Duplicate comments

[CathyVI]

Hi,

I have a  long comment variable and I will like to separate any duplicates comment from non-duplicate comments. I just don't want to use  first.comment because some comment are the same at the initial sentences but later are different.( That is the entire sentence case is not the same). I have provided a sample data. Can I use proc sql if yes please provide a sample code. Thanks

data have;
length comment $ 1000;
input comment$ 1 - 1000;
datalines;
Since it is impossible to completely and completely quit using dependent drugs, it is better to face drug abuse with a positive, rational and practical attitude and point of view
Substance abuse refers to the intermittent or persistent excessive use of psychoactive substances that is contrary to social norms or is irrelevant or inconsistent with accepted medical practice.Such abuse is far from tentative use, social entertainment or situational need, but gradually turns into a state of intensive use, which leads to the formation of dependence.
Substance abuse refers to the intermittent or persistent excessive use of psychoactive substances that is contrary to social norms or is irrelevant or inconsistent with accepted medical practice.Such abuse is far from tentative use, social entertainment or situational need, but gradually turns into a state of intensive use, which leads to the formation of dependence.
Need safe spaces and safe disposable sites
Need safe spaces and safe disposable sites
More attention has been paid to the concept of harm reduction.
NO
No opinion
Marijuana should be legal.
Need safe spaces and safe disposable sites
The use of drugs should be vigorously suppressed
The use of drugs should be vigorously suppressed
Very bad behavior
Need safe spaces and safe disposable sites
Increase control
nothing
nothing
I just hope there will be when I need it
I just hope there will be when I need it
Increase control
No opinion
;
run;

[Tom]

Make a new variable and use it sort/group the data.

Example:

data want;
  set have;
  row+1;
  ucomment = upcase(compbl(comment));
run;

proc sort;
  by ucomment row;
run;

data want;
  set want;
  by ucomment;
  if first.ucomment;
run;

proc sort;
  by row;
run;

proc print;
 var row comment;
run;

Result: