"For example, if there are only two kinds of textbooks taken in by a system (such as math and history), and the probability of observing a math book is .90, versus .10 for a history book, then there is little uncertainty in the classification of textbooks. On the other hand, if the probability of a math book being observed is .50 and likewise for a history book, then uncertainty is maximum (for when there are two alternatives). An H measure from information theory can be used to indicate the uncertainy in a probability distibution of discrete alternatives." (Simulosophy Group Report, pg.56)
Another good example of this is in the Cabinet of the U.S. President. The President is presented with a wide range of advice and information from his Cabinet members and has to select what information to incorporate into his policy.