Statistical validation of a surrogate endpoint or surrogate marker has been studied for three decades. However, there is still no consensus on how to evaluate surrogate endpoints or surrogate markers. There is much confusion on which method should be used for the evaluation of a surrogate endpoint or surrogate marker. In this article, we clarify the statistical definitions of the surrogate endpoint and surrogate marker and introduce the concept of the validity and efficiency of a surrogate marker. We suggest the proportion of information gain can be used to evaluate the validity of a surrogate marker. Copyright © 2012 John Wiley & Sons, Ltd.