@cmoore wrote: Hi, I have received an output of data from a survey that was recently sent out. Please see a data example below;   I basically want to reorder the Data by rank order in columns depending what number is in each cell. Any ideas of the best way to do this?   Thanks Chris Have you examined your data so that you can verify that there are never more than 5 "notranked" for any respondent? That none of the respondents ranked any items with the same rank? [ I know that may not have been intended but some surveys / survey software end up allowing such depending on how carefully the data collection staff was trained or the survey script programmed.] Are there any actually missing values that need to treated differently than "not ranked"?   The values you show imply that the variables would have to be character but there is a chance they are actually numeric and "not ranked" is missing and that text applied with a custom format. Is that the case?   If everything has exactly 5 ranked, with full ranks of 1 to 5, and 5 not ranked, and the variables are in the order as shown and adjacent this might get you started: Untested as NO data set provided and I'm not going to make one. data want; set have; array s Capability -- Reputation; array r{10} $15 Rank1-Rank5 NotRanked1-NotRanked5; notcount=0; do = 1 to dim(s); j= input(s[i],f1.); if j>0 then r[j]= vname(s[i]); else do; notcount=notcount+1; r[5+notcount] = vname(s[i]); end; end; keep Respondent_id Rank1-Rank5 NotRanked1-NotRanked5; run; ... View more

I always use connect using these days. So much easier...   libname SQLSLIB odbc database= server= schema=  etc;    proc sql;  connect using SQLSLIB ;  create table XX as select * from connection to SQLSLIB  (  etc  ); ... View more

Hi,   I would be wanting the data to create a dataset and build up the history of all news alerts that come through. I will then use another internal dataset I have to merge together the data. I'm using SAS 9.4.   Thanks Chris ... View more

SAS stands for "Statistic Analysis Systems". Where do you see "scheduler" in that name?   Use the proper tool for your task, which means a scheduling system. ... View more

Have you looked into the FINDW and SCAN functions? ... View more

@cmoore If you've got the SAS Data Quality Server licensed http://support.sas.com/software/products/dataqual/ then you could try code as below. options dqlocale=(ENUSA); data want(drop=_:); set have; /* showTokens=DQPARSEINFOGET('NAME', 'ENUSA');*/ stdName=dqStandardize(name, 'Name'); _parsedValue=dqParse(stdName, 'NAME', 'ENUSA'); nm_prefix=dqParseTokenGet(_parsedValue, 'Name Prefix', 'NAME', 'ENUSA'); nm_given=dqParseTokenGet(_parsedValue, 'Given Name', 'NAME', 'ENUSA'); nm_Middle=dqParseTokenGet(_parsedValue, 'Middle Name', 'NAME', 'ENUSA'); nm_Family=dqParseTokenGet(_parsedValue, 'Family Name', 'NAME', 'ENUSA'); nm_Suffix=dqParseTokenGet(_parsedValue, 'Name Suffix', 'NAME', 'ENUSA'); nm_Appendage=dqParseTokenGet(_parsedValue, 'Name Appendage', 'NAME', 'ENUSA'); run; ... View more