Hi Ben,
An star schema consist of a fact table and dimensions. Basically, a fact table contains values measured and the keys that connect to the dimensions to give meaning to that measures. On the other hand, a dimension is something like a catalog that contains attributes that describe and provide meaning to the fact.
There is another type of fact table called Factless fact table that doesn't contain any measure and are used to keep a track of relations between dimensions. For instance, if you need to track student attendance, you create a record every time a student attends a class.
Your star schema design, depends more on the business process that you're trying to model. In this case, what type of analysis do you want to perform? what data do you have available?, what is the level of granularity in your data?...the answer to this type of questions will help you to determine the design of your data model (star schema)...y recommend you to visit Kimballgroup.com for more information about techniques that you can use in your dimensional modeling design.
Regards,
-- Karla
