Table 2 Multilevel mixed-effects logistic regression models of the students’ enrolment decision to a master programme.

From: Understanding the influence of business innovation context on intentions of enrolment in master education of STEM students: a multi-level choice model

 

Model I

Model II

β

\(\exp ({\beta}^{\rm{a}} )\)

β

\(\exp (\beta )\)

Gender (women)

Men

0.074

1.07

0.013

1.013

Age

−0.013

0.98

−0.045

0.95

Residence area (urban)

Rural

−0.400*

0.67

−0.392*

0.675

Grade

0.250*

1.28

0.261**

1.29

Father education (Low)

Medium

0.308

1.36

0.248

1.28

High

0.754**

2.12

0.709*

2.03

Subjective income (Easy)

With difficulty

0.203

1.22

0.204

1.22

Working contract (Not working)

Full-time

1.073***

2.92

1.03***

2.80

Part-time

0.374

1.45

0.369

1.44

Seniority

−0.244**

0.78

−0.231**

0.79

Perceived master graduates wage

0.0001

1.00

0.00009

1.00

Perceived master graduates unemployed %

0.006

1.00

0.00438

1.00

Field of study (STEM)

Non-STEM

  

−0.269

0.763

University score

  

0.047

1.05

Constant

−0.165

0.85

0.725

2.06

Observations

476

 

476

 

No. of groups

10

 

10

 

Log-likelihood

−300.89

 

−299.91

 

Variance (Constant)

0.89

 

0.736

 

(Intercept variance)

(standard error)

0.57

 

0.472

 

Variance at University Levelb (%)

21.44%

 

18.28%

 

LR test

30.66***

 

23.10***

 
  1. All coefficients are compared to the benchmark category, shown in brackets. aOdds ratio. bVariance partition coefficient: measures the proportion of the total residual variance that is due to between-group variation.
  2. ***p  < 0.01; **p < 0.05; *p < 0.1.