SYMGET
in [SAS]
|
Prev: ODS Tagsets ExcelXP
Next: Help - Gini's Coefficient
From: Gaetan on 19 May 2006 03:52 Hello, excuse me for the language, I am french. I have a trouble : I have a colum in a data base and with values of this column, I would like to create new var. ex : Name Sexe John Man Rita Woman I would like to have in my database : Name Sexe Man Woman John Man 1 0 Rita Woman 0 1 My trouble is : I have lots of differents values like (men, women, men1, men2, men3 , ...) I have tried with macos functions SYMGET, SYMPUT to get automaticely the values and create new columns bit I don't manage because the row : symget("var")=1 doesn't work. I hope you can understand what I say. Thank you Gaétan
From: Eric Eyre on 19 May 2006 05:03 Bonjour Gaetan Je suis familier avec [men, women] mais que sont les [men1, men2, men3, etc] ? Un nouveau type de genre? SVP aidez-nous ? comprendre ce nouveau et difficile concept...vraiment, vraimant, comme totalement pour s?r dude Erique "qui est exact, je suis une personne de maths, je ne suis pas un de 4 sur 3 personnes qui ont des ennuis avec des fractions" "Gaetan" <gaetan.soulas(a)ifen.fr> wrote in message news:1148025175.083257.322680(a)i39g2000cwa.googlegroups.com... Hello, excuse me for the language, I am french. I have a trouble : I have a colum in a data base and with values of this column, I would like to create new var. ex : Name Sexe John Man Rita Woman I would like to have in my database : Name Sexe Man Woman John Man 1 0 Rita Woman 0 1 My trouble is : I have lots of differents values like (men, women, men1, men2, men3 , ...) I have tried with macos functions SYMGET, SYMPUT to get automaticely the values and create new columns bit I don't manage because the row : symget("var")=1 doesn't work. I hope you can understand what I say. Thank you Ga?tan
From: Joep on 19 May 2006 05:15 Gaetan, You may want to look into Proc Transpose: data tall; length name $ 10; name = 'Jean'; sexe = 'm'; output; name = 'Jeannette'; sexe = 'f'; output; name = 'Pierre'; sexe = 'm'; output; name = 'Catherine'; sexe = 'f'; output; run; proc sort data=tall; by name; run; proc transpose data= tall out=wide (drop = _name_); by name; id sexe; var sexe; run; Although this example does not exactly give you your desired output. Another thing is, do you really want to transpose datarows in a column into several columns. At present I'm actually doing the opposite, because the data structure of columns representing data elements is to rigid. If a data element is added then I have to alter the datastructure, which I think is not a good idea. In my new data structure, I just add a new row. Hope this is helpful, Joep
From: Gaetan on 19 May 2006 06:15 En fait, je vais donner une explication beaucoup plus simple : je pense que la meilleure solution pour moi est de passer par un PROC TABULATE mais un nouveau probleme se pose : J'aimerai récupérer à la suite de la PROC TABULATE , une table "resultat" de la meme forme que dans la fenêtre SORTIE, c'est à dire un tableau croisé dynamique. Parce que lorsque j'utilse l'option OUT = resultat de la PROC TABULATE , la table "resultat" n'a pas la forme d'un tableau croisé dynamique. Merci
From: Venky Chakravarthy on 19 May 2006 08:54
On Fri, 19 May 2006 00:52:55 -0700, Gaetan <gaetan.soulas(a)IFEN.FR> wrote: >Hello, > >excuse me for the language, I am french. > >I have a trouble : > >I have a colum in a data base and with values of this column, I would >like to create new var. > >ex : > >Name Sexe > >John Man >Rita Woman > >I would like to have in my database : > >Name Sexe Man Woman > >John Man 1 0 >Rita Woman 0 1 > >My trouble is : I have lots of differents values like (men, women, >men1, men2, men3 , ...) > >I have tried with macos functions SYMGET, SYMPUT to get automaticely >the values and create new columns bit I don't manage because the row : > >symget("var")=1 > >doesn't work. > >I hope you can understand what I say. > >Thank you > Ga?tan Gaetan, Can you present a small sample with the data that you HAVE |