Question

I have a table with four fields that are usually filled in:

`animal`
- id
- type
- name
- weight
- location

Three additional fields are filled in if the animal type = ‘person’. This happens about 5% of the time. The additional table would be:

`person_additional`
- animal_id (FK)
- IQ
- native_language
- handedness

Is the suggested practice in db design to store this in two tables or one table? It almost makes no difference to me, but I was curious about best practices and why one would be preferable over the other.

Answer 1

Two tables is probably the right approach, but I might suggest a different second table. I would define it as:

`animal_additional`
- animal_id (FK)
- Trait (this would enumerate allowable traits)
- value

This would give you more flexibility in having different traits for different types, or even different traits for the same type.