Table 13 Summary statistics for the estimated propensity scores, the assessment of common support, and the sample size

From: Wage subsidies targeted to jobseekers with disabilities: subsequent employment and disability retirement

	Propensity score			Outside common support			Number of observations
Analysis	Mean	Min	Max	p=1	p=0	\(p^{1}>p^{0}_{\text {max}}\)	Total	Unique
Main analysis
Participants (D=1)	0.104	0.002	0.717	0	0	1	2739	2738
Non-participants (D=0)	0.040	0.000	0.718	0	469	N/A	61,704	29,515
Sensitivity analysis I^a
Participants (D=1)	0.128	0.002	0.898	0	0	1	2739	2738
Non-participants (D=0)	0.045	0.000	0.821	0	389	N/A	53,010	25,122
Sensitivity analysis II^b
Participants (D=1)	0.069	0.001	0.610	0	0	1	2853	2852
Non-participants (D=0)	0.028	0.000	0.544	0	401	N/A	95,403	41,096
Sensitivity analysis III^c
Participants (D=1)	0.030	0.000	0.740	0	0	0	3010	3008
Non-participants (D=0)	0.003	0.000	0.768	0	2386	N/A	1,134,477	486,029
Sensitivity analysis IV^d
Participants (D=1)	0.099	0.000	0.600	0	0	3	1273	1273
Non-participants (D=0)	0.031	0.000	0.524	0	933	N/A	37,706	18,180
Sensitivity analysis V^e
Participants (D=1)	0.323	0.002	1.000	1	0	16	2,232	2231
Non-participants (D=0)	0.104	0.000	0.935	0	214	N/A	14,818	7091
Sensitivity analysis VI^f
Participants (D=1)	0.092	0.001	0.420	0	0	0	2739	2738
Non-participants (D=0)	0.041	0.000	0.500	0	644	N/A	61,704	29,515

^aThe counterfactual case is changed from “non-participation” to “never-participation”.
^bThe timing of disability coding is ignored.
^cThe disability coding is ignored altogether.
^dJobseekers who were not full-time unemployed at the time of registration at the PES are dropped.
^eBoth participation and non-participation is conditional on being employed in the month of November.
^fLabour market histories earlier than the year immediately preceding the year of (potential) program participation are not included in the estimation of the propensity scores

Back to article page