Since you have three hashes, I would think you would want three separate loops, unless you are comparing a values to b and c values.

You may also consider reorganizing your data to be in a single hash-of-arrays structure, since your 3 hashes have identical keys. You would still need the same looping for your comparisons, but a single structure would guarantee coherence, and the relationship amongst your data sets would be more apparent.

for (keys %a) 
   print("$a{$_}\n");
   print("$b{$_}\n");
   print("$c{$_}\n");
}
[download]

If you want to loop over your hashes, you can use references:

for( \%a, \%b, \%c ){
   print sort keys %$_, "\n"; 
}
[download]

You can mix all hashes in one
(See above, toolic's node)

#!/usr/bin/perl
use strict;
use warnings;
use Data::Dumper;

my $a = { 1 => 'a', 2 => 'b', 3 => 'c' };
my $b = { 1 => 'd', 2 => 'e', 3 => 'f' };
my $c = { 1 => 'g', 2 => 'h', 3 => 'i' };

my $abc;

for my $h_ref ( $a, $b, $c ){
    for my $key (keys %$a){
        push @{ $abc->{$key} }, $h_ref->{$key};
    }   
}

print Dumper $abc;

__END__

$VAR1 = {
          '1' => [
                   'a',
                   'd',
                   'g'
                 ],
          '3' => [
                   'c',
                   'f',
                   'i'
                 ],
          '2' => [
                   'b',
                   'e',
                   'h'
                 ]
        };
[download]

hth,
PooLpi