Merging two data sets

[mroy01]

Hi there,

I have a dataset with a number of same id and different observations for another dummy variable.  I hope to make it so that there's one row for each id and the count becomes a sum of all of the observations. Such as

1 1 

2 1

3 0

1 1

4 1

2 1

3 1

4 0

5 0

1 1

Becoming

1 3

2 2

3 1

4 1

5 0

Gratefully, 

SAS 9.3

[Astounding]

SAS knows how to do that:

proc summary data=have nway;

   class id;

   var dummy;

   output out=want (keep=id total) sum=total;

run;

[mroy01]

Thank you for this tip!  I forgot to mention that I'm hoping to keep other variables in the summary as well.  Any ideas to achieve that?

Kindly,

Matt

[Astounding]

You can keep other variables.  But you have to choose from among more than one value.  When you get a sum, that comes from many observations, but only one of the values of those other variables can be saved.

Take a look at MINID and MAXID.  You can select whether you want the minimum or the maximum value of the "other" variables kept in the summary data set.

[kiranv_]

something like below

proc sql;

select id, sum(secondcolumn) as secondcolumn

from yourtable

group by id;