Skip to main content

Table 1 Sample characteristics: Dependent and independent variables and confounders

From: The effect of co-ethnic social capital on immigrants' labor market integration: a natural experiment

Variables

Mean/Share

(SD)

Sample size

Dependent variables

   

Entered first job

0.86

 

309

Duration until first job

3.10

(3.02)

249

Real hourly wages in first job

6.44

(2.61)

219

Independent variables

   

Co-ethnic community size

0.01

(0.01)

309

Use of social contacts for job search

0.46

 

309

Confounders

   

Female

0.38

 

309

Partner at arrival

0.66

 

309

Children at arrival

0.48

 

309

Age at arrival

29.40

(9.76)

309

Country group of origin a

   

Western, Eastern, and Southeastern Europe

0.13

 

309

Former USSR

0.45

 

309

Asia and the Middle East (incl. Turkey)

0.25

 

309

Africa

0.17

 

309

Years since immigration a

13.06

(6.10)

309

Refugees

0.53

 

309

Ethnic Germans

0.25

 

309

Reason for immigration

   

Political

0.39

 

309

Family

0.20

 

309

Economic

0.28

 

309

Other (including unclear)

0.13

 

309

Pre-migration connections to Germany

0.53

 

309

Educational attainment before immigration

   

Low

0.60

 

305

Medium

0.25

 

305

High

0.15

 

305

(Very) good German language proficiency before immigration b

0.07

 

309

Good math score in school (above median)

0.20

(0.40)

273

Worked before immigration

0.68

 

309

New educational degree

0.03

 

309

Recognition of foreign educational degree

0.04

 

309

Unemployment rate in Germany in the year before immigration

9.38

(1.77)

309

  1. Data source: IAB-SOEP-Migration Sample 2015, own calculations. Design weights are used
  2. Standard deviaton (SD) in parentheses. Variation in the sample size (column 4) is due to the differences in missing data across variables. In the multivariate model, we control for missing values in the variables of interest. a These variables are only presented for illustrative purposes and are not used in the multivariate models (at least in the form presented here). b For the sake of interpretation, German language proficiency is coded 1 if German language proficiency is equal to 4 or above (“good” or “very good”) and is coded 0 otherwise