Hi,

All depends. If you have core data and optional data, it may become easier
to write validations rules with one-to-one tables.  As example, assume a
table of employees. Some, but not all, take a life insurance and if they do
so, then there must be a not null value under the field "Beneficiary".

Employees
EmployeeID , ...

LifeInsurance
EmployeeID, Beneficiary, Premium, ...

With that design, make a one-to-one relation from Employees to LifeInsurance
(an employee cannot take two, or more, life insurance) and oblige
Beneficiary, and Premium, to be NOT NULL. In addition, if an employee take a
life insurance, then a record is added in the table, and filled. If not, the
employee is not in the table LifeInsurance, that's all, since no data is
required.

To get the "whole" picture, a simple outer join work fine:

SELECT ...
FROM Employees LEFT JOIN LifeInsurance ON
Employees.EmployeeID=LifeInsurance.EmployeeID

If you have just one table, making a table validation rule about the Premium
cannot be null if there is a Beneficiary is much more complex to formulate,
and to maintain. That is where a one-to-one relation may become usefull.

Hoping it may help,
Vanderghast, Access MVP