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

Re: st: Comparison of two or more observations

From   Philipp Rehm <>
Subject   Re: st: Comparison of two or more observations
Date   Wed, 19 Oct 2005 15:30:04 +0200

You could do this:

collapse (sd) fvc fev1, by(patid)

/* this calculates the standard deviation on each variable of interest for each patient. Clearly, you want the standard deviation to be zero */

drop if fvc==0 & fev1==0

/* this drops all observations with identical entries and leaves only partients in the data-set with mistyped data (assuming, of course, that the data have not been mistyped twice */


Philip Ryan wrote:

Separate your data set into two and use Tom Steichen's -cf3- program, available
using -ssc-


Quoting Seb Buechte <>:

Dear all,

we have been entering lung function data over the course of the last
few month. To be able to identify mistyped data we have entered all
data twice. So for a patient who received a lung function measurement
on 12 Jan 2002 I have two observation holding all the data gained from
that measurement.

I am now wondering wether there is an easy way to compare those two
observations? The data could look like this

patid	lufudate	fvc	fev1
1	10jan2002	5.1	4.8
1	10jan2002	5.1	4.3
2	09jan2002	4.3	3.7
2	09jan2002	4.3	3.7
3	08jan2002	2.9	2.6
3	08jan2002	2.9	6.2

(This is articial data that I just made up. The "real" data contains
approx. 20 parameters)

So is there any program out there that would help me with this task?
Thank you for your suggestions!

Kind regards,

- Seb F Buechte
- Stay tuned!

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index