To many people, surveys are synonymous with political polling and have been since the inception of scientific polls in the 1930s. It is not surprising, therefore, that there exists a considerable amount of over-time data about political subjects. The time series are especially long and complete with respect to party identification and the nation's "most important problem." But very interesting time series also exist with respect to voting for certain kinds of individuals for president on both general and specific questions of governmental responsibilities. An additional concept--liberalism/ conservatism--is mostly a recent discovery, though some early measurements exist.
Here we trace party identification in two series, the first beginning in 1952 and the second in 1972 (Table 1.1).1 For more than twelve years, through 1964, there was relatively little movement in the distribution. The number of "apoliticals" declined as blacks, especially in the South, began to feel more involved in the party system.2 but there were no significant changes in the proportions of Republicans, Democrats, and independents. By 1966, the number of independents____________________