efficicient coding

[archibald]

Hello,

I have a dataset of patients who reported alcohol consumption during the last  3 months,see below

What would be the most efficient  way of coding the patients into  three categories

1- No alcohol drink during the last 3 months

2-moderate  alcohol consumption ( those who reported drinking 25 glasses per month)

3-Excessive alcohol consumption ( those who reported drinking more than 25 glasses  per month 

Thank you.

data alcohol;
  input ID month1 month2 month3;
  datalines;

30300	0	0	3
30301	3	3	6
30302	0	0	0
30303	10	35	35
30390	20	22	21
30391	0	0	0
30392	9	20	22
30393	67	90	45
30501	12	11	32
30500	4	12	19
30502	2	4	2
30503	1	2	2

;

[PGStats · Accepted solution]

mm = max(of month:);
if mm > 25 then cat = 3;
else if mm > 0 then cat = 2;
else cat = 1;
drop mm;

[archibald]

This is great!! Thank you very much.

One more question, if  the month variable was such as alco_month_1, alco_month_2, and alco_month_last, how can this be  tranformed mm = max(of month:) ?

[PGStats]

mm = max(of alco_month_:);

[archibald]

Thanks!!