Skip to main content

Table 7 Ordered logit on improvement in perceived learning and improvement in grades by country, odds ratios estimates

From: The effect of the COVID-19 disruption on the gender gap in students’ performance: a cross-country analysis

Variables

Russian Federation

United Arab Emirates

Slovenia

Uzbekistan

 

IPL

IG

IPL

IG

IPL

IG

IPL

IG

female

0.762***

0.752***

0.832*

0.810***

0.729***

0.868*

0.729***

0.874*

 

(0.065)

(0.050)

(0.078)

(0.060)

(0.067)

(0.068)

(0.060)

(0.064)

age

1.069

1.023

1.123**

1.101

0.999

0.859

1.220**

0.974

 

(0.010)

(0.059)

(0.061)

(0.066)

(0.123)

(0.102)

(0.110)

(0.090)

ses

1.006

0.998

0.978***

0.991*

0.999

1.001

1.002

0.988***

 

(0.005)

(0.005)

(0.005)

(0.005)

(0.004)

(0.005)

(0.005)

(0.005)

lang_istr

0.999

0.775*

1.035

0.909

0.970

1.185

1.313*

1.112

 

(0.135)

(0.101)

(0.098)

(0.089)

(0.165)

(0.164)

(0.191)

(0.154)

he_parents

0.971

1.039

1.187*

1.081

1.189*

0.923

0.952

1.037

 

(0.080)

(0.098)

(0.109)

(0.121)

(0.113)

(0.087)

(0.085)

(0.094)

cut1

0.874

0.154*

0.562

0.380

0.303

0.0164**

5.107

0.0288**

 

(1.151)

(0.142)

(0.475)

(0.314)

(0.534)

(0.028)

(6.419)

(0.040)

cut2

9.385

1.073

2.818

1.915

2.353

0.010

28.890***

0.160

 

(12.23)

(0.987)

(2.372)

(1.589)

(4.141)

(0.169)

(36.080)

(0.217)

cut3

38.040**

4.521

12.060***

9.710***

10.270

0.535

115.3***

0.952

 

(50.310)

(4.155)

(10.440)

(8.143)

(18.130)

(0.907)

(145.300)

(1.299)

Pseudo R2

0.002

0.003

0.006

0.004

0.004

0.001

0.005

0.002

Observations

3411

3399

2695

2685

2388

2381

2878

2873

  1. ***Indicates significance at the 1% level, ** at the 5% level and * at the 10% level. IPL indicates the dependent variable improvement in perceived learning, while IG the improvement in grades. Coefficients are reported in odds ratios. Jackknife standard errors in parentheses. Results are adjusted for survey setting with 75 Jackknife replications. Pseudo R2 is not provided for survey setting. The values of Pseudo R2 are taken from simple ordered logistic models, without survey parameters