¹ For statements of this argument see, in particular, Key, V. O. Jr.,, Southern Politics in State and Nation (New York: Alfred A. Knopf, 1949), pp. 298–311Google Scholar; Lockard, Duane, New England State Politics (Princeton: Princeton University Press, 1959), pp. 320–340Google Scholar; and Dawson, Richard E. and Gray, Virginia, “State Welfare Policies,” in Politics in the American States: A Comparative Analysis, ed. Jacob, Herbert and Vines, Kenneth N. (Boston: Little, Brown and Company, 1971), p. 461Google Scholar.

² This argument is stated most succinctly in Cnudde, Charles F. and McCrone, Donald J., “Party Competition and Welfare Policies in the American States,” American Political Science Review, 63 (September, 1969), 858CrossRefGoogle Scholar.

³ Ibid.

⁴ See especially Dawson, Richard E. and Robinson, James A., “Inter-Party Competition, Economic Variables and Welfare Policies in the American States,” Journal of Politics, 25 (May, 1963), 265–289CrossRefGoogle Scholar; Dye, Thomas R., Politics, Economics and the Public: Policy Outcomes in the American States (Chicago: Rand McNally and Company, 1966), pp. 143, 144Google Scholar; and Hofferbert, Richard I., “The Relationship between Public Policy and Some Structural and Environmental Variables in the American States,” American Political Science Review, 60 (March, 1966), 73–82CrossRefGoogle Scholar. For an analysis which uses a more extended time period and comes to essentially the same conclusion see Dawson, Richard E., “Social Development, Party Competition, and Policy,” in The American Party Systems: Stages of Political Development, ed. Chambers, William Nisbet and Burnham, Walter Dean (New York: Oxford University Press, 1967), pp. 203–237Google Scholar.

⁵ See, in particular, Sharkansky, Ira, Spending in the American States (Chicago: Rand McNally, 1968)Google Scholar; Sharkansky, Ira and Hofferbert, Richard I., “Dimensions of State Politics, Economics, and Public Policy,” American Political Science Review, 63 (September, 1969), 867–879CrossRefGoogle Scholar; Lockard, Duane, “State Party Systems and Policy Outputs,” in Political Research and Political Theory, ed. Garceau, Oliver (Cambridge: Harvard University Press, 1968), pp. 190–220Google Scholar. Also relevant to this point is Cnudde and McCrone, pp. 858–866.

⁶ For discussions of these conflicting findings and explanations see Fenton, John H. and Chamberlayne, Donald W., “The Literature Dealing with the Relationships between Political Processes, Socio-economic Conditions and Public Policies in the American States: A Bibliographical Essay,” Polity, 1 (Spring, 1969), 388–394CrossRefGoogle Scholar; Jacob, Herbert and Lipsky, Michael, “Outputs, Structure, and Power: An Assessment of Changes in the Study of State and Local Politics,” Journal of Politics, 30 (May, 1968), 511–519CrossRefGoogle Scholar; Fry, Brian R. and Winters, Richard F., “The Politics of Redistribution,” American Political Science Review, 64 (June, 1970), 508–522CrossRefGoogle Scholar; and Sullivan, John L., “A Note on Redistributive Politics,” American Political Science Review, 66 (December, 1972), 1301–1305CrossRefGoogle Scholar.

⁷ Throughout this paper when correlations and differences between correlations are described as significant, I do not mean to imply that these correlations are statistically significant at a given level. This classical notion of statistical inference from sample to population would be inappropriate in this case, since the data came from the population of states, not a sample thereof. Rather, my use of the term “significant” means simply that the correlation or difference between two correlations is not a trivial one; that it is substantively important. The statistics provided in this paper, therefore, are intended to make inferences from the data to specified theoretical notions, not to estimate population parameters.

⁸ Ranney, Austin, “Parties in States Politics,” in Politics in the American States, ed. Jacob, and Vines, , pp. 82–121Google Scholar. The index appears on page 87.

⁹ All index scores above .5000 were subtracted from 1.0000 in order to transform the index from one of degree of Democratic control to degree of party competition, with higher scores representing greater degrees of competition.

¹⁰ Grumm, John G., “The Effects of Legislative Structure on Legislative Performance” in State and Urban Politics: Readings in Comparative Public Policy, ed. Hofferbert, Richard I. and Sharkansky, Ira (Boston: Little, Brown, and Company, 1971), pp. 298–322Google Scholar. The index appears on page 317. Evidence relevant to the construct validity of Grumm's index appears in The Citizens Conference on State Legislatures, State Legislatures: An Evaluation of Their Effectiveness (New York: Praeger Publishers, 1971), p. 70Google Scholar. Their analysis shows that Grumm's index is significantly related to a measure of legislative capability which they devised for each state.

¹¹ Among states high in legislative professionalism, then, these three welfare expenditures (of the seven studied) are the most significantly affected by interparty competition. In this sense, they may be described as the most politically controversial and perhaps publically salient of the policies in the welfare domain. While it may be obvious that aid to dependent children and percentage of public assistance supplied by state and local governments are central elements in the welfare domain, it seems somewhat more difficult to justify the centrality of educational expenditures in terms of the struggle between the haves and have-nots. A possible reason for its centrality in the present case is that the measure of educational expenditures used in this paper does not include capital expenditures but reflects instead, such controversial items as the salaries of teachers, cost of books, special education costs, etc. See Cnudde and McCrone, pp. 859, 862, and 865 for a more extended discussion and analysis of the relation of educational expenditures to the struggle between the haves and have-nots.

¹² See Schuessler, Karl, Analyzing Social Data: A Statistical Orientation (Boston: Houghton Mifflin Company, 1971), p. 24Google Scholar.

¹³ The comparison of beta-coefficients across different samples or different subsets within a sample can be distorted by unequal variances. However, an examination of the unstandardized partial regression coefficients, which are not affected by unequal variances, confirms the interpretation offered here in all essential respects. For a discussion of the use of standardized versus unstandardized regression coefficients see Blalock, H. M. Jr.,, “Causal Inferences, Closed Populations, and Measures of Association,” in Causal Models in the Social Sciences, ed. Blalock, H. M. Jr., (New York: Aldine Atherton, 1971), pp. 139–151Google Scholar.

¹⁴ The states were also trichotomized according to their scores on the legislative professionalism index and relevant correlations were computed between interparty competition and welfare expenditures within each set of states. As would be expected on the basis of findings reported in this paper, the correlations became stronger as the set of states became more professionalized.