On a Common Mistake in the Description of the Photoelectric Effect