The outcome of coaching and training for self-employment. A statistical evaluation of outside assistance support programs for unemployed business founders in Germany

Oberschachtsiek, Dirk; Scioch, Patrycja

doi:10.1007/s12651-014-0161-6

Journal for Labour Market Research

Table 7 Treatment effects for ‘coaching’ (sub groups and robustness checks)

From: The outcome of coaching and training for self-employment. A statistical evaluation of outside assistance support programs for unemployed business founders in GermanyDie Erfolgswirkung von Gründercoaching und Gründertrainings. Eine Evaluation von Fördermaßnahmen bei Gründungen aus der Arbeitslosigkeit in Deutschland

	On support^a		Matched^a		ATT^b	Inference			Balance (MSB)^c		Test^d		Exit^e
	Nj	Ni	Nj	Ni		se	se^r/se I	se^r/se II	Before	After	Before	After
Single nearest neighbor
All types	10,107	180,283	10,107	8,611	− 0.012	0.008	2.237	1.603	28.604	1.107	0.000	0.003	0.473
Unempl.	10,107	180,283	10,107	8,611	− 0.004	0.007	2.166	1.435	28.604	1.107	0.000	0.003	0.329
Empl.	10,107	180,283	10,107	8,611	− 0.016	0.005	1.392	1.371	28.604	1.107	0.000	0.003	0.0838
Caliper matching
All types	9,393	180,283	9,393	8,347	− 0.006	0.008	2.237	1.389	28.604	0.998	0.000	0.010	0.477
Unempl.	9,393	180,283	9,393	8,347	− 0.003	0.007	2.166	1.426	28.604	0.998	0.000	0.010	0.331
Empl.	9,393	180,283	9,393	8,347	− 0.012	0.005	1.392	1.225	28.604	0.998	0.000	0.010	0.086
Kernel PS-matching
All types	10,107	180,283	10,107	8,611	− 0.012	0.008	2.237	1.603	28.604	1.107	0.000	0.003	0.473
Unempl.	10,107	180,283	10,107	8,611	− 0.004	0.007	2.166	1.435	28.604	1.107	0.000	0.003	0.329
Empl.	10,107	180,283	10,107	8,611	− 0.016	0.005	1.392	1.371	28.604	1.107	0.000	0.003	0.084
All, but excluding information
All types	6,906	177,573	6,906	23,810	0.002	0.008	2.608	1.411	29.724	0.916	0.000	0.919	0.460
Unempl.	6,906	177,573	6,906	23,810	0.011	0.008	2.334	1.071	29.724	0.916	0.000	0.919	0.322
Empl.	6,906	177,573	6,906	23,810	− 0.017	0.005	1.454	1.164	29.724	0.916	0.000	0.919	0.074
Western Germany
All types	2,935	129,836	2,935	18,630	0.014	0.011	1.519	1.022	28.336	0.702	0.000	1.000	0.501
Unempl.	2,935	129,836	2,935	18,630	0.005	0.010	1.335	1.101	28.336	0.702	0.000	1.000	0.301
Empl.	2,935	129,836	2,935	18,630	− 0.007	0.007	1.710	1.007	28.336	0.702	0.000	1.000	0.107
Western Germany excluding regions with high ESF state funding
All types	1,615	34,537	1,615	13,288	0.019	0.015	1.336	0.613	23.574	1.289	0.000	0.999	0.536
Unempl.	1,615	34,537	1,615	13,288	0.016	0.014	1.065	0.872	23.574	1.289	0.000	0.999	30.22
Empl.	1,615	34,537	1,615	13,288	− 0.007	0.009	1.369	1.481	23.574	1.289	0.000	0.999	0.108
Eastern Germany
All types	4,269	47,737	4,269	9,497	− 0.008	0.011	1.931	0.846	17.989	0.754	0.000	0.957	0.426
Unempl.	4,269	47,737	4,269	9,497	0.006	0.010	1.955	0.833	17.989	0.754	0.000	0.957	0.338
Empl.	4,269	47,737	4,269	9,497	− 0.017	0.005	1.088	0.831	17.989	0.754	0.000	0.957	0.0855
Male population
All types	5,405	124,239	5,405	28,342	− 0.002	0.009	2.541	1.684	30.354	1.114	0.000	0.901	0.441
Unempl.	5,405	124,239	5,405	28,342	0.002	0.008	2.120	1.377	30.354	1.114	0.000	0.901	0.333
Empl.	5,405	124,239	5,405	28,342	− 0.010	0.005	1.268	1.247	30.354	1.114	0.000	0.901	0.070
Female population
All types	2,667	43,564	2,667	12,747	0.006	0.012	1.437	1.244	23.458	1.304	0.000	0.965	0.522
Unempl.	2,667	43,564	2,667	12,747	0.018	0.012	1.777	1.365	23.458	1.304	0.000	0.965	0.338
Empl.	2,667	43,564	2,667	12,747	− 0.022	0.008	1.217	0.894	23.458	1.304	0.000	0.965	0.090
Low treatment disposition
All types	340	74,383	340	1,583	− 0.017	0.034	2.237	0.867	15.003	4.011	0.000	0.809	0.497
Unempl.	340	74,383	340	1,583	0.002	0.032	2.166	0.965	15.003	4.011	0.000	0.809	0.332
Empl.	340	74,383	340	1,583	− 0.047	0.021	1.392	0.794	15.003	4.011	0.000	0.809	0.082
High treatment disposition
All types	3,133	31,542	3,133	4,836	− 0.002	0.013	2.237	0.766	12.474	1.243	0.000	0.795	0.422
Unempl.	3,133	31,542	3,133	4,836	0.003	0.012	2.166	0.888	12.474	1.243	0.000	0.795	0.321
Empl.	3,133	31,542	3,133	4,836	− 0.009	0.007	1.392	1.314	12.474	1.243	0.000	0.795	0.064

Table reports selected statistics of the evaluation for subgroups and selected robustness checks; source: IEB, own calculations
^aj and i are indicators for the population (i = treated population; j = untreated persons)
^bATT stands for the average treatment effect on the treated; the ATT is calculated on the basis of formula (4): Pr(T ^k ≤ 36)
^cThe balancing property is calculated as the averaged mean standardized bias based on individual and regional variables as well as the three propensity scores
^dThe test used is an F-test of the joint insignificance of all the regressors before and after matching (prob values)
^eReports the probability to observe quitting self-employment within a period of 36 months related to the employment status for the matched population

Back to article page