Comparison of the Average Critical Point and the Power Function Methods of Adjudging Relative Efficiency of Statistical Tests