¹ Committee on Political Parties, American Political Science Association, “Toward A More Responsible Two-Party System,” this Review, 44 (09, 1950), SupplementGoogle Scholar.

² Flinn, Thomas A., “Party Responsibility in the States: Some Causal Factors,” this Review, 58 (03, 1964), 60Google Scholar.

³ Jewell, Malcolm E., “Party Votingin American State Legislatures,” this Review, 49 (09, 1955), 773–791Google Scholar. Jewell chose eight competitive states and limited his research to sessions in which the party balance in the legislature was reasonably close, for he assumed that there would be less party voting even in strong two-party states during the years of one-sided legislative control.He discovered, however, that the three states with the highest levels of party loyalty had long records of Republican legislative control, with only a few Democratic years, while the Democrats frequently held large legislative minorities and (except in Pennsylvania) won control of the governorship. In the states with acloser balance between the parties in the legislature, a lower level of party unity was found. He used all the roll call votes in a session and paid no attention to their sponsorship.

⁴ Lockard, W. Duane, New England State Politics (Princeton: Princeton University Press, 1959)CrossRefGoogle Scholar. In the then essentially non-competitive states of New Hampshire, Maine and Vermont, policy-making was primarily a function of the dominant party in the legislature. The factions within this party, both inside and outside the legislature, determined the fate of legislation. In the competitive states of Massachusetts, Connecticut and Rhode Island, Lockard discovered considerable discipline in both legislative parties. He used votes on which there was disagreement between the parties.

⁵ Wahlke, John C.et al., The Legislative System (New York: John Wiley and Sons Inc.,1962)Google Scholar.

⁶ Derge, David R., “Metropolitan and Outstate Alignments in Illinois and Missouri Legislative Delegations,” this Review, 52 (12, 1958), 1051–1065Google Scholar. Since this study, it has been acknowledged that party is a more important determinant than rural-urban conflict, and the subsequent studies have dealt with the presumed relationship between similarity of district and degree of ideological cohesion within the party contingent.

⁷ Flinn, op. cit., p. 63.

⁸ MacRae, Duncan, “The Relationship Between Roll Call Votes and Constituencies in Massachusetts,” this Review, 46 (12, 1952), 1046–1055Google Scholar. MacRae maintained that those legislators who come from districts which are most typical of their parties tend to show highest party loyalty on roll calls. Those who come from districts atypical of their party tend to cross party lines more often. We do not know whether the roll calls selected by MacRae were administration-supported measures or whether deviations were sizable enough to prevent passage.

⁹ Dye, Thomas R., “A Comparison of Constituency Influences in the Upper and Lower Chambers of a State Legislature,” Western Political Quarterly, 14 (06, 1961),473–481CrossRefGoogle Scholar; Flinn, op. cit., pp. 64–69. Both writers used all party opposition votes whether or not the issues were economic. Dye measured voting deviation and Flinn measured voting support.

¹⁰ MacRae, op. cit.; Dye, op. cit., p. 477; and Patterson, Samuel C., “The Role of the Deviant in the State Legislative System: The Wisconsin Assembly,” The Western Political Quarterly, 14 (06, 1961), 460–472CrossRefGoogle Scholar.

¹¹ Lowell, A. Lawrence, “The Influence of Party Upon Legislation in England and America,” Annual Report of The American Historical Association for 1901, Vol. I (Washington: American Historical Association, 1902), p. 323Google Scholar; and Rice, Stuart, Quantitative Methods in Politics (New York: Alfred A. Knopf, 1928), pp. 209–211Google Scholar.

¹² Riker, William H., “A Method for Determining the Significance of Roll Calls in Voting Bodies,” Legislative Behavior, Wahlke, John C. and Eulau, Heinz (eds.), (Glencoe, Illinois: Free Press, 1959), pp. 377–384Google Scholar. He defined the most significant roll call as one in which (1) all members vote, and (2) the difference between the minority and the majority is the minimum possible under the voting rules.

¹³ Truman, David B., The Congressional Party (New York: John Wiley and Sons, 1959), p. 327Google Scholar.

¹⁴ As a second try, all the governors from 1946–1960 were contacted and asked for lists of program bills from their personal files. In all, 101 governors were contacted and replies were received from 72. Only one governor had such a list: Governor Williams of Michigan sent his worksheet “Priority Points of Democratic Legislators and Governor Williams” for 1960. A considerable amount of strategy was admitted by some governors. Some declared the necessity of including certain items in a message as an aid to secure theenactment of legislation more seriously sought. Capital newspapers could provide the type of information needed. Usually capitol reporters either know specifically what measures are aimed at enacting part of the Governor's program or make their own evaluations in covering the legislative session.

¹⁵ The substantive importance of these vetoes varies. It must be remembered that the governor vetoes frivolous as well as major substantive legislation. For instance, the one veto of the late Governor Stevenson which wasput to a roll call vote in the Illinois Senate in 1949 was S.B. No. 93, an act to provide protection to insectivorous birds by restraining cats. (Illinois, , Journal of the Senate, 1949, p. 541Google Scholar). In the following session of 1951, on the other hand, the one veto roll call was on S.B. No. 102, an act to protect against sub-versive activities. (Illinois, , Journal of the Senate, 1951, pp. 1946–50Google Scholar). Regardless of the substance of the veto, the roll call vote puts on record the degree of party support the governor can muster. Not all veto votes are taken by roll call, but the assumption is that most votes of real controversy are recorded in this way.

¹⁶ As an additional test of the ability of a governor to have his veto upheld against opposition, the following screening procedure was used: (1) 80% or more of the legislature must vote on the veto; and (2) There must be 10% or more of those voting in disagreement. We hoped in this way to reject low-interest, low-tension vetoes. A large percentage of these are technical vetoes. In California, for instance, this method cut 185 out of 189 roll call votes in the seven sessions studied. It seems logical that in a state where so little party voting exists, a governor's veto may be the only check on hasty and ill-advised legislation.

¹⁷ Rice, op. cit., pp. 208–209.

¹⁸ Arizona, , Journal of the House of Representatives, 1949, p. 638Google Scholar.

¹⁹ Truman, op. cit., p. 60. Truman used the arithmetical mean of the legislators' individual administration support indices as the average index for the session.

²⁰ Iowa, , Journal of the House of Representatives, 1953, p. 1234Google Scholar. (H.B. 123, relating to exemptions for debts).

²¹ With respect to measuring inter-party competition see: Golembiewski, Robert I., “A Taxonomic Approach to State Political Party Strength, I” Western Political Quarterly, 11 (09, 1958), 499CrossRefGoogle Scholar; Key, V. O., American State Politics: An Introduction (New York: Alfred A. Knopf, 1956), p. 98Google Scholar; Ranney, Austin and Kendall, Willmoore, “The American Party Systems,” this Review, 48 (06, 1954), 477–485Google Scholar; Ransone, Coleman B., The Office of Governor in the United States (University, Alabama: University of Alabama Press, 1956), pp. 12–94Google Scholar; Schlesinger, Joseph A., “The Structure of Competition For Office in the American States,” Behavioral Science, 5 (07, 1960), 197–210CrossRefGoogle Scholar; and Standing, William H. and Robinson, James A., “Inter-Party Competition and Primary Contesting: The Case of Indiana,” this Review, 52 (12, 1958), 1066–1077Google Scholar. For measures of party cohesion based on primary voting see: V. O. Key, op. cit., pp. 109–118; and Schlesinger, Joseph A., How They Became Governor (East Lansing: Governmental Research Bureau, Michigan State University, 1957), p. 27Google Scholar.

²² Curtis, Richard F. and Jackson, Elton F., “Multiple Indicators in Survey Research,” American Journal of Sociology, 68 (09, 1962), 195–204CrossRefGoogle Scholar.

²³ Ibid., p. 195.

²⁴ The eleven southern states were not included because they did not offer sufficient two-party competition to be directly comparable with the northern states. Nebraska and Minnesota were excluded because their legislatures are elected on a non-partisan basis. New Mexico did not publish a record of the legislative proceedings during this period.

²⁵ The following information was collected for the 140 legislative sessions and processed on an IBM 709 Computer.

²⁶ An additional reason for the high negative correlations between the first two electoral variables and support for the Democrats is that there are high positive correlations between the electoral and house seat measures for the Democrats but not for the Republicans. Thus the variables for elections may be measuring the same phenomena as seats.

²⁷ A good explanation of the technique appears in Blalock, Hubert M., Social Statistics (New York: McGraw Hill Book Company, Inc., 1960), Chapters 17–19Google Scholar. For two applications see Stokes, Donald E., Campbell, Angus and Miller, Warren E., “Components of Electoral Decision,” this Review, 52 (06, 1958), 367–387Google Scholar; and Alker, Hayward R. Jr., “Dimensions of Conflict in the General Assembly,” this Review, 58 (09, 1964), 642–657Google Scholar.

²⁸ The regression equation for explaining a governor'sindex of support (Sup.) was: Sûp. i = K + b₁ G.E._i + b₂ Prim._i + b₃ Seats_i + b₄ Maj._i. In this equation the i subscript may indicate any of the 34 sessions being analyzed. K is an appropriate constant; G.E._i equals the governor's percent of the post-session election; Prim._i, equals the governor's percent of the post-session primary vote; Seats_i equals the percent of seats the governor's party has for the session; Maj._i equals one if the governor's party has 50% or more of the House, zero if it does not. K equals zero when the explanatory variables are standardized. Asimilar equation was used for explaining a governor's index of success (Suc).

²⁹ The states and number of sessions used are: Republican: R.I. (1), Maryland (1), Idaho (1), Oregon (1), Illinois (1), N.H. (1), Mass. (1), Wisconsin (3), Kansas (1), Ohio (2), and Michigan (1). Democratic: Maine (1), Ohio (3), Conn. (1), Iowa (1), Kansas (2), Michigan (4), N.Y. (1), Nevada (1), California (1), Colorado (1), Mass. (2), Montana (1) and R. I. (1). There were 16 sessions in which the governor had a minority of the legislature: (4 Republican: 12 Democratic). There were 18 sessions in which he had a majority: (10 Republican: 8 Democratic).

³⁰ In these equations b-coefficients are given with the independent variables; beta weights are given below them in parentheses. Beta weights are obtained by multiplying a concrete b-coefficient times the standard deviation of its independent variable and dividing by the standard deviation of the dependent variable.

³¹ Multiple correlation represents the zero-order correlation between the actual values obtained for the dependent variable and those values predicted from the equation. If all of the points are on a least-squares line in a scattergram, which is a geometrical interpretation of the equation, the actual and predicted values will coincide and the multiple correlation will be unity. The greater the scatter about the line, the lower the correlation between the actual and predicted values.

³² A way of deciding between alternative models of the causal interrelationships among variables has been elaborated by Blalock, Hubert M., “Correlation and Causality: The Multivariate Case,” Social Forces, 39 (03, 1961), 246–251CrossRefGoogle Scholar. It was used by Miller, Warren E. and Stokes, Donald E., “Constituency Influence in Congress,” this Review, 57 (03, 1963), 45–56Google Scholar.

³³ Flinn, op. cit., pp. 61–66.