OK, I gave it a try and SAS seems to be ignoring what I tell it. Here is the code I ran:

PROC IMPORT OUT=ace0211 FILE="c:\temp\ace0211.xls" DBMS=EXCEL REPLACE; RANGE='A1:L32765'; GETNAMES=YES; DBDSOPTS="DBTYPE=(Reason_Codes='Char(8)')"; RUN;

The Reason_Codes column is still showing as numeric when I list the columns.

Jeff

On Wed, 28 Sep 2011 08:13:40 -0500, Joe Matise <snoopy369@GMAIL.COM> wrote:

>You can force SAS to read them as text using the DBSASTYPE option. It's >something like this: >proc import file="whatever.xls" data=whatever dbms=excel replace; >dbdsopts=(dbsastype("var1='NUMERIC(8)' var2='CHAR(3)')); >run; > >I don't tend to remember the exact combination of ( and = and " for this, so >the above is probably slightly off, but you can google it and get the right >combination, or when I get into work i'll look at some code I have that does >it properly. > >-Joe > >On Wed, Sep 28, 2011 at 7:44 AM, Jeff Kroll <jeffrey.kroll@us.ing.com>wrote: > >> I have multiple spreadsheets Im importing. On one of the spreadsheets a >> column is all numbers on other spreadsheets its mixed. When I bring these >> together I get an error saying the field has been defined as both numeric >> and character. On the spreadsheet that has the values as all numeric I >> tried formatting the column as text but SAS seems to ignore it. Is there a >> way to force SAS to see the column as text? I even tried inserting a row >> after the heading with X's in that particular column but that didnt work >> either. I have many spreadsheets and dont want to have to edit each of >> them. >> >> Thanks >> >> Jeff Kroll >>