Previous research identifies three variables that might bridge the inevitable conflict between the president and Congress: political party, political ideology, and presidential popularity. While the literature provides unambiguous evidence that party and ideology affect presidential support in Congress, the evidence that public approval affects support is mixed. The study reported in this article seeks to clarify the relationship using a research design that corrects some of the limitations of previous work. The analysis reveals that variables within Congress—party and ideology—have the strongest effect on presidential support. Although presidential popularity exerts statistically significant effects, the substantive effects are marginal. Public approval has slightly stronger effects on foreign-policy issues than on economic issues, and the effects are generally stronger on members of the president's party than on members of the opposition. The marginal effects do not increase in strength as we refine the measure of popularity to approach the relevant public for members of Congress.
Get full access to this article
View all access options for this article.
References
1.
1. This list, of course, is not exhaustive. Although other factors such as the president's leadership and the complexity of the policy agenda are no doubt important, we have no way to measure them reliably. Moreover, their effects are likely to be at the margins, in specific cases with particular individuals, rather than general, systematic effects across a representative sample of cases. Because we are seeking to develop general explanations, focusing on party, ideology, and public approval is appropriate.
2.
Paul C. Light , The President's Agenda: Domestic Policy Choice from Kennedy to Carter (Baltimore, MD: Johns Hopkins University Press, 1982).
3.
On the decline of presidential coattails, see John A. Ferejohn and Randall L. Calvert, “Presidential Coattails in Historical Perspective,”American Journal of Political Science, 28:127-146 (Feb. 1984).
4.
4. John Kingdon, Congressmen's Voting Decisions, 2d ed. (New York: Harper & Row, 1981), p. 180.
5.
5. Gary C. Jacobson and Samuel Kernell, Strategy and Choice in Congressional Elections, 2d ed. (New Haven, CT: Yale University Press, 1983).
6.
6. George C. Edwards, “Presidential Party Leadership in Congress,” in Presidents and Their Parties: Leadership or Neglect? ed. Robert H. Harmel (New York: Praeger, 1984), pp. 184-187.
7.
7. Richard Neustadt, Presidential Power: The Politics of Leadership (New York: John Wiley, 1960), p. 187.
8.
8. Edwards, “Presidential Party Leadership,” pp. 191-192.
9.
9. Sidney M. Milkis, “Presidential Party Purges: With Special Emphasis on the Lessons of 1938,” in Presidents and Their Parties: Leadership or Neglect? ed. Harmel.
10.
10. David W. Brady and Phillip Althoff, “Party Voting in the U.S. House of Representatives: Elements of a Responsible Party System,”Journal of Politics, 36:753-775 (Aug. 1974).
11.
Mellisa P. Collie and David W. Brady, “The Decline of Partisan Voting Coalitions in the House of Representatives,” in Congress Reconsidered, 3d ed., ed. Lawrence C. Dodd and Bruce I. Oppenheimer (Washington, DC: CQ Press, 1985). Note, however, that party unity has increased during the Reagan presidency.
12.
12. Morris P. Fiorina, “The Presidency and the Contemporary Electoral System,” in The Presidency and the Political System, ed. Michael Nelson (Washington, DC: CQ Press, 1984).
13.
13. Kingdon, Congressmen's Voting Decisions, p. 271.
14.
14. Jerrold E. Schneider, Ideological Coalitions in Congress (Westport, CT: Greenwood Press, 1979), p. 30.
15.
15. Kingdon, Congressmen's Voting Decisions, pp. 269-270.
16.
16. Richard F. Fenno, Home Style: House Members in Their Districts (Boston: Little, Brown, 1978), p. 144.
17.
John R. Johannes and John C. McAdams, “The Congressional Incumbency Effect: Is It Casework, Policy Compatibility, or Something Else?”American Journal of Political Science, 25:512-542 (Aug. 1981).
18.
18. Kingdon, Congressmen's Voting Decisions, pp. 75-82, 270-271.
19.
19. Bond and Fleisher, “Limits of Presidential Popularity.”
20.
20. Kingdon, Congressmen's Voting Decisions, p. 268.
21.
21. Edwards, Presidential Influence in Congress, p. 88.
22.
22. Neustadt, Presidential Power, p. 86.
23.
23. Roger H. Davidson, The Role of the Congressman (Indianapolis: Pegasus, 1969), pp. 118-119.
24.
the quote is in ibid., p. 109.
25.
Douglas Rivers and Nancy L. Rose, “Passing the President's Program: Public Opinion and Presidential Influence in Congress,”American Journal of Political Science, 29:183-196 (May 1985).
26.
26. Paul C. Light, “Passing Nonincremental Policy: Presidential Influence in Congress, Kennedy to Carter,”Congress and the Presidency, 9:61-82 (Winter 1981-82).
27.
27. Bond and Fleisher, “Limits of Presidential Popularity.” Ideological conflict is the distance between the president and each member on a liberal-conservative continuum, measured as the absolute difference between the president's and each House member's support for Conservative Coalition votes.
28.
28. Ibid., p. 75.
29.
29. Nelson Polsby, Congress and the Presidency, 4th ed. (Englewood Cliffs, NJ: Prentice-Hall, 1986), p. 207.
30.
Another study of presidential-congressional relations that analyzes votes rather than individuals finds that presidential popularity has only marginal effects on whether the president wins or loses. See Bond and Fleisher, “Presidential Popularity and Congressional Voting.”
32. Bond and Fleisher, “Presidential Popularity in Congressional Voting,” p. 294.
33.
Bond and Fleisher, “Limits of Presidential Popularity.”
34.
34. This discussion argues that correlating individual support scores with an independent variable aggregated across individual constituencies to a national average is conceptually the same as correlating averages with averages. It can be shown that the two procedures are mathematically equivalent. The slopes and intercepts of the regressions are identical; the correlations differ depending upon how much variance there is around the mean of the true individual-level variable within each year. As variance around the mean decreases, the mean more accurately reflects the variance of the individual behavior, and the correlations from the two procedures become closer.
35.
35. Edwards, Presidential Influence in Congress, p. 93.
36.
36. Edwards also reports a partial correlation between partisan opinion and partisan support controlling for the party of the president. The partial correlation solves the spuriousness problem, but the interpretation of a partial is not the same as the interpretation of a bivariate correlation.
37.
Edwards reports a correlation of -.48 between Democratic and Republican approval of the president. Edwards, Presidential Influence in Congress, p. 93. This correlation is misleading, however, because it results from change in the partisan evaluations of the president as the presidential party changes over time. Holding president's party constant, partisan opinion varies together, although the opposition party identifiers' approval rate is lower.
38.
38. We did not calculate support scores if fewer than five votes of the policy type occurred during the six-month period. Periods for which support scores could not be computed were dropped from the analysis of that policy type. Because there were more periods with fewer than five economic scores, the analysis of economic support scores is based on a smaller number of cases than is the analysis for foreign policy support.
39.
39. Thomas E. Cronin, The State of the Presidency, 2d ed. (Boston: Little, Brown, 1980), pp. 145-150.
40.
40. George Gallup, The Gallup Opinion Index (Princeton, NJ: Gallup Poll, annually 1959-83).
41.
Bond and Fleisher, “Presidential Popularity and Congressional Voting.”
42.
42. Gallup reports regional breakdowns several times each year. The states included in each category are as follows. The East comprises Connecticut, Delaware, Maine, Maryland, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont, and West Virginia; the Midwest: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, Ohio, North Dakota, South Dakota, and Wisconsin; the South: Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, and Virginia; and the West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming.
43.
Conservative Coalition scores for members of Congress are from Congressional Quarterly Almanac (1959-83). The positions of the Conservative Coalition of Republicans and Southern Democrats are what we typically associate with a conservative stand in American politics. Although CQ does not report a Conservative Coalition support score for the president, we calculated one based on his publicly expressed positions on Conservative Coalition votes. We recognize that this measure introduces an element of circularity into the analysis. Because some presidential support score votes are also Conservative Coalition votes, some of the same votes are used to construct measures on both sides of the equation, which inflates the correlation between ideology and support. Because ideology is relatively stable for members of Congress, however, other measures purged of presidential votes would be highly correlated and would not produce substantially different results.
44.
44. Bond and Fleisher, “Limits of Presidential Popularity.”
45.
45. Neustadt, Presidential Power, p. 87.
46.
Richard Fleisher , “Economic Benefit, Ideology, and Senate Voting on the B-1 Bomber,”American Politics Quarterly, 13:200-211 (Apr. 1985).
47.
47. Significance tests are appropriate in this analysis because the data constitute a random sample. Given the large sample size, however, significance tests are mainly useful as a basis to reject weak coefficients. Coefficients that fail to attain statistical significance can be confidently rejected, but we must be cautious before accepting a statistically significant coefficient as substantively important.
48.
48. Edwards, Presidential Influence in Congress, chap. 4.
49.
49. Ibid.
50.
Aage R. Clausen , How Congressmen Decide: A Policy Focus (New York: St. Martin's Press, 1973), chap. 8.
