As I mentioned earlier in this chapter, cross-validation is when a data scientist applies rules to data in a data pool.
Validity checking is the most common form of statistical data cleansing and is a process that both the data developer and the data scientist will most likely be (at least somewhat) familiar with.
There can be any number of validity rules used to clean the data, and these rules will depend upon the intended purpose or objective of the data scientist. Examples of these rules include: data-typing (for example, a field must be a numeric), range limitations (where numbers or dates must fall within a certain range), required (a value cannot be empty or missing), uniqueness (a field, or a combination of fields, ...
