¹ Kallenbach, Joseph E., The American Chief Executive (New York: Harper and Row, 1966), p. 104 Google Scholar.

² Notable examples include: Key, V. O. Jr., American State Politics: An Introduction (New York: Alfred A. Knopf, 1956)Google Scholar; Jacob, Herbert and Vines, Kenneth N., eds., Politics in the American States (Boston: Little, Brown and Company, 1965)Google Scholar; Fenton, John H., Midwest Politics (New York: Holt, Rinehart, and Winston, 1966)Google Scholar; Lockard, Duane, New England State Politics (Princeton: Princeton University Press, 1959)CrossRefGoogle Scholar; and Dye, Thomas R., Politics, Economics and the Public (Chicago: Rand McNally and Company, 1966)Google Scholar.

³ For example, Campbell, Angus, Converse, Philip E., Miller, Warren E., and Stokes, Donald E., The American Voter (New York: John Wiley and Sons, 1960)Google Scholar. Notable exceptions include: Stokes, Donald E. and Miller, Warren E., “Party Government and the Saliency of Congress,” Public Opinion Quarterly, 26 (Winter, 1962) 531–546 CrossRefGoogle Scholar; Campbell, Angus, “Surge and Decline: A Study of Electoral Change,” Public Opinion Quarterly, 24 (Fall, 1960) 397–418 CrossRefGoogle Scholar. These studies, however, are not concerned with state political contests.

⁴ Although the selection is small, some few state sample surveys have been conducted. A four-state study is analyzed in Public Opinion and Congressional Elections, ed. McPhee, William N. and Glaser, William A. (New York: The Free Press of Glencoe, 1962)Google Scholar. Also, the Comparative State Elections Project of the Institute for Research in Social Science, University of North Carolina, conducted a thirteen-state study during the 1968 national elections.

⁵ Adrian, Charles R., Governing Our Fifty States and Their Communities (New York: McGraw-Hill Book Company, Inc., 1963), p. 112 Google Scholar.

⁶ Stokes, Donald E. and Miller, Warren E., “Party Government and the Saliency of Congress,” in Campbell, Angus, Converse, Philip E., Miller, Warren E., and Stokes, Donald E., Elections and the Political Order (New York: John Wiley and Sons, Inc., 1966), p. 204 Google Scholar.

⁷ The appendix contains a technical description of the construction of these indices.

⁸ For a discussion of the conditions under which the pyramiding of several national samples can be justified, see Cowart, Andrew T., Electoral Decision-Making in the American States: Reconciling Individual Predispositions and Aggregate Patterns (Ph.D. dissertation, University of Michigan, 1971)Google Scholar, chap. 2.

⁹ Campbell et al., The American Voter; Stokes, Donald E., “Some Dynamic Elements of Contests for the Presidency,” American Political Science Review, 60 (March, 1966) 19–28 CrossRefGoogle Scholar.

¹⁰ The question of whether these patterns are indeed linear is an interesting one. Although there are some perturbations on the extremes, I am satisfied with some earlier work in which I utilized a linear multiple regression model to partition the departure of the aggregate vote from 0.5 (the assumed result for evenly matched contests) into two components: one indicating the net contributions of attitudes toward the parties, the other indicating the net contribution of the incumbency advantage. I have since, however, considered an alternative hypothesis which suggests an S-shaped curve for which the effects of the attitude decrease slightly as the attitude becomes more extreme in either partisan direction. A reasonably good fit is obtained with a least squares polynomial of the third degree, and the behavior of the first and second derivatives of the obtained function rule, at various points on the attitude scale, is quite consistent with certain derived consequences of the non-linear theoretical formulation. That work is in progress.

¹¹ See Jennings, M. Kent and Zeigler, Harmon, “The Salience of American State Politics,” American Political Science Review, 64 (June, 1970) 526–535 CrossRefGoogle Scholar. While the findings question the usual low level of saliency attributed to state politics, the saliency advantage which national-level politics holds over state politics is clear from their presentations.

¹² The coefficients are grouped by year rather than by election type. For any particular election, the vote may not be adequately accounted for because of some unmeasured factor peculiar to that election year. Thus, rather than examining whether the magnitudes of all R's for nonincumbent contests exceed those for all incumbent contests, the question is asked whether, for each election year, the R for nonincumbent contests exceeds the R's for the incumbent contests.

¹³ Correlation coefficients computed across different sample subsets of varying size may fluctuate purely as a result of changing magnitudes of the variances of the independent variables. For excessively small N's, the correlation may be quite low as a result of a reduction in variance. It should be noted that there is no correlation between the size of the subsamples and the magnitude of these multiple correlation coefficients. In fact, the lowest N's in 1960 and 1964 for senatorial voting are found in the nonincumbent races where the multiple correlation coefficients are highest. However, the exceptionally low R's for 1968 gubernatorial voting in both incumbent races may have been affected, in part, by reduced variances for the smaller subsamples in the incumbent categories.

¹⁴ Campbell, Angus and Miller, Warren E., “The Motivational Basis for Straight and Split Ticket Voting,” American Political Science Review, 51 (June 1957) 293–312 CrossRefGoogle Scholar.

¹⁵ The procedure for constructing these indices of presidential candidate appeal is the same as that for constructing the four indices of partisan attitude, except that the search is made for free-answer content suggesting a reaction to the personal qualities of the candidates.

¹⁶ Angus Campbell, “Surge and Decline.”

¹⁷ The arguments against the modification of normative theory on the basis of empirical findings are made by several writers in Apolitical Politics: A Critique of Behavioralism, ed. McCoy, Charles A. and Playford, John (New York: Thomas Y. Crowell Company, 1967), especially pp. 185–219 Google Scholar.

¹⁸ Donald E. Stokes, “Some Dynamic Elements of Contests for the Presidency.”