Transposing Data without Proc Transpose

ali_hash — Sun, 16 Aug 2015 23:31:16 GMT

Hi there,

I have the following data set (screenshot). What I want to do is to collapse the IDs, Semesters, Adjusted, Raw, and Scores.

Want:

ID Semester Type Score

1 Fall Raw xx

1 Fall Adj xx

1 Spring Raw xx

1 Spring Adj xx

...

I have started something like this...but it doesn't seem to be working out well. And I know the retain has to be there somewhere as well..

data want;

set have;

by id ;

if FIRST.ID THEN DO;

IF FALL_SEMESTER_RAW>0 THEN SEMESTER= 'FALL';

IF SPRING_SEMESTER_RAW>0 THEN SEMESTER= 'SPRING';

ELSE;

*if SPRING_SEMESTER_RAW>0 THEN SEMESTER=SPRING_SEMESTER_ADJ;

END;

RUN;

	FALL_SEMESTER_RAW	FALL_SEMESTER_ADJ	SPRING_SEMESTER_RAW	SPRING_SEMESTER_ADJ
1	89	92.2	93	95.6
2	83	85.6	85	87.8
3	96	98.9	90	93.3
4	93	95.6	96	97.8
5	66	68.9	67	71.1

Re: Transposing Data without Proc Transpose

RW9 — Mon, 17 Aug 2015 12:43:45 GMT

Sorry, could you clarify hy you do not want to use Transpose? Seeing as that procedure is specifically built to transpose data as fast and consisely as possible it makes no sense to not use it:

data have;

input id semester $ type $ score;

id_var=catx("_",upcase(semester),"SEMESTER",upcase(type));

datalines;

1 Fall Raw 89

1 Fall Adj 92.2

1 Spring Raw 93

1 Spring Adj 95.6

2 Fall Raw 83

2 Fall Adj 85.6

2 Spring Raw 85

2 Spring Adj 87.8

;

run;

proc sort data=have;

by id id_var;

run;

proc transpose data=have out=want;

by id;

var score;

id id_var;

idlabel id_var;

run;

topic Re: Transposing Data without Proc Transpose in Statistical Procedures

Transposing Data without Proc Transpose

Re: Transposing Data without Proc Transpose