options ls=80;

* The options statement allows the linesize (ls) to
  be 80 characters wide;

data hospital;

input days progindx gender $;

* DAYS : Number of days spent in the hospital;
* PROGINDX : Prognosis index on how well the patient
             will recover - higher means more likely
             to make a full recovery;
* GENDER: M=Male  F=Female Note: The $ sign is needed
           to indicate that gender is character data;
cards;

2 54 M
5 50  M
7 45 F
10 37 F
14 35 M
19 25 F
26 20 F
31 16 M
34 18 F
38 13 F
45 8 F
52 11 M
53 8 M
60 4 M
65 6 F
;

 *PROC UNIVARIATE calculates summary statistics 
    for each listed quantitative variable;

proc univariate data=hospital;
  var days progindx;

 *Here is a histogram;

 proc univariate data = hospital noprint;
  histogram progindx / midpoints = 0 to 60 by 5 vscale = count ;

* PROC PLOT provides a scatter plot of Y*X;

proc plot data=hospital;
  plot progindx*days;

* PROC CORR provides the correlation between 
   the listed variables;

proc corr;
  var days progindx;

* PROC SORT sorts the data by the given variable;

proc sort; 
  by gender;

* PROC FREQ provides counts of character data.  
    The datasset must be first sorted by the 
    variable that you want counts for (gender);

proc freq;
  tables gender;

proc print;

* PROC PRINT prints the data in the output file 
    (not to the printer!);

run;