Table 5.
Check the sensitivity of results to unobservable selection.
| Rosenbaum bounds for delta (N = 783 matched pairs) income inequality |
Rosenbaum bounds for delta (N = 682 matched pairs) educational inequality |
||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gamma | sig+ | sig- | t-hat+ | t-hat- | CI+. | CI- | Gamma | sig+ | sig- | t-hat+ | t-hat- | CI+. | CI- |
| 1 | 0 | 0 | −7.665 | −7.665 | −8.635 | −6.695 | 1 | 0 | 0 | .145,135 | .145,135 | .119,655 | .1727 |
| 1.05 | 0 | 0 | −7.96 | −7.37 | −8.935 | −6.4 | 1.05 | 0 | 0 | .13,786 | .15,281 | .1126 | .180,815 |
| 1.1 | 0 | 0 | −8.24 | −7.09 | −9.22 | −6.135 | 1.1 | 0 | 0 | .13,106 | .160,087 | .105,924 | .18,853 |
| 1.15 | 0 | 0 | −8.505 | −6.825 | −9.495 | −5.865 | 1.15 | 0 | 0 | .12,453 | .16,725 | .09956 | .195,955 |
| 1.2 | 0 | 0 | −8.76 | −6.57 | −9.77 | −5.62 | 1.2 | 0 | 0 | .118,393 | .174,145 | .093,545 | .203,155 |
| 1.25 | 0 | 0 | −9.01 | −6.335 | −10.02 | −5.385 | 1.25 | 0 | 0 | .112,565 | .18,085 | .087,695 | .21,017 |
| 1.3 | 0 | 0 | −9.245 | −6.11 | −10.27 | −5.15 | 1.3 | 0 | 0 | .106,978 | .18,724 | .08246 | .216,685 |
| 1.35 | 0 | 0 | −9.475 | −5.885 | −10.505 | −4.935 | 1.35 | 1.1e-16 | 0 | .101,575 | .193,395 | .077,404 | .223,105 |
| 1.4 | 0 | 0 | −9.71 | −5.675 | −10.73 | −4.72 | 1.4 | 1.8e-15 | 0 | .096,486 | .1995 | .07271 | .229,105 |
| 1.45 | 0 | 0 | −9.92 | −5.475 | −10.955 | −4.515 | 1.45 | 3.1e-14 | 0 | .09151 | .20,537 | .068,071 | .23,507 |
| 1.5 | 0 | 0 | −10.13 | −5.275 | −11.165 | −4.32 | 1.5 | 4.4e-13 | 0 | .086,945 | .211,135 | .06365 | .24,109 |
| 1.55 | 0 | 0 | −10.335 | −5.085 | −11.365 | −4.135 | 1.55 | 4.9e-12 | 0 | .08265 | .216,495 | .059,625 | .24,671 |
| 1.6 | 0 | 0 | −10.53 | −4.91 | −11.575 | −3.95 | 1.6 | 4.5e-11 | 0 | .07852 | .221,775 | .055,587 | .252,215 |
| 1.65 | 0 | 0 | −10.715 | −4.735 | −11.775 | −3.775 | 1.65 | 3.4e-10 | 0 | .074,565 | .22,678 | .051,656 | .25,758 |
| 1.7 | 0 | 0 | −10.9 | −4.565 | −11.965 | −3.605 | 1.7 | 2.2e-09 | 0 | .070,776 | .23,162 | .047,903 | .262,955 |
| 1.75 | 0 | 1.1e-16 | −11.08 | −4.4 | −12.15 | −3.44 | 1.75 | 1.3e-08 | 0 | .066,965 | .236,535 | .044,095 | .268,212 |
| 1.8 | 0 | 1.0e-15 | −11.25 | −4.245 | −12.33 | −3.275 | 1.8 | 6.2e-08 | 0 | .063,415 | .241,445 | .04066 | .27,285 |
| 1.85 | 0 | 9.7e-15 | −11.42 | −4.095 | −12.5 | −3.125 | 1.85 | 2.7e-07 | 0 | .06016 | .245,972 | .03733 | .27,756 |
| 1.9 | 0 | 8.1e-14 | −11.59 | −3.94 | −12.67 | −2.975 | 1.9 | 1.1e-06 | 0 | .056,845 | .25,046 | .033,996 | .28,222 |
| 1.95 | 0 | 5.9e-13 | −11.75 | −3.795 | −12.84 | −2.83 | 1.95 | 3.7e-06 | 0 | .053,715 | .254,816 | .03077 | .28,684 |
| 2 | 0 | 3.8e-12 | −11.91 | −3.655 | −13.005 | −2.685 | 2 | .000012 | 0 | .050,605 | .25,916 | .02769 | .291,175 |
*gamma - log odds of differential assignment due to unobserved factors.
Sig + -upper bound significance level.
Sig--lower bound significance level.
t-hat-upper bound Hodges-Lehmann point estimate.
t-hat-lower bound Hodges-Lehmann point estimate.
CI+.-upper bound confidence interval (a = 0.95).
CI--lower bound confidence interval (a = 0.95).