How are the names to be "selected"?
How many unique names are there in the data?
What if there are not 10 names available after processing the first XXX number of groups?
If you sort your dataset2 by name you can group them by any number of names desired. Here is a short example with a data set you should have available, SASHELP.CLASS provided by SAS.
proc sort data=sashelp.class out=work.class;
by name;
run;
data work.nameseq;
set work.class;
by name;
if first.name then seq+1;
group = int((seq-1)/10); /* the 10 groups records in to values of 10 based on sequence in the data set)
run;
The sort will get the names together if there are multiple instances.
The BY statement in the second data step adds automatic variables to indicate the first or last of a group of same names. These are accessed using First.variable or Last.variable and have values of 1 when true and 0 when false.
The Seq+1 implies retaining the value across data steps and means that each name gets a unique sequence number. This set does not have repeats but you can test with your data.
The Group calculation will group the names into groups of 10 (except the last group may be smaller if there aren't enough names).
You would use the Group variable with a BY statement for most things after that.
An example of writing each group to a different sheet in excel: (change the drive/path to something on your system)
ods excel file="x:\text.xlsx" options(sheet_interval='BYGROUP');
proc print data=work.nameseq noobs;
by group;
var name sex age height weight;
run;
ods excel close;
You don't say what the XLSX is to be used for. It may be easier to do it ONCE in SAS with a grouped data set before any export.
