topic Re: how to get the mean in the form of diagonal format to show that in the wide format? in SAS Programming

how to get the mean in the form of diagonal format to show that in the wide format?

bhr-q — Mon, 23 Sep 2024 21:37:38 GMT

Hello all,

Many thanks for helping me to convert the first design to the second design.

Id B w g

1 8 . .

1 7 . .

1 . 6 .

1 . 5 .

1 . . 4

1 . . 5

2 4 . .

2 3 . .

2 . 6 .

2 . 2 .

2 . . 8

2 . . 4

Second design: the number here is the mean number of each event (B,w,g ) for each id_recoded

Id B w g

1 7.5 5.5 4.5

2 3.5 4 6

Thank you in advance

Re: how to get the mean in the form of diagonal format to show that in the wide format?

ballardw — Mon, 23 Sep 2024 21:46:23 GMT

Proc summary data=have nway;
   class id;
   var b w g;
   output out=want (drop= _: ) mean=;
run;

Looks likely.

The NWAY means that only the highest level of the _type_ variable goes into the output.

The drop is because otherwise two additional variables, _type_, indicating combinations of the CLASS variables, and _freq_ would appear.

Mean= used this way with this procedure says "name the statistic mean with the original name of the variable". You could rename the results if desired.

Re: how to get the mean in the form of diagonal format to show that in the wide format?

data_null__ — Mon, 23 Sep 2024 21:46:33 GMT

data test;
   input Id B w g;
   cards;
1             8        .        .
1             7        .        .
1             .        6        .       
1             .        5        .
1             .        .         4 
1             .        .         5
2           4        .          .
2           3        .          . 
2           .         6          .
2           .         2         .
2          .          .          8
2           .        .           4
;;;;
   run;
proc print;
   run;
proc summary data=test nway;
   class id;
   output out=means mean(b w g)=;
   run;
proc print;
   run;

Re: how to get the mean in the form of diagonal format to show that in the wide format?

bhr-q — Tue, 24 Sep 2024 01:43:37 GMT

Thanks so much, it was helpful