Re: st: Ungroup grouped data

 From Nikolaos Pandis To statalist@hsphsun2.harvard.edu Subject Re: st: Ungroup grouped data Date Mon, 28 Dec 2009 08:08:26 -0800 (PST)

```Dear Martin,

Many thanks, this was great!

The counts variable shows the original count for every row of covariate (var1 and var2) pattern. How could I change this to show only 0 or 1 depending on absence or presence of the outcome?

For example for covariate pattern 00 we have 7 cases from which we have 4 individuals with the outcome and 3 without. I would like each row to show 0 or 1, in other words for the covariate pattern 00 there should be 4 rows of 1 and 3 of zero value instead of having the value 4 for all rows for this covariate pattern.
I hope I am clear.
Again, many thanks and happy New year.
Nick

--- On Mon, 12/28/09, Martin Weiss <Martin.Weiss1@gmx.de> wrote:

> From: Martin Weiss <Martin.Weiss1@gmx.de>
> Subject: Re: st: Ungroup grouped data
> To: statalist@hsphsun2.harvard.edu
> Date: Monday, December 28, 2009, 5:49 PM
> <>
>
> Here is a reasonably short solution for you! Have a happy
> new year :-)
>
> ***
> clear*
>
> inp byte(counts cases    var1 var2)
> 1    1    0
> 2
> 2    2    1
> 2
> 4    7    0
> 0
> 4    12
> 1    1
> 1    3    2
> 2
> 2    7    1
> 0
> 0    2    2
> 0
> 0    13
> 2    1
> end
>
> //to 47 cases
> expand cases
>
> //outcome must be assigned separately
> bys var*: gen byte outcome=inrange(_n,1,counts)
> la def mylabel 0 "outcome not observed" 1 "outcome
> observed"
> la val outcome mylabel
>
> //get rid of "counts" and "cases"
> drop c*
>
> li, noo sepby(var*)
>
> ***
>
>
> HTH
> Martin
> -------- Original-Nachricht --------
> > Datum: Mon, 28 Dec 2009 07:37:57 -0800 (PST)
> > Von: Nikolaos Pandis <npandis@yahoo.com>
> > An: statalist@hsphsun2.harvard.edu
> > CC: npandis@yahoo.com
> > Betreff: st: Ungroup grouped data
>
> > Hi to all,
> >
> > I have the dataset below which I would like to expand
> and list each
> > observation in one row.
> >
> > The counts variable shows the number of individuals
> with the outcome.
> > The cases variable shows the number of individuals
> that share the
> > covariate pattern for the variables var1 and var2.
> > The sum of cases (47) shows the total number of
> exposed individuals and
> > the sum of counts (14) shows the total number of
> individuals,from the
> > exposed, with the outcome.
> >
> > The fully expanded dataset should list 47 observations
> (=sum of cases).
> >
> > counts    cases
> var1    var2
> > 1    1
> 0    2
> > 2    2
> 1    2
> > 4    7
> 0    0
> > 4    12
> 1    1
> > 1    3
> 2    2
> > 2    7
> 1    0
> > 0    2
> 2    0
> > 0    13
> 2    1
> >
> > Your help would be appreciated.
> >
> > Best wishes,
> >
> > Nick
> >
> >
> >
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
>
> --
> Preisknaller: GMX DSL Flatrate für nur 16,99 Euro/mtl.!
> http://portal.gmx.net/de/go/dsl02
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```