different frequencies before and after CASESTOVARS

[Tanya Temkin]

Hi SPSSers,

I'm having trouble with *apparent*  data discrepancies following a file
restructuring via CASESTOVARS.
I saved my unrestructured file, opened it and saved it as a new file, then
did my restructuring on the new file.

My unrestructured file had a structure in which each diagnosis received by
a patient was a record. This is a case-control study, so I counted the
number of cases and controls before restructuring, and re-counted the
unduplicated diagnosis frequency (unduplicated for the same patient, that
is).

This was the file structure (CA_CON indicates case or control status):
Pt_ID           CA_CON  undup_dx
10023           CA              headache
10023           CA              abd pain
10023           CA              burns
10036           CO              chest pain
10036           CO              headache
10047           CO              fx lower limb
10047           CO              asthma
10047           CO              fx upper limb
10049           CO              other GI
10049           CO              abd pain

I used this syntax to restructure my file:
CASESTOVARS
  /ID=Pt_ID
  /rename undup_dx=dx
  /fixed=CA_CON.
freq var CA_CON.

....and it seemed to work. I got, as intended, a string of variables --
dx.1 to dx.29 (one patient had 29 dx's):

Pt_ID   CA_CON    dx.1          dx.2            dx.3
10023   CA           headache   abd pain        burns
10036   CO           chest pain headache
10047   CO           fx lower limb      asthma          fx upper limb
10049   CO           other GI   abd pain
....etc.

Next I set up a string of dummy  variables, each corresponding to one of
the diagnostic categories. that will hold dichotomous values (1/0) for
presence or absence of the diagnosis for that patient. After initializing
all these variables' values to 0, I used the syntax
if any(dx.1 TO dx.29,"abd pain") dx_abd=1.
if any(dx.1 TO dx.29,"chest pain") dx_cp=1.
[etc.]

After entering a few of these I ran crosstabs (e.g. dx_abd by CA_CON) .to
see if they matched the diagnostic frequencies I had (via crosstabs for
undup_dx by CA_CON) before the CASESTOVARS restructure. They don't match -
current n's are much lower . I already know I have the same number of
cases and controls as before restructuring, so that's not the problem.
Perhaps my "any" syntax is?

I appreciate any insights on this. I'm stumped.


Tanya Temkin
Research Associate
AACC Reporting
Northern California Regional Office
The Permanente Medical Group
[phone]

NOTICE TO RECIPIENT:  If you are not the intended recipient of this
e-mail, you are prohibited from sharing, copying, or otherwise using or
disclosing its contents.  If you have received this e-mail in error,
please notify the sender immediately by reply e-mail and permanently
delete this e-mail and any attachments without reading, forwarding or
saving them.  Thank you.

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

[Tanya Temkin]

Ooops -- my bad. Bad syntax, that is.  I realized (right after I posted my
previous note) what I should've  been using wasn't IF ANY at all.
I tried
do repeat dxa=dx.1 TO dx.29.
  if dxa="abdominal pain" dx_abd=1.
end repeat.
crosstabs tables dx_abd by CA_CON.

Much much better. I got the N's I expected, same as in unrestructured
file.

Tanya Temkin
Research Associate
AACC Reporting
Northern California Regional Office
The Permanente Medical Group
[phone]

NOTICE TO RECIPIENT:  If you are not the intended recipient of this
e-mail, you are prohibited from sharing, copying, or otherwise using or
disclosing its contents.  If you have received this e-mail in error,
please notify the sender immediately by reply e-mail and permanently
delete this e-mail and any attachments without reading, forwarding or
saving them.  Thank you.

----- Forwarded by Tanya L TemKin/CA/KAIPERM on 10/25/2007 04:48 PM -----

Tanya L TemKin/CA/KAIPERM
10/25/2007 03:03 PM

To
[hidden email]
cc

Subject
different frequencies before and after CASESTOVARS





Hi SPSSers,

I'm having trouble with *apparent*  data discrepancies following a file
restructuring via CASESTOVARS.
I saved my unrestructured file, opened it and saved it as a new file, then
did my restructuring on the new file.

My unrestructured file had a structure in which each diagnosis received by
a patient was a record. This is a case-control study, so I counted the
number of cases and controls before restructuring, and re-counted the
unduplicated diagnosis frequency (unduplicated for the same patient, that
is).

This was the file structure (CA_CON indicates case or control status):
Pt_ID           CA_CON  undup_dx
10023           CA              headache
10023           CA              abd pain
10023           CA              burns
10036           CO              chest pain
10036           CO              headache
10047           CO              fx lower limb
10047           CO              asthma
10047           CO              fx upper limb
10049           CO              other GI
10049           CO              abd pain

I used this syntax to restructure my file:
CASESTOVARS
  /ID=Pt_ID
  /rename undup_dx=dx
  /fixed=CA_CON.
freq var CA_CON.

....and it seemed to work. I got, as intended, a string of variables --
dx.1 to dx.29 (one patient had 29 dx's):

Pt_ID   CA_CON    dx.1          dx.2            dx.3
10023   CA           headache   abd pain        burns
10036   CO           chest pain headache
10047   CO           fx lower limb      asthma          fx upper limb
10049   CO           other GI   abd pain
....etc.

Next I set up a string of dummy  variables, each corresponding to one of
the diagnostic categories. that will hold dichotomous values (1/0) for
presence or absence of the diagnosis for that patient. After initializing
all these variables' values to 0, I used the syntax
if any(dx.1 TO dx.29,"abd pain") dx_abd=1.
if any(dx.1 TO dx.29,"chest pain") dx_cp=1.
[etc.]

After entering a few of these I ran crosstabs (e.g. dx_abd by CA_CON) .to
see if they matched the diagnostic frequencies I had (via crosstabs for
undup_dx by CA_CON) before the CASESTOVARS restructure. They don't match -
current n's are much lower . I already know I have the same number of
cases and controls as before restructuring, so that's not the problem.
Perhaps my "any" syntax is?

I appreciate any insights on this. I'm stumped.


Tanya Temkin
Research Associate
AACC Reporting
Northern California Regional Office
The Permanente Medical Group
[phone]

NOTICE TO RECIPIENT:  If you are not the intended recipient of this
e-mail, you are prohibited from sharing, copying, or otherwise using or
disclosing its contents.  If you have received this e-mail in error,
please notify the sender immediately by reply e-mail and permanently
delete this e-mail and any attachments without reading, forwarding or
saving them.  Thank you.

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

[Melissa Ives]

ANY should have worked, however it looks like you had the arguments
backwards--should have the test value (i.e. the diagnosis text, first
followed by the variable list.

The syntax "if any(dx.1 TO dx.29,"abd pain")." look for any value
matching dx.1 in dx.2 to dx.29 and in abd pain.
This would have worked better:
        if any("abd pain",dx.1 TO dx.29).

Per SPSS help....
ANY. ANY(test,value[,value,...]). Logical. Returns 1 or true if the
value of test matches any of the subsequent values; returns 0 or false
otherwise. This function requires two or more arguments.

Melissa

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Tanya Temkin
Sent: Thursday, October 25, 2007 6:54 PM
To: [hidden email]
Subject: Re: [SPSSX-L] different frequencies before and after
CASESTOVARS

Ooops -- my bad. Bad syntax, that is.  I realized (right after I posted
my previous note) what I should've  been using wasn't IF ANY at all.
I tried
do repeat dxa=dx.1 TO dx.29.
  if dxa="abdominal pain" dx_abd=1.
end repeat.
crosstabs tables dx_abd by CA_CON.

Much much better. I got the N's I expected, same as in unrestructured
file.

Tanya Temkin
Research Associate
AACC Reporting
Northern California Regional Office
The Permanente Medical Group
[phone]

NOTICE TO RECIPIENT:  If you are not the intended recipient of this
e-mail, you are prohibited from sharing, copying, or otherwise using or
disclosing its contents.  If you have received this e-mail in error,
please notify the sender immediately by reply e-mail and permanently
delete this e-mail and any attachments without reading, forwarding or
saving them.  Thank you.

----- Forwarded by Tanya L TemKin/CA/KAIPERM on 10/25/2007 04:48 PM
-----

Tanya L TemKin/CA/KAIPERM
10/25/2007 03:03 PM

To
[hidden email]
cc

Subject
different frequencies before and after CASESTOVARS





Hi SPSSers,

I'm having trouble with *apparent*  data discrepancies following a file
restructuring via CASESTOVARS.
I saved my unrestructured file, opened it and saved it as a new file,
then did my restructuring on the new file.

My unrestructured file had a structure in which each diagnosis received
by a patient was a record. This is a case-control study, so I counted
the number of cases and controls before restructuring, and re-counted
the unduplicated diagnosis frequency (unduplicated for the same patient,
that is).

This was the file structure (CA_CON indicates case or control status):
Pt_ID           CA_CON  undup_dx
10023           CA              headache
10023           CA              abd pain
10023           CA              burns
10036           CO              chest pain
10036           CO              headache
10047           CO              fx lower limb
10047           CO              asthma
10047           CO              fx upper limb
10049           CO              other GI
10049           CO              abd pain

I used this syntax to restructure my file:
CASESTOVARS
  /ID=Pt_ID
  /rename undup_dx=dx
  /fixed=CA_CON.
freq var CA_CON.

....and it seemed to work. I got, as intended, a string of variables --
dx.1 to dx.29 (one patient had 29 dx's):

Pt_ID   CA_CON    dx.1          dx.2            dx.3
10023   CA           headache   abd pain        burns
10036   CO           chest pain headache
10047   CO           fx lower limb      asthma          fx upper limb
10049   CO           other GI   abd pain
....etc.

Next I set up a string of dummy  variables, each corresponding to one of
the diagnostic categories. that will hold dichotomous values (1/0) for
presence or absence of the diagnosis for that patient. After
initializing all these variables' values to 0, I used the syntax if
any(dx.1 TO dx.29,"abd pain") dx_abd=1.
if any(dx.1 TO dx.29,"chest pain") dx_cp=1.
[etc.]

After entering a few of these I ran crosstabs (e.g. dx_abd by CA_CON)
.to see if they matched the diagnostic frequencies I had (via crosstabs
for undup_dx by CA_CON) before the CASESTOVARS restructure. They don't
match - current n's are much lower . I already know I have the same
number of cases and controls as before restructuring, so that's not the
problem.
Perhaps my "any" syntax is?

I appreciate any insights on this. I'm stumped.


Tanya Temkin
Research Associate
AACC Reporting
Northern California Regional Office
The Permanente Medical Group
[phone]

NOTICE TO RECIPIENT:  If you are not the intended recipient of this
e-mail, you are prohibited from sharing, copying, or otherwise using or
disclosing its contents.  If you have received this e-mail in error,
please notify the sender immediately by reply e-mail and permanently
delete this e-mail and any attachments without reading, forwarding or
saving them.  Thank you.

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


PRIVILEGED AND CONFIDENTIAL INFORMATION
This transmittal and any attachments may contain PRIVILEGED AND
CONFIDENTIAL information and is intended only for the use of the
addressee. If you are not the designated recipient, or an employee
or agent authorized to deliver such transmittals to the designated
recipient, you are hereby notified that any dissemination,
copying or publication of this transmittal is strictly prohibited. If
you have received this transmittal in error, please notify us
immediately by replying to the sender and delete this copy from your
system. You may also call us at [phone] for assistance.

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