**FAP Floor Vote Models and Their Results**
No new explanatory variables seem appropriate for modeling the four 92nd Congress FAP floor votes. LIBERAL, SOUTH, and REPUBLICAN are as in the 91st Congress FAP vote models. EGALITARIAN is also as in the earlier models and so is defined only for the Senate vote. ANTIWAR is the proportion of votes that a member cast for the antiwar anti-administration position of the Vietnam war measures - five each in the House and Senate - that ADA recorded during 1971. Except for the case of REPUBLICAN in the Senate floor votes, the expected signs of the coefficients of each variable are the same as those in the 91st Congress FAP votes. Since Nixon backed neither FAP version in the Senate floor votes, two-sided null hypotheses seem plausible for the coefficients of REPUBLICAN in those two votes.
Table 5 presents results from using maximum likelihood coefficient estimation in single equation multiple logistic regression models, with all explanatory variables entered linearly, of the four 92nd Congress FAP floor votes. Except for that of the coefficient of EGALITARIAN in the second Senate floor vote, once again (descriptively) the sign of each coefficient estimate is what previous narratives and political logic suggest. To interpret the results causally one must again at least assume that the model functional forms are correct, the list of operational variables includes all that are of causal importance, and nominal are indistinguishable from actual p-values.
INSERT TABLE 5 HERE
That the p-values for the 2 tests are < .0001 for all four votes means that, at any conventional -level, variability in explanatory variables explains variability in log (odds of a vote for FAP), in the sense of "explain" that statisticians employ. The coefficients of EGALITARIAN in the Senate votes differ from zero at no conventional a-level, only one of the four ANTIWAR coefficients differs from zero at the .05 -level, and only one of the two one-sided null hypotheses for REPUBLICAN is rejected at the .05 -level. These results suggest that a narrative of the 92nd Congress FAP defeat should probably not stress much the roles of advocates of more re-distributional Senate bills, Vietnam-engendered hostility to Nixon, or political party.
The SOUTH coefficients for the House votes differ from zero at the. 001 level, but for the Senate votes differ from zero at no conventional a-level. These results suggest that a narrative of the 92nd Congress FAP defeat should probably not stress much the role of region in the Senate __floor __votes. That all four of the LIBERAL coefficients differ from zero at the .0005 level suggests that a narrative account of the 92nd Congress FAP defeat should stress the role of political tendency. Since again each explanatory variable ranges from 0 to 1, it again makes sense to informally compare I estimated coefficient I values. Such a comparison estimates LIBERAL has at least 2.5 times the effect on log (odds of a vote for FAP) as the next most influential explanatory variable and at least 4 times the effect of SOUTH. That suggests that a narrative of the 92nd Congress FAP defeat should stress the role of political tendency more than that of region.
**Share with your friends:** |