Appendix 6.A.2
This appendix replicates Table 6.2 with different cutoff points from the POLITY2 index for democracy. The reader can verify that the results remain essentially highly significant despite the reduction in the number of cases. In addition, statistical significance increases with the substantive significance of amendments.
Table 6.A.2.1 Comparison of three models of effects of constitutional rigidity (null, mean only, and heteroskedastic) on amendment rate for POLITY2 ≥ 6 threshold (ninety-five countries; likelihood ratio tests)
| Significance | Models | Chi-square | p(> Chi-square) |
|---|---|---|---|
| Null vs. mean only | 10.00 | 0.00157 | |
| All amendments | Mean only vs. heteroskedastic | 2.08 | 0.14880 |
| Null vs. heteroskedastic | 12.08 | 0.00238 | |
| Null vs. mean only | 6.60 | 0.01022 | |
| Significant and fundamental | Mean only vs. heteroskedastic | 12.26 | 0.00046 |
| Null vs. heteroskedastic | 18.85 | 0.00008 | |
| Null vs. mean only | 4.19 | 0.04077 | |
| Fundamental amendments | Mean only vs. heteroskedastic | 68.46 | 1.00E+00 |
| Null vs. heteroskedastic | 72.65 | 1.00E+00 |
Table 6.A.2.2 Comparison of three models of effects of constitutional rigidity (null, mean only, and heteroskedastic) on amendment rate for POLITY2 ≥ 7 threshold (eighty-three countries; likelihood ratio tests)
| Significance | Models | Chi-square | p(> Chi-square) |
|---|---|---|---|
| Null vs. mean only | 5.23 | 0.02222 | |
| All amendments | Mean only vs. heteroskedastic | 1.13 | 0.28679 |
| Null vs. heteroskedastic | 6.36 | 0.04151 | |
| Null vs. mean only | 3.74 | 0.05317 | |
| Significant and fundamental | Mean only vs. heteroskedastic | 10.30 | 0.00133 |
| Null vs. heteroskedastic | 14.04 | 0.00089 | |
| Null vs. mean only | 1.93 | 0.16431 | |
| Fundamental amendments | Mean only vs. heteroskedastic | 49.81 | 1.00E+00 |
| Null vs. heteroskedastic | 51.74 | 1.00E+00 |
Table 6.A.2.3 Comparison of three models of effects of constitutional rigidity (null, mean only, and heteroskedastic) on amendment rate for POLITY2 ≥ 8 threshold (seventy-two countries; likelihood ratio tests)
| Significance | Models | Chi-square | p(> Chi-square) |
|---|---|---|---|
| Null vs. mean only | 4.34 | 0.03719 | |
| All amendments | Mean only vs. heteroskedastic | 0.34 | 0.55854 |
| Null vs. heteroskedastic | 4.68 | 0.09613 | |
| Null vs. mean only | 3.10 | 0.07825 | |
| Significant and fundamental | Mean only vs. heteroskedastic | 7.16 | 0.00745 |
| Null vs. heteroskedastic | 10.26 | 0.00591 | |
| Null vs. mean only | 1.55 | 0.21365 | |
| Fundamental amendments | Mean only vs. heteroskedastic | 46.82 | 1.00E+00 |
| Null vs. heteroskedastic | 48.37 | 1.00E+00 |
Table 6.A.2.4 Comparison of three models of effects of constitutional rigidity (null, mean only, and heteroskedastic) on amendment rate for POLITY2 ≥ 9 threshold (fifty-four countries; likelihood ratio tests)
| Significance | Models | Chi-square | p(> Chi-square) |
|---|---|---|---|
| Null vs. mean only | 7.26 | 0.00703 | |
| All amendments | Mean only vs. heteroskedastic | 2.11 | 0.14666 |
| Null vs. heteroskedastic | 9.37 | 0.00923 | |
| Null vs. mean only | 3.53 | 0.06028 | |
| Significant and fundamental | Mean only vs. heteroskedastic | 12.32 | 0.00045 |
| Null vs. heteroskedastic | 15.84 | 0.00036 | |
| Null vs. mean only | 1.47 | 0.22585 | |
| Fundamental amendments | Mean only vs. heteroskedastic | 38.94 | 1.00E+00 |
| Null vs. heteroskedastic | 40.40 | 1.00E+00 |
Table 6.A.2.5 Comparison of three models of effects of constitutional rigidity (null, mean only, and heteroskedastic) on amendment rate for POLITY2 ≥ 10 threshold (thirty-five countries; likelihood ratio tests)
| Significance | Models | Chi-square | p(> Chi-square) |
|---|---|---|---|
| Null vs. mean only | 5.00 | 0.02538 | |
| All amendments | Mean only vs. heteroskedastic | 0.78 | 0.37602 |
| Null vs. heteroskedastic | 5.78 | 0.05554 | |
| Null vs. mean only | 2.84 | 0.09168 | |
| Significant and fundamental | Mean only vs. heteroskedastic | 8.39 | 0.00377 |
| Null vs. heteroskedastic | 11.23 | 0.00364 | |
| Null vs. mean only | 1.88 | 0.17067 | |
| Fundamental amendments | Mean only vs. heteroskedastic | 38.60 | 1.00E+00 |
| Null vs. heteroskedastic | 40.48 | 1.00E+00 |
Open access