Hostname: page-component-586b7cd67f-t7fkt Total loading time: 0 Render date: 2024-11-27T20:56:13.976Z Has data issue: false hasContentIssue false
  • English
  • Français

The House and the Federal Role: A Computer Simulation of Roll-Call Voting*

Published online by Cambridge University Press:  01 August 2014

Michael J. Shapiro
Michael J. Shapiro*
Affiliation:
University of Hawaii
Get access Rights & Permissions [Opens in a new window]

Extract

Much of the business of the U.S. Congress in the post war period has involved issues concerning the size and scope of activities of the federal government. The legislation in this area can be traced, for the most part, to measures which originated during the period of the New Deal in response to the Great Depression and to measures enacted during World War II to meet the short-run exigencies attendant to rapid economic and social mobilization. From the point of view of the expansion of the federal role, the Eisenhower years are of some moment. While they marked a lull in the expansionist trend witnessed under the Democratic presidencies of Roosevelt and Truman, their significance lies in the fact that despite the change in adminsitrations, there was no reversal of the policies begun during the Roosevelt years. While most of the Republican legislators were on record in opposition to the expansion of the federal role, the failure of the Republican Party to introduce and enact legislation to reverse the trend of federal expansion resulted in a new plateau of federal activity from which the congressional dialogue was to proceed during the Kennedy and Johnson Administrations.

While the 87th Congress, meeting during Kennedy's first two years in the White House, did not enact the quantity of legislation expanding the federal role that Kennedy had called for in his inaugural, In the 88th Congress both parties supported a larger federal role to a greater extent than they had previously. In fact the first sessions of the 88th Congress as it bears on the federal role has been summed up as follows: “At no time did the majority of both parties reject a larger federal role.” (Congressional Quarterly Almanac, 1963, p. 724) With two exceptions, the statement holds true for the second session in 1964.

Type
Research Article
Information
American Political Science Review , Volume 62 , Issue 2 , June 1968 , pp. 494 - 517
Copyright
Copyright © American Political Science Association 1968

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

*

The author is indebted to Dr. Cleo Cherryholmes of the Department of Political Science at Michigan State University who shared in the design of the simulation model and rendered valuable advice and assistance in its application reported here. The author wishes, in addition, to thank Dr. Harold Guetzkow and Dr. Kenneth Janda of the Department of Political Science at Northwestern University who aided immeasureably in the conception and execution of the research upon which this article is based.

References

1 The attributes of the bills were determined from their descriptions in the Congressional Quarterly Almanac for 1963 and 1964. The bills chosen for the simulation study are essentially the same as those identified by the Congressional Quarterly Service as pertaining to the federal-role issue area. A few recommital motions on these bills have been added to provide a larger number of bills for purposes of analysis in general and to allow, in particular, for comparisons between bills introduced by the Democratic Administration and those introduced by Republicans (the latter having been reflected, for the most part, in recommital motions).

2 For a number of different perspectives on the effects of personal ideology on legislative roll-call voting, see Brimhall, D. R. and Otis, A. S., “Consistency of Voting in Our Congresses,” Journal of Applied Psychology, 32 (1948), 1–14 CrossRefGoogle Scholar; Farris, Charles “A Method for Determining Ideological Groupings in the Congress,” Journal of Politics, 20 (1958), 308–338 CrossRefGoogle Scholar; MacRae, Duncan Dimensions of Congressional Voting: A Statistical Study of the House of Representatives in the Eighty-First Congress (Berkeley: University of California Press, 1958)Google Scholar; and Froman, Lewis A. Jr. Congressmen and Their Constituencies (Chicago: Rand McNally, 1963).Google Scholar

3 From the standpoint of face validity, the case for the unidimensionality of congressional attitudes on the federal role is a good one. In recent years it has been one of the few issue areas that has consistently distinguished the platforms of the two parties. This consistency has been reflected by the increasing tendency for both casual observers and systematic investigators of politics to employ this issue dimension as a liberal-conservative index. In one case, the use of representatives' past votes on the federal role as a liberal-conservative index has provided an instance of empirical validation of the issue area's unidimensionality (Froman, op. cit.). It was found that congressmen from constituencies associated with conservative voting are low on the index (at the conservative end), while those from constituencies associated with liberal voting are high on the index (at the liberal end). This type of validation is not completely satisfying in as much as Froman's criteria for liberal and conservative voting with which the constituencies were purported to be associated are not entirely independent of the factors or bills that make up the liberal-conservative index employed. A better case for the unidimensionality of representatives' attitudes (inferred from voting behavior) towards the federal role is based upon inferences from the results of Guttman scaling techniques applied to congressional roll-call votes. This scaling technique, based upon rank-ordering procedures, was designed to tap the unidimensionality of attitude clusters. See Guttman, Louis “The Basis for Scalogram Analysis,” In Stouffer, S. A. (ed.), Measurement and Prediction (Princeton, N. J.: Princeton University Press, 1950).Google Scholar MacRae, in his Guttman scale analysis of roll-call votes in the 81st Congress, found what he termed “Fair Deal” and “Welfare State” scales for Democrats and Republicans respectively (op. cit.). Within both of these scales were most of the roll-call votes concerned with the federal role. Anderson obtained similar scales in his analyses of four succeeding congresses. See Anderson, Lee “Variability in the Unidimensionality of Legislative Voting,” Journal of Politics, 26 (1964), 568–585.CrossRefGoogle Scholar Finally, a Guttman scaling analysis was conducted on the 21 bills included in my simulation study. When two bills were eliminated from this analysis, a reproducibility coefficient of .90 was obtained. The two bills that did not scale were not coded for a memory effect in the simulation.

4 After some testing of the model, predisposition weights of plus or minus two were assigned to representatives who had voted for the expansion of the federal role either eighty-five percent of the time or more or fifteen percent of the time or less. Predisposition scores of plus or minus one were assigned for percentages between sixty-five and eighty-five percent and between fifteen and thirty-five percent. Predisposition scores based upon past voting were calculated only for those representatives who had voted on at least ten bills in this issue area. In general, our measure of past voting behavior on the federal role issue dimension was based on the previous two House sessions. All other weights for variables in the predisposition phase of the model are plus or minus one. It can be observed, then, that the predisposition phase of the model provides a theoretical framework for combining propositions about legislative voting behavior that have emerged from more inductive legislative studies. This theoretical framework is based upon the assumptions that predispositions are additive and that all the predisposing variables have approximately equal weight. It is important to note that this last assumption relates to initial predispositions and not to voting behavior. As will be noted below, these variables appear in the communication phase of the model where their effects are not equal.

5 Two of these types of predisposition calculations require some clarification. On the basis of our knowledge of the pervasiveness of party conflict in legislatures I have constructed the model such that a coding of a bill for party sponsorship which predisposes all the members on one party to vote for the bill (adding one to all their predisposition scores) has the effect of predisposing all the members of the other party in the opposite direction (subtracting one from all their predisposition scores). In the case of a bill coded from committees, all members of a committee through which the bill has passed are predisposed for the bill except those representatives who signed the minority report. These are predisposed against the bill.

6 It should be noted that when bills are coded such that some representatives acquire a predisposition as a result of an attribute of their constituency, it is then determined whether or not their constituency is competitive. If the representative comes from a competitive constituency, he acquires an additional unit of predisposition in the same direction. Competitive constituencies were designated as those that had not returned a representative from the same party in the last four consecutive congressional elections. The constituency attributes included in this study were percent urban, percent rural farm, percent rural non-farm, percent negro, percent foreign stock, percent owner occupied dwellings, percent white collar, and degree of competitiveness. With the excpetion of the competitiveness variable, the raw data on these variables was obtained from the Congressional District Data Book, Districts of the 88th Congress, 1963.

7 For evidence with respect to the attitude extremity proposition see Hovland, Carl, Harvey, O. J., and Sherif, Muzafer, “Assimilative and Contrast Effects in Reactions to Communication and Attitude Change,” Journal of Abnormal and Social Psychology, 55 (1957), 244–252.CrossRefGoogle ScholarPubMed The proposition on immunity to influence of representatives with two influences predisposing them in the same direction is suggested by Turner, Julius Party and Constituency: Pressures on Congress (Baltimore: The Johns Hopkins Press, 1951) p. 125.Google Scholar

8 Bauer, Raymond, Dexter, Lewis, and Pool, Ithiel de Sola American Business and Public Policy (New York: Atherton Press, 1964).Google Scholar

9 Wahlke, John, Eulau, Heinz, Buchanan, William and Ferguson, Leroy, The Legislative System: Explorations in Legislative Behavior (New York: Wiley, 1962).Google Scholar

10 Abelson, Robert and Bernstein, A., “A Computer Simulation Model of Community Referendum Controversies,” Public Opinion Quarterly, 27 (1963), 93–122.CrossRefGoogle Scholar

11 See for example Richard Neustadt “Presidency and Legislation,” this Review, 49 (1955), 980–1021.

12 Truman, David, The Congressional Party (New York: Wiley, 1959).Google Scholar

13 Wahlke et al., op. cit., p. 225.

14 Randall Ripley, “The Party Whip Organizations in the United States House of Representatives,” this Review, 58 (1964), 561–576.

15 Wahlke et al., op. cit., p. 226.

16 Ripley, op. cit.

17 See Wahlke et. al., op. cit., p. 222 for data on the dozen basic interactions. The additional interactions are based upon the findings of Truman, op. cit., pp. 145–275.

18 See Routt, Garland “Interpersonal Relationships and the Legislative Process,” Annals of the American Academy of Political and Social Sciience, 19 (1938), 129–136 CrossRefGoogle Scholar, and Wahlke et. al., op. cit., p. 224.

19 See Truman, op. cit., p. 253; Matthews, Donald U. S. Senators and Their World (Chapel Hill: University of North Carolina Press, 1960)Google Scholar; Kessel, John, “The Washington Congressional Delegation,” Mid-West Journal of Political Science, 8 (1964), 1–21 CrossRefGoogle Scholar; Lewis Froman, Jr. and Randall Ripley, “Conditions for Party Leadership: The Case of the House Democrats,” this Review, 59 (1965), 52–63; and Ross, Marc, “Some Correlates of Voting Support for the Leadership in the House of Representatives” (Mimeo, Department of Political Science, Northwestern University, 1965).Google Scholar

20 Routt, op. cit., p. 135; Patterson, Samuel, “Patterns of Interpersonal Relations in a State Legislative Group: The Wisconsin Assembly,” Public Opinion Quarterly, 23 (1959), 101–109 CrossRefGoogle Scholar; Ross, op. cit.

21 Fiellin, Allan “The Functions of Informal Groups: A State Delegation,” in Peabody, R. L. and Polsby, N. W. (eds.), New Perspectives on the House of Representatives (Chicago: Rand McNally, 1963), p. 66 Google Scholar; and Truman, op. cit., p. 210.

22 Truman, op. cit., p. 277.

23 See Wahlke et al., op. cit., p. 224; and Routt, op. cit., p. 135.

24 Truman, op. cit., p. 286; Routt, op. cit., p. 135.

25 Ibid., p. 240.

26 Matthews, op. cit., p. 252.

27 Ross, op. cit.

28 See Goldberg, S. C., “Three Situational Determinants of Conformity to Social Norms,” Journal of Abnormal and Social Psychology, 59 (1954), 325–329 CrossRefGoogle Scholar; Hovland, C. I. and Pritzker, H. A., “Extent of Opinion Change as a Function of Amount of Change Advocated,” Journal of Abnormal and Social Psychology, 54 (1957), 257–261 CrossRefGoogle ScholarPubMed; C. I. Hovland, O. J. Harvey, and M. Sherif, op. cit.; Fisher, S. and Lubin, A., “Distance as a Determinant of Influence in a Two-Person Serial Interaction Situation,” Journal of Abnormal and Social Psychology, 56 (1958), 230–238 CrossRefGoogle Scholar; Zimbardo, P. G., “Involvement, Communication Discrepancy, and Conformity,” Journal of Abnormal and Social Psychology, 60 (1960), 86–94.CrossRefGoogle ScholarPubMed

29 If the sum of the yes and no votes are used as raw scores instead of the splits between the yes's and no's on each vote, correlation coefficients become .95 and .96 respectively. It should be noted, however, that because the product-moment correlation coefficient adjusts for the situation in which the range of the values for one variable is higher than that for the other, the coefficient we have obtained slightly overstates the degree of fit between the simulated and actual roll-call voting outcomes.

30 To determine this variation two bills were run ten times each, one being a bill with a large number of representatives entering the communication phase and the other with relatively few. In each case, the amount of variation in the percentage of representatives correctly simulated did not exceed one percent. Thus any difference in percentages that is two percent or greater can be treated as an actual difference.

31 In keeping with previous findings cited in a number of the works noted above, the structure of the communication process provides for more communicative interactions within the majority party.

32 See Froman and Ripley, op. cit.

33 These percentages are computed for only those twelve bills on which constituency effects are coded.

34 See Oliver Garceau and Corrine Silverman, “A Pressure Group and the Pressured: A Case Report,” this Review, 48 (1954), 672–691; Truman, op. cit.; Froman, op. cit.

35 If this inquiry were directed toward a rigorous examination of such effects as seniority, we could control for confounding effects and operate the model with these controls to obtain these more specific insights into the effects of individual mechanisms.

36 Pool, Ithiel de Sola, Abelson, Robert, and Popkin, Samuel, Candidates, Strategies, and Issues: A Computer Simulation of the 1960 and 1964 Presidential Elections (Cambridge: MIT Press, 1964), p. 64.Google Scholar

37 See Turner, op. cit.; MacRae, op. cit.; Truman, op. cit.; and Froman, op. cit.

38 Turner, op. cit.; Duncan MacRae “The Relation Between Roll-Call Votes and Constituencies in the Massachusetts House of Representatives,” this Review, 46 (1952), 1046–1055; MacRae, op. cit., 1958; Pennock, J. Roland “Party and Constituency in Postwar Agricultural Price Support Legislation,” Journal of Politics, 18 (1956), 167–210 CrossRefGoogle Scholar; Froman, op. cit.

39 Warren Miller and Donald Stokes “Constituency Influence in Congress,” this Review, 57 (1963), 45–56.

40 When we compute the correlation between the actual splits and the “without constituency” splits across only the twelve bills on which constituency was originally coded we obtain a coefficient of .78 as compared with a .95 obtained with the original model for the same bills.

41 See for example H. G. Roach “Sectionalism in Congress (1870–1890),” this Review, 19 (1925), 500–526; Grassmuck, G., Sectional Biases in Congress on Foreign Policy (Baltimore: John Hopkins Press, 1951)Google Scholar; Key, V. O., Southern Politics (New York: Alfred A. Knopf, 1949).Google Scholar

42 When we compute the correlation between the actual splits and the “without region” splits for only the four bills on which region was originally coded we obtain a coefficient of .62 as compared with a .99 obtained with the original model for the same bills.

43 See footnote 2 above.

44 Deutsch, K. W., Singer, J. D., and Smith, K.. “The Organizing Efficiency of Theories,” American Behavioral Scientist, 9 (1965), 30–33.CrossRefGoogle Scholar

45 For a more extensive treatment of the model which includes an assessment of its performance on foreign affairs legislation, see Cherryholmes, Cleo H. and Shapiro, Michael J. Representatives and Roll-Calls: A Computer Simulation of Voting in the Eighty-Eighth Congress (Indianapolis, Indiana: Bobbs-Merrill, 1968)Google Scholar, forthcoming.

Submit a response

Comments

No Comments have been published for this article.