I have the following scenario: $array1 = array("12", "34", "56", "78", "90"); $array2 = array("12", "23", "56", "78", "89"); $result = array_diff($array1, $array2); print_r($result); This returns: Array ( [1] => 34 [4] => 90 ) However what I really want is a two-way comparison. I want elements that don't exist in either to be returned: 34 and 90 because they don't exist in $array2, AND 23 and 89 because they don't exist in $array1. So, is that a two step process of first doing an array_diff($array1, $array2) then reverse it by doing array_diff($array2, $array1) and merge/unique the results? Any caveats with that? $array1 = array("12", "34", "56", "78", "90"); $array2 = array("12", "23", "56", "78", "89"); $diff1 = array_diff($array1, $array2); $diff2 = array_diff($array2, $array1); $result = array_unique(array_merge($diff1, $diff2)); print_r($result); -- A