r = 1 & ETHN3_1cm  = 1) ETHN3_1cm1 = 1 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1cm  = 2) ETHN3_1cm1 = 2 .
EXECUTE .
IF (AgeG2r = 1 & ETHN3_1cm  = 3) ETHN3_1cm1 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'young white both'
                       2 'young afam both'
  3 'young other both'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1cm1
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .

COMMENT both old
IF (AgeG2r = 2 & ETHN3_1cm  = 1) ETHN3_1cm2 = 1 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1cm  = 2) ETHN3_1cm2 = 2 .
EXECUTE .
IF (AgeG2r = 2 & ETHN3_1cm  = 3) ETHN3_1cm2 = 3 .
EXECUTE .
VALUE LABEL ENV 1 'old white both'
                       2 'old afam both'
  3 'old other both'
EXECUTE.
FREQUENCIES
  VARIABLES=ETHN3_1cm2
  /BARCHART  FREQ
  /ORDER=  ANALYSIS .



_____

Tato zpráva a všechny připojené soubory jsou důvěrné a určené výlučně adresátovi(-ům). Jestliže nejste oprávněným adresátem, je zakázáno jakékoliv zveřejňování, zprostředkování nebo jiné použití těchto informací. Jestliže jste tento mail dostali neoprávněně, prosím, uvědomte odesilatele a smažte zprávu i přiložené soubory. Odesilatel nezodpovídá za jakékoliv chyby nebo opomenutí způsobené tímto přenosem.

This message and any attached files are confidential and intended solely for the addressee(s). Any publication, transmission or other use of the information by a person or entity other than the intended addressee is prohibited. If you receive this in error please contact the sender and delete the message as well as all attached documents. The sender does not accept liability for any errors or omissions as a result of the transmission.

-.- --

[Richard Ristow]

At 01:41 AM 9/11/2007, Leslie wrote:

>I am recoding three environment variables (civilian, military, blend
>military/civilian) into two age groups (1-young, 2-old) and find that
>on the recode the total number between the groups is not equal to the
>original number, in addition, one environment appears to have no value.

First, of course, Jan Spousta's suggestion to close all your COMMENT
statements with period. I believe that, by not doing so, you lose the
following statements:

>COMMENT civilian young
>IF (AgeG2r = 1 & ETHN3_1c  = 1) ETHN3_1c1 = 1 .

>COMMENT civilian old
>IF (AgeG2r = 2 & ETHN3_1c  = 1) ETHN3_1c2 = 1 .

>COMMENT military young
>IF (AgeG2r = 1 & ETHN3_1m  = 1) ETHN3_1m1 = 1 .

>COMMENT military old
>IF (AgeG2r = 2 & ETHN3_1m  = 1) ETHN3_1m2 = 1 .

>COMMENT both young
>IF (AgeG2r = 1 & ETHN3_1cm  = 1) ETHN3_1cm1 = 1 .

That'll mean none of those variables ever have value 1, which seems to
be what happens; so that may well be your whole problem.

Second, if you don't get what you want when you're creating one
categorical variable from two, CROSSTABS is a useful diagnostic. For
the first, for example,

CROSSTABS AgeG2r BY ETHN3_1c.

That'll show you all the categories directly, and you'll often spot
ones that are empty because of missing values or the like.

Third, and a side issue: none of those EXECUTE statements are doing
anything for you. With 1,000 cases, I'll bet you see a noticeable
speed-up if you remove them.

-Best of luck,
  Richard Ristow

[Dale Glaser]

Hi all....I have a colleague who is trying to find a software that will convert handwriting to text for qualitative analysis...........though I did a cursory search on the web, I did find one software that does such but have no idea if it is worth purchasing....any recommendations for such a type of software would be most appreciated.

  Thank you...Dale


Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com

[ViAnn Beadle]

Most handwriting recognition software is designed for onscreen capture
(PDAs, Tablet PCs, etc.) from a single writer and must be "trained". The
training is typically aimed at the user and not the software -- you keep
trying until it recognizes what you've written. (sigh)

That which works well is based upon "stylized" printed characters (think
Graffiti). Cardiff Software will recognize printed characters placed in
little boxes on scanned forms. IIRC, they claim over 90% accuracy but still
suggest that each complete field be manually checked for a real person.

If your colleague is also expecting accurate results from cursive script
that is an order of magnitude more difficult.

I'd be interested in how the search turns out.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Dale Glaser
Sent: Tuesday, September 11, 2007 1:33 PM
To: [hidden email]
Subject: Software for Qual Res (handwriting to text)

Hi all....I have a colleague who is trying to find a software that will
convert handwriting to text for qualitative analysis...........though I did
a cursory search on the web, I did find one software that does such but have
no idea if it is worth purchasing....any recommendations for such a type of
software would be most appreciated.

  Thank you...Dale


Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com