Riker's method for assessing the significance of roll call votes

Abstract

In the course of various studies of the political factors influencing the outcomes of monetary policy decision making in the United States, (2) the author has attempted to apply the ingenious method devised a few years ago by William H. Riker to assessing the published record of voting outcomes of the nearly two hundred decisions on current economic policy made in the last decade by the Federal Open Market Committee of the Federal Reserve System, especially the roughly one-fourth of the decisions in which dissents have occurred. It has been useful to generalize Riker's work by uncovering the properties of characteristic matrices of decisive voting outcomes and the relationships between these matrices and Riker's coefficient of significance of roll call votes. Several sources of a priori bias in Riker's specific measure of significance have been brought to light, particularly when the measure is used to make comparisons for the same or different voting bodies where any of the Riker coefficients lie at the high or low ends of his scale or where the differences among the coefficients is relatively small.