Skip to main content

Table 2 Synthesis of estimates

From: Gender wage gap in China: a large meta-analysis

Study type

Number of estimates (K)

(a) Traditional synthesis

(b) Heterogeneity test and measures

(c) Unrestricted weighted least squares average (UWA)

Fixed-effect model (z value)a

Random-effects model (z value) a

Cochrane Q test of homogeneity

(p value)b

I2 statisticc

H2 statisticd

UWA of all estimates (t value)a,e

Number of the adequately powered estimatesf

WAAP (weighted average of the adequately powered estimates) (t value)a

Median S.E. of estimates (MSE)

Median statistical power (MSP)

All studies (Hypothesis H1)

1472

− 0.092*** (− 354.98)

− 0.122*** (− 52.21)

85,041.040*** (0.000)

99.96

2833.7

− 0.092 (− 46.69)***

1243

− 0.092*** (− 42.90)

0.019

0.998

Target region (Hypothesis H2)

 Urban region

919

− 0.087*** (− 281.08)

− 0.114*** (− 43.43)

40,279.710*** (0.000)

99.93

1341.66

− 0.086*** (− 42.45)

757

− 0.086*** (− 38.65)

0.018

0.997

 Rural region

135

− 0.153*** (− 94.17)

− 0.174*** (− 18.31)

3751.320*** (0.000)

99.71

340.03

− 0.153*** (− 17.80)

126

− 0.153*** (− 17.21)

0.023

1.000

 Region unspecified

418

− 0.101*** (− 200.44)

− 0.124*** (− 26.02)

38,976.320*** (0.000)

99.82

563.87

− 0.101*** (− 20.73)

382

− 0.101*** (− 19.80)

0.018

1.000

Target corporate sector (Hypothesis H3)

 Public sector

24

− 0.080*** (− 23.54)

− 0.090*** (− 10.62)

103.050*** (0.000)

71.86

2.55

− 0.080*** (− 11.12)

16

− 0.078*** (− 10.18)

0.020

0.976

 Private sector

34

− 0.128*** (− 27.51)

− 0.123*** (− 7.72)

366.920*** (0.000)

92.10

11.65

− 0.128*** (− 8.25)

21

− 0.127*** (− 6.32)

0.026

0.998

 Corporate sector unspecified

1414

− 0.092*** (− 353.23)

− 0.122*** (− 51.12)

84,499.000*** (0.000)

99.96

2815.63

− 0.092*** (− 45.68)

1207

− 0.092*** (− 42.20)

0.019

0.999

Target period (Hypothesis H4)

 1990s or before

315

− 0.075*** (− 152.87)

− 0.105*** (− 25.70)

14,735.520*** (0.000)

99.80

490.18

− 0.075*** (− 22.31)

231

− 0.074*** (− 19.40)

0.018

0.989

 2000s

689

− 0.085*** (− 238.58)

− 0.111*** (− 37.60)

31,563.520*** (0.000)

99.95

1971.72

− 0.085*** (− 35.23)

535

− 0.085*** (− 31.08)

0.019

0.993

 2010s

468

− 0.138*** (− 230.70)

− 0.148*** (− 29.95)

31,239.770*** (0.000)

99.95

2081.65

− 0.137*** (− 28.18)

425

− 0.138*** (− 26.86)

0.019

1.000

  1. ***, **, and * denote statistical significance at the 1, 5, and 10% levels, respectively
  2. aNull hypothesis: The synthesized effect size is zero
  3. bNull hypothesis: Effect sizes are homogeneous
  4. cRanges between 0 and 100% with larger scores indicating heterogeneity
  5. dTakes zero in the case of homogeneity
  6. eSynthesis method advocated by Stanley and Doucouliagos (2017) and Stanley et al. (2017)
  7. fDenotes number of estimates with statistical power of 0.80 or more which is computed referring to the UWA of all collected estimates