Skip to main content

Table 2 Percentage bias before and after matching procedure

From: Gender gap in STEM pathways: the role of secondary curricula in a highly differentiated school system—the case of Chile

Variable

Status

Treated

Control

% Bias

% Reduced bias

Women

Unmatched

0.484

0.547

− 12.7

88.5

Matched

0.485

0.478

1.5

Socioeconomic status (St)

Unmatched

0.750

0.714

3.3

− 10.7

Matched

0.745

0.784

− 3.7

Ln income per capita (St)

Unmatched

10.794

11.273

− 69.4

97.2

Matched

10.797

10.783

1.9

Father’s education (St)

Unmatched

− 0.480

0.175

− 74.2

96.5

Matched

− 0.476

− 0.499

2.6

Mother’s education (St)

Unmatched

− 0.479

0.180

− 74.3

95.2

Matched

− 0.475

− 0.507

3.5

Attendance 10th grade (St)

Unmatched

− 0.041

0.125

− 17.8

83.0

Matched

− 0.041

− 0.013

− 3.0

GPA 10th grade (St)

Unmatched

− 0.292

0.184

− 50.8

98.3

Matched

− 0.288

− 0.280

− 0.9

SIMCE language (St)

Unmatched

− 0.393

0.224

− 66.7

97.8

Matched

− 0.389

− 0.376

− 1.5

SIMCE math (St)

Unmatched

− 0.465

0.227

− 77.6

99.4

Matched

− 0.462

− 0.466

0.5

School SES (St)

Unmatched

− 0.665

0.155

− 151.6

92.3

Matched

− 0.660

− 0.723

9.6

Rural school

Unmatched

0.052

0.017

19.1

74.7

Matched

0.052

0.060

− 4.8

  1. Ln income per capita = Logarithm of the family income per capita; St indicates that the variable was standardized to a mean of 0 and a standard deviation of 1