Date:         Fri, 29 Aug 2008 10:42:11 -0400
Reply-To:     Arthur Tabachneck <art297@NETSCAPE.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Arthur Tabachneck <art297@NETSCAPE.NET>
Subject:      Re: proc import question
Comments: To: "Mark S." <marksmolkin@GMAIL.COM>

Mark,

I agree with Ron that the problem may well be that the column of concern
has a blank in one of the records.

I usually get around such problems by using the excel libname engine,
rather than proc import, and control what is read by preparing a template
of the data types I want.

For example:

libname xlsIN 'c:\exceldata.xls';

data template;
  format a 2.;
  format b 2.;
  format c 2.;
  input a b c;
  cards;
1 1 1
;

data want;
  set template xlsIN.'sheet1$'n;
  if _n_ gt 1 then output;
run ;

HTH,
Art
--------
On Fri, 29 Aug 2008 10:02:50 -0400, Mark S. <marksmolkin@GMAIL.COM> wrote:

>Hi,
>
>I am trying to learn the basics of proc import and am slowly making
>progress. I often need to work on somewhat dirty spreadsheets in Excel and
>so I learned to use the mixed=yes option in proc import and additionally
>have changed the value of 'typeguessrows' in the Windows registry to a
>very large value...My question: With the latest dataset I am working with,
>there are numerous numeric rows that are correctly getting read into SAS
>as numeric, and, mixed rows appropriately being created as character. For
>some reason, though, there are other seemingly 'numeric' rows in the Excel
>file (without any apparent character values) that are stubbornly being
>forced to character when read into SAS. I can't figure out why. Has
>anybody noticed this too and is there some other attribute in these Excel
>data columns that is forcing SAS to have this effect? Thanks for the
>help....
>
>
>Mark