A Scalogram Analysis of Election Returns

In this paper the author uses Guttman's technique, which has been called “scalogram analysis,” to show that there is one major dimension in election returns. The data submitted to analysis are Quebec provincial elections returns, by constituencies, from 1936 to 1966. Election years and constituencies are ranked in such a way as to form the best scale, that is to say, the scale where the number of “errors” is minimized. As a matter of fact, this number is small enough, in the present case, to conclude that election returns may be explained by one very major dimension. Moreover, many of the errors are quite significant. But the problem is to determine what is the major dimension in the scale. For that purpose, the author isolates four dimensions: the socio-economic, the partisan, the political, and the personal. After reviewing each of them, he comes to the conclusion that no dimension other than the political one can be the major dimension in the scale. It is to be noted that the late V. O. Key, Jr., came to the same conclusion, by another way, in his The Responsible Electorate. The last part of the paper contains some remarks on the limits of scalogram analysis, as applied to election returns, plus some clues on the practical advantages in such analysis for political parties.

The Comparative Analysis of Data

For several years political scientists have been studying the possibilities of using computers to advance the study of comparative politics. Some have urged that the development of a truly empirical political theory would benefit from the use of computers on a massive scale. This proposition is explored through careful examination of data assembled under the direction of Banks and Textor and published as A Cross-Polity Survey. This discussion is placed in the context of recent developments in the field of comparative politics. It is argued that the term “comparative politics” must be understood more clearly before one can estimate the advances that may be made in this field through the use of computers.