Hello ScarletRoxanne, welcome to PerlMonks and Perl!

See #4 on the Basic debugging checklist: print your data structures. For example, if I add use Data::Dump; dd \%anaHash; just after the first loop, and enter Hello and World, the output I get is:

{ dlorw => ["world"], ehllo => ["hello"] }
[download]

Whereas if I enter restful and fluster, I get:

{ eflrstu => ["fluster", "restful"] }
[download]

Now if you look at your second loop, you're iterating over values %anaHash, so I hope it becomes clear why the loop is executing twice?

I'm not exactly sure about what your requirements are, but I'm not sure why you need the second loop? Checking that the phrases are identical could be done with eq on $aWord and $bWord, and whether or not the words are anagrams could be checked by looking at the number of keys in the hash, no?

Some more general tips:

• Instead of #!/usr/bin/perl -w, do use warnings; (What's wrong with -w and $^W Update: Link fixed, thanks Laurent_R!).
• $|=1; I have to use this to make STDIN work. IDK why. - I'm not sure either, $| doesn't have to do with STDIN, I removed it and it worked fine on my end.
• It's better if you don't declare all your variables at the top of the script, because then they're not much better than global variables. Instead, declare them at the smallest scope they are needed, such as foreach my $sortWords (...) or my $sortWord = join '', ...
• $word =~ tr/A-Z/a-z/; - you might want to look at the lc function.
• @words = ($bWord, $aWord); foreach $word (@words) can be written as foreach $word ($bWord, $aWord).
• $aWord = <STDIN>; chomp $aWord; is fine, but is more commonly written as chomp( my $aWord = <STDIN> );.

I can't get the script to print FALSE by, for example, setting a variable to '' or 0, etc. or saying return ''. Ideally, I wouldn't have to put "print TRUE/FALSE" in the script at all.

I'm not sure what you mean by this - Perl doesn't have any "TRUE" and "FALSE" constants like other languages, instead there are simply a couple of values that Perl interprets as being "false" in a "boolean context" (for example, in an if (...) condition) - see Truth and Falsehood. These values don't get converted internally into anything else. If you want Perl to print the strings TRUE and FALSE, you'll need to tell it to do that, such as in print $value?"TRUE":"FALSE".

I'm curious why my m// expression doesn't work.

If I enter hello and hell o, then the regular expressions will be / +hello/ and / +hell o/. Because both of them require spaces before the h, which neither of the input strings have, neither regex will match. I suspect you may have wanted to match on the keys instead, but even then, I think you need to rethink the overall approach, because there would still be further modifications needed to make it work with regexes. (Hint: I would've made a copy of the strings and removed all the spaces.) Another regex tip: If you don't anchor your regular expressions with ^ and $, then remember that the regular expression will match anywhere in the string (e.g. "foobar"=~/a/ is true).

Update 2019-08-17: Updated the link to "Truth and Falsehood".

Discipulus used the following text to restore the OP (thanks Laurent_R) and added a readmore

Here's the script! I'm sorry it's so long

#!/usr/bin/perl
use strict;
use warnings;
use feature 'say';

say "\nReady to play the anagram game? Excellent.\n\nType your first w
+ord or phrase, then hit Enter.\n\n";
chomp (my $aWord = <STDIN>);
say "\n\nThanks! Now type your second word or phrase and hit Enter.\n\
+n";
chomp (my $bWord = <STDIN>);
say "You typed the same word or phrase twice" and exit if $aWord eq $b
+Word;
say is_anagram($aWord, $bWord) ? "True" : "False";

sub is_anagram {
    s/^ +| +$//g for @_;    # remove leading or trailing spaces
    return 0 if length $_[0] != length $_[1]; # words are not anagrams
+ if not the same length
    my @words = map {join '', sort split //, lc} @_;
    return $words[0] eq $words[1];
}
[download]

If the two words are not anagrams then %anaHash will have two keys as the $sortWord will be different for the two words entered. Which means you go around your for each twice for none anagrams and only once if they are anagrams.

Depending on your future plans to expand on this script. You may be able to do away with the loop and replace it with an if which identifies the $sortWord is the same else print not an anagram. If you have a reason for the loop I would suggest moving the that else statement outside the loop and make it an if conditional on a $anagramFound flag populated by the earlier checks.

This was crossposted - also at StackOverflow. :-P