* SAS has powerful procedures for reshaping data, such as Proc TRANSPOSE. However, for a 
  situation where multiple adjacent records all belong to the same subject, the following 
  approach can be used to convert data in univarate format into multivariate format ;


data prob3 ;
length agegroup $ 12 ;
input agegroup $ rep time consump1 / agegroup2 $ rep2 time2 consump2 /
  agegroup3 $ rep3 time3 consump3 / agegroup4 $ rep4 time4 consump4 / 
  agegroup5 $ rep5 time5 consump5 ;
 keep agegroup rep consump1 consump2 consump3 consump4 consump5 ;
cards ;
adolescents 1 1 0.83 
adolescents 1 2 0.79 
adolescents 1 3 0.80 
adolescents 1 4 0.78 
adolescents 1 5 0.73 
children 1 1 0.78 
children 1 2 0.76 
children 1 3 0.76 
children 1 4 0.68 
children 1 5 0.67 
youngadults 1 1 0.81 
youngadults 1 2 0.82 
youngadults 1 3 0.78 
youngadults 1 4 0.83 
youngadults 1 5 0.85 
       ;
proc print data = prob3 ; run ;