Skip to main content

Table 2 Treatment effects (Mahalanobis-Distance-Kernel-Matching). (Source: IEB, own calculations)

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 supporta

Matcheda

ATTb

Inference

Balance (MSB)c

F-testd

 

Exite

Treatment/ type of exit

Nj

Ni

Nj

Ni

 

se

ser/se, I

ser/se, II

Before

After

Before

After

 

Training

All types

1,555

118,236

1,555

32,968

0.006

0.015

1.799

0.818

24.866

2.380

0.000

0.631

0.486

Unempl.

1,555

118,236

1,555

32,968

0.023f

0.014

1.364

1.031

24.866

2.380

0.000

0.631

0.328

Employment

1,555

118,236

1,555

32,968

− 0.013

0.009

1.163

1.020

24.866

2.380

0.000

0.631

0.103

Coaching

All types

7,204

177,573

7,204

27,529

0.002

0.008

2.237

1.623

28.573

0.970

0.000

0.823

0.462

Unempl.

7,204

177,573

7,204

27,529

0.007

0.007

2.166

1.179

28.573

0.970

0.000

0.823

0.331

Employment

7,204

177,573

7,204

27,529

− 0.013g

0.005

1.392

1.060

28.573

0.970

0.000

0.823

0.076

Discr. start-up support (DSUS)

All types

8,942

206,189

8,942

22,033

0.010

0.007

3.633

1.042

24.773

0.885

0.000

0.523

0.487

Unempl.

8,942

206,189

8,942

22,033

0.021g

0.007

2.329

0.888

24.773

0.885

0.000

0.523

0.312

Employment

8,942

206,189

8,942

22,033

− 0.011g

0.005

1.942

1.358

24.773

0.885

0.000

0.523

0.112

  1. aj and i are indicators for the population (i = treated population; j = untreated persons)
  2. bATT stands for the average treatment effect on the treated; the ATT is calculated on the basis of Formula (4): Pr(T k ≤ 36)
  3. cThe balancing property is calculated as the averaged mean standardized bias based on individual and regional variables as well as on the three propensity scores
  4. dThe test used is an F-test of the joint insignificance of all regressors before and after matching
  5. eReports the probability to observe quitting self-employment within a period of 36 months related to the employment status for the matched population
  6. fIndicates statistical significance at the 90 % level
  7. gIndicates statistical significance at the 95 % level