Skip to main content

Table 6 Meta-regression analysis of publication selection bias: All studies

From: Gender wage gap in European emerging markets: a meta-analytic perspective

Estimator

Unrestricted WLS

WLS with bootstrapped standard errors

Cluster-robust

Random-effects panel ML

IV

WLS

 

[1]

[2]

[3]

[4]

[5]

(a). FAT–PET test (Equation: t = γ0 + γ1(1/SE) + v)

 

 Intercept (FAT: H0: γ 0 = 0)

− 6.4514***

(1.038)

− 6.4514***

(0.996)

− 6.4514*

( 3.831)

− 9.4834

( 7.115)

− 3.4118**

(1.734)

 1/SE (PET: H0: γ 1 = 0)

− 0.0875***

(0.013)

− 0.0875***

(0.013)

− 0.0875*

(0.049)

− 0.0569**

(0.027)

− 0.1076***

(0.009)

 K

670

670

670

670

670

 R2

0.4

0.399

0.4

0.4

0.379

Estimator

Unrestricted WLS

WLS with bootstrapped standard errors

Cluster-robust

Random-effects panel ML

IV

WLS

Model

[6]

[7]

[8]

[9]

[10]

(b). EESE approach (Equation: t = γ 0SE + γ 1(1/SE) + v)

 SE

− 111.5465***

(18.967)

− 111.5465***

(16.454)

− 111.547

(68.935)

165.6966***

(31.308)

388.5471

(494.315)

 1/SE (H0: γ 1 = 0)

− 0.0970***

(0.012)

− 0.0970***

(0.011)

− 0.0970**

(0.044)

− 0.0513***

(0.006)

− 0.4378

(0.32)

 K

670

670

670

670

670

 R2

0.524

0.523

0.524

–

-

  1. This table reports the test results for publication selection bias in estimates of the gender dummy variable reported in 53 research works listed in Additional file 1: Table S1 and Additional file 2: Bibliography of studies subject to meta-analysis using the FAT–PET–PEESE procudure. See Table 1 for the descriptive statistics of the estimates. The Appendix describes the methodology applied in this table. Figures in parentheses beneath the regression coefficients are standard errors. Models [3], [4], and [8] report standard errors clustered by study. Models [5] and [10] use the inverse of the square root of the number of observations used as an instrument of the standard error
  2. *** and ** denote statistical significance at the 1% and 5% levels, respectively
  3. aHausman test: χ2 = 0.55, p = 0.4571