> Are you trying to identify all records which violate > your validation rule (4 digits), using the single 9 > simply as an example, or are you trying to > differentiate cases among bad values (eg, single 9 > versus two 9's, or single 8)? > If you are trying to standardize by eliminating > insignificant blanks, and if some of the values to > be processed contain two or three digits and > possibly embedded blanks, use the COMPRESS function. > For example > if compress badnum = '23'; > will dredge not only '23' but '2 3' and ' 23'.

That is precisely what i want to do. As you say, not only I need to scan for values that violate the 4 digit rule and as in your example above but I also need to identify any posible violation of that rule. The purpose is to identify and locate the errots adn if possible clean up and/or massage the data to be used. I dont need to identify blanks.

thank you

Hernan

> >4 digit number entered as a character. > >Due to data entry error the field may have 1, 2, or > 3 > >digits and in any position. ie "9 " or " 9 " or > " > >9 " or " 9". > >I do have a list of those numbers and i need to > know > >what record they belong to but not knowing in what > >position the number or numbers may be I was > wondering > >if there is a simpeler method that can look for the > >number or numbers any of the 4 positions in the > field? > > > >I have been using the following for each position > >which is time consuming. Is there an easier way? > > > > > >data temp; > >set numberslist; > >if badnum ="9 " or badnum = " 9 " or badnum=" 9 > " > >or badnum= " 9"; > >run; > > > > > >Cheers > > > > > > > >__________________________________ > >Do you Yahoo!? > >New and Improved Yahoo! Mail - 100MB free storage! > >http://promotions.yahoo.com/new_mail >

__________________________________ Do you Yahoo!? Yahoo! Mail - 50x more storage than other providers! http://promotions.yahoo.com/new_mail