My results are similar to choroba's. I suspect rocroc is being a bit coy with us about the code or data actually being used because I don't get the same output from the the posted code of Re^3: scoping problem?. I am using a  data file with the OPed data plus a blank line at the end.

I have made minor code changes as follows:

1. The substitution  s/"//g; is changed to  s/\x22//g; (0x22 is the hex value of ASCII " (double-quote)) because my little command-line editor gets very confused by unbalanced double-quotes.
2. I have added single-quotes around the variables in the diagnostic print statements to clearly delineate the strings being printed.

As you can see, the 'agag' record is detected (as well as the blank line, of course).

>perl -wMstrict -le
"my $username;
 my $color;
 while(<>){
     chomp;
     s/\x22//g;
     ($username,$color) = (split /,/,$_)[2,3];
     print STDOUT \"test of username: '$username'\n\";
     print STDOUT \"test of dollar-underscore: '$_'\n\";
     if (\"agag\" =~ m/($username)/){
     print STDOUT \"here is the username: '$username'\n\";
     print STDOUT \"here is dollar-underscore: '$_'\n\";
     }
 }
"  data
test of username: 'adad'

test of dollar-underscore: 'ADELMAN,John,adad,Ray'

test of username: 'agag'

test of dollar-underscore: 'AGAN,John,agag,Aditya'

here is the username: 'agag'

here is dollar-underscore: 'AGAN,John,agag,Aditya'

test of username: 'ahah'

test of dollar-underscore: 'AHMED,John,ahah,Conor'

Use of uninitialized value $username in concatenation (.) or string at
+ -e line 1, <> line 4.
test of username: ''

test of dollar-underscore: '    '

Use of uninitialized value $username in regexp compilation at -e line 
+1, <> line 4.
Use of uninitialized value $username in concatenation (.) or string at
+ -e line 1, <> line 4.
here is the username: ''

here is dollar-underscore: '    '
[download]