I am using the ECR-Scale to assess the attachment styles of university students. I would like to use the secure attachment as a moderator. Can I measure secure attachment by reversing coding the items (except the items already reverse coded) of the anxious and avoidant?