Inter-Method Reliability of Progression Sizes in a Hypothetical Purchase Task: Implications for Empirical Public Policy