Delimiting Codes in a Single Variable into Multiple Variables

[art297]

I'm not sure which approach you finally took and knowing that will be necessary to provide the best advice.  If you know have a series of 1 and 0 coded variables, for example, you would probably be best assigning variable labels rather than trying to use a format and only possibly use a format to show the values of 1 and 0 as checked or not checked.

What does your final datafile cuurently look like.  Can you provide a couple of its lines?

[Mgarret]

Thanks Tom and Art.

Tom, your version was perfect for analysis. I can now effectively count the frequencies of all different interventions. This is great!

Art, your version is perfect for manipulating the data to send out to agencies. Please see the version i used below . Thanks so Much!

data Pre.Q60 (keep=Agency_Name Case_No Case_FirstName Case_LastName cra60_);

  set Pre.Presheet1;

  i=1;

  do while (scan(cra60,i,";") ne "");

    cra60_=input(scan(cra60,i,";"),12.);

    output;

    i+1;

  end;

Proc sort data= Pre.Q60;

by Case_No;

proc transpose

data=Pre.Q60

out=Pre.Q60 (drop=_:)

prefix= cra60_;

  var cra60_  ;

  format cra60_ cra60format.;

  by Case_No;

run;

[art297]

I'm still just guessing as to what your resulting file looks like, but I think that the following (or something quite similar) is what you will need to finish the task:

data labels;

  informat response $50.;

  input Code $ Response &;

  cards;

cra60_1  Intensive Home Based Family PreservationServices

cra60_2  Family has been provided Emergency Shelter

cra60_3  Family has been safely moved to a DV Shelter

cra60_4  Caretaker has been moved to a safe environment

cra60_5  Authorization of Emergency Food/Cash

cra60_6  Judicial Intervention

cra60_7  Order of Protection

cra60_8  Law Enforcement involvement

cra60_9  Emergency Medical Services

cra60_10 Crisis Mental Health Services have beenprovided

cra60_11 Emergency In-patient Mental Health Serviceshave been provided

cra60_12 Immediate Supervision/Monitoring

cra60_13 Emergency Alcohol Abuse Services have beenprovided

cra60_14 Emergency Drug Abuse Services

;

proc sql noprint;

  select "label "||Code||'="'||Response||'"'

    into :labels

      separated by " "

        from labels

  ;

quit;

/*What I think you currently have*/

data have;

  input cra60_1-cra60_14;

  cards;

1 0 1 0 1 1 1 1 1 1 0 1 0 1

0 1 1 1 0 0 0 1 1 0 1 1 1 1

;

data want;

  set have;

  &labels.;

run;

[Tom]

The INPUT() function is used to convert the characters in the variable NEXT to a number in the variable I.

The IF statement is so that only numbers will cause SAS to execute the assignment statement.  You might want to make it more specific than just testing for not missing.  Try IF 1 <= I <= dim(cra60_) THEN ....

When I is missing the PUT statement will run and you will see in the log the values of the variables.  The most likely characters that you might see that could have caused the errors you got before are formatting characters like tab,line feed or carriage return. The HEX code representation of those three characters are 09,0A and 0D , respectively.

Here is a link to the manual page for $HEX format.

http://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000205099.htm

[Mgarret · Accepted solution]

Thanks Tom and Art.

Tom, your version was perfect for analysis. I can now effectively count the frequencies of all different interventions. This is great!

Art, your version is perfect for manipulating the data to send out to agencies. Please see the version i used below . Thanks so Much!

data Pre.Q60 (keep=Agency_Name Case_No Case_FirstName Case_LastName cra60_);

  set Pre.Presheet1;

  i=1;

  do while (scan(cra60,i,";") ne "");

    cra60_=input(scan(cra60,i,";"),12.);

    output;

    i+1;

  end;

Proc sort data= Pre.Q60;

by Case_No;

proc transpose

data=Pre.Q60

out=Pre.Q60 (drop=_:)

prefix= cra60_;

  var cra60_  ;

  format cra60_ cra60format.;

  by Case_No;

run;

[Ksharp]

It is another way.

*Get magic number;
data _null_;
input;
retain max;
max=max(max,count(_infile_,';'));
call symputx('max',max+1);
cards4;
case1 1;3;7
case2 2
case3 5;7;12;14
case4 1;2;4
case5 3
case6 6;8;10;12;13
case7 4;8
;;;;
run;


data want;
infile cards dlm=' ;' truncover;
input Case $ (Q16_1-Q16_&max) ($);
cards4;
case1 1;3;7
case2 2
case3 5;7;12;14
case4 1;2;4
case5 3
case6 6;8;10;12;13
case7 4;8
;;;;
run;

Ksharp