Functional Dependency is a property of the primary key. A good FD determines all attributes in relation. A bad FD refers to an attribute that is the key to other attributes in relation. Such sets of attributes are called sub keys of the relation (Jewett, 2006).
Normalization is following a set of rules to ensure that the database has the right design. Normalization rules eliminate unnecessary data in a database. Redundant data results from sub keys and should, therefore, be eliminated. The process of normalization is done in three steps. In the first normal form, repeated and multi-valued attributes are eliminated. The second step involves the removal of sub keys. In the third step, sub keys without primary keys are eliminated (Jewett, 2006).
Contact databases and Phone books exhibit the repeated attribute. They store phone numbers and addresses. Care is taken to prevent leaving fields blank or NULL. In the discussed example, there is one repeated attributed that is the phone. It shows whether the phone number is for work, cell or home. Multi-valued attributes hold additional data for non-formal use such as hobbies. The term hobbies in plural, and entering different hobbies in the same column creates the problem. It creates problems when one wants to query the hobbies in the table or a particular hobby of a person (Jewett,