Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: data set reconfiguration


From   Philip Ryan <philip.ryan@adelaide.edu.au>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: data set reconfiguration
Date   Sun, 9 Nov 2003 00:27:44 +1030

Clint

It's a little hard to be sure exactly how to approach this most efficiently as 
your data all *appear* as strings, but of course may just be labels hiding 
numbers. In what follows I will assume that the outcome variables (q1_1 and 
q1_2) are in fact numeric - otherwise you won't be able to do a signed rank 
test - but if they are strings you had better first change them to numeric.  I 
will however assume that the variable "eye" holding values "OD" and "OS" is 
string, only because this leads to a demonstration of a required extra option 
in the -reshape- command.  (may one guess that "OD" signifies the right - 
dextra- eye, and "OS" the left - sinistra - eye??).  

Anyway, assuming that "None" for the outcome variables might really be a score 
of 0 in some, at least ordinal, scoring system:

. list , nolabel

     +------------------------------------------+
     | jmec_no   eye    iol   q1_1   q1_2   sex |
     |------------------------------------------|
  1. |   30332    OD   SA60      0      0     F |
  2. |   30332    OS   AR40      0      0     F |
  3. |   46199    OD   SA60      0      0     F |
  4. |   46199    OS   AR40      0      0     F |
     +------------------------------------------+

then

. reshape wide iol q1*, i(jmec_no) j(eye) string
(note: j = OD OS)

Data                               long   ->   wide
-----------------------------------------------------------------------------
Number of obs.                        4   ->       2
Number of variables                   6   ->       8
j variable (2 values)               eye   ->   (dropped)
xij variables:
                                    iol   ->   iolOD iolOS
                                   q1_1   ->   q1_1OD q1_1OS
                                   q1_2   ->   q1_2OD q1_2OS
-----------------------------------------------------------------------------

Note that:

1.  I had to specify the -string- option, because the j index is ordinarily 
expected to be numeric.

2.  I had to either drop the iol variable or (as I ended up choosing) include 
it in the list of variables I wanted reshaped, since iol is not constant within

each case identifier.


. list, nolabel

     +-------------------------------------------------------------------+
     | jmec_no   iolOD   q1_1OD   q1_2OD   iolOS   q1_1OS   q1_2OS   sex |
     |-------------------------------------------------------------------|
  1. |   30332    SA60        0        0    AR40        0        0     F |
  2. |   46199    SA60        0        0    AR40        0        0     F |
     +-------------------------------------------------------------------+

and, given some more data to make it sensible, one could:

. signrank  q1_1OD =  q1_1OS

to do a paired test on outcome q1_1 between eyes, and

. signrank  q1_2OD =  q1_2OS

to do the same on outcome q1_2


Phil


Quoting cthompson@dfpm.utah.edu:

> Greetings STATA-listers,
> I have a dataset examining eye surgeries and I want to examine 
> how responses to certain variables (e.g. q1_1, q1_2, etc.) 
> between each eye vary.  The problem is, my dataset is 
> structured as pasted below:
> 
>      +------------------------------------------+
>      | jmec_no   eye    iol   sex   q1_1   q1_2 |
>      |------------------------------------------|
>   1. |   30332    OD   SA60     F   None   None |
>   2. |   30332    OS   AR40     F   None   None |
>   3. |   46199    OD   SA60     F   None   None |
>   4. |   46199    OS   AR40     F   None   None |
>      +------------------------------------------+
> 
> How do I restructure the dataset so that the Wilcoxon sign rank 
> test can be performed??  That is, I recognize that I need to 
> reshape it or create an array or do some manipulation, but I am 
> at a loss as how to proceed (even after a perusal of the 
> reference manual and STATA help feature) - any suggestions?
> 
> Many thanks,
> Clint Thompson  
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 


-- 
Philip Ryan
Associate Professor
Department of Public Health
University of Adelaide
5005 South Australia
AUSTRALIA
CRICOS Provider Number 00123M
----------------------------------------------------------- 
This email message is intended only for the addressee(s) 
and contains information that may be confidential and/or 
copyright.  If you are not the intended recipient please 
notify the sender by reply email and immediately delete 
this email. Use, disclosure or reproduction of this email 
by anyone other than the intended recipient(s) is strictly 
prohibited. No representation is made that this email or 
any attachments are free of viruses. Virus scanning is 
recommended and is the responsibility of the recipient.
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index