We propose a sample average approximation (SAA) method for stochastic programming problems involving an expected value constraint. Such problems arise, for example, in portfolio selection with constraints on conditional value-at-risk (CVaR). Our contributions include an analysis of the convergence rate and a statistical validation scheme for the proposed SAA method. Computational results using a portfolio selection problem with a CVaR constraint are presented. Key words: Sample average approximation; Expected value constrained stochastic program; Conditional value-at-risk; Convergence rate; Validation scheme; Portfolio optimization