The political implications of state political ideology: a measure tested

American Journal of Economics and Sociology, The,  April, 1997  by Marshall H. Medoff

• E-mail
• Print
• Link

3. The Americans for Democratic Action records a "no" vote, if a representative did not vote. To avoid penalizing representatives for failure to vote the ADA scores were recalculated as the percentage of times the representative voted in the direction supported by the ADA only for those votes the representative was present. The votes were cast in 1993 and supplied by Americans for Democratic Action. The correlation between the ADA rating and the conservative group Americans for Constitutional Action rating is .96. Thus the choice of a rating group makes little difference.

4. On election day November 8, 1994 the Democrats lost 50 congressional representative seats to Republicans. Over half these losses were in eleven states located in the South and Southwest. According to the Americans for Democratic Action the losses in those states were conservative Democrats being replaced by conservative Republicans. In the other states moderate Republicans replaced moderate Democrats. Thus the relative state political ideology scores estimated in this paper should be robust even with the change in representation. When a state had only one representative the ADA scores of the state's Senators were included.

5. The state ideology scores and rankings remain invariant whether the weights used are eligible voters, registered voters, or actual voters in a congressional district.

6. All the economic variables for each congressional district were obtained from the U.S. Census of the Population: 1990, Congressional Districts of the 103rd Congress.

7. The empirical results are (t-statistics in parentheses):

[ADA.sub.i] = 5.5649 + .0256 %Manufacturing + .0037 %Urban (4.62) (2.41) (.92) + .1371 %Unemployment - .1271% Married, [R.sup.2] = .36. (3.69) (-7.63)

8. If congressional representative i is female (F), black (B), Catholic (CATH), Jewish (J), or Fundamentalist Christian (FUND) the respective variable was set equal to one and zero otherwise. [All the personal data on congressional representatives were obtained from Barone and Ujifusa (1993).] The empirical results are (t-statistics in parentheses):

[ADA.sub.i] = 4.918 + .0294 %Manufacturing - .0018 %Urban + .1363 %Unemployment (4.18) (2.82) (-.49) (3.58) - .1151 %Married + .8371 F + .8251 B + 1.7052J (-7.21) (3.57) (2.34) (5.07) + .3482 CATH - .5004 FUND, [R.sup.2] = .45. (2.35) (-2.36)

9. The empirical results are (t-statistics in parentheses):

[ADA.sub.i] = 3.6786 + .0096 %Manufacturing + .0059 %Urban - .0323 %Unemployment (4.36) (1.24) (2.04) (-1.15) - .0582 %Married + .5649 F + .8211 B + .8491 J + .2773 CATH (-5.24) (3.08) (3.09) (3.36) (2.36) - .5828 FUND - 2.6345 Republican, [R.sup.2] = .81. (-4.09) (-20.44)

10. The mean ideological scores between the Northeast and Pacific regions were not statistically significantly different from zero.

11. The mean ideological scores between the Mountain and South regions were not statistically significantly different from zero. The null hypothesis of equality of ideological mean scores between either of the two liberal regions (Northeast, Pacific) and either of the two conservative regions (Mountain, South) was rejected at the .05 level of significance.