Skip to main content

Table 3 Name, definition, and descriptive statistics of meta-independent variables

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

Variable name

Definition

Descriptive statistics

Mean

Median

S.D

2007 and 2013 member

1 = if the target country joined the EU either in 2007 or 2013, 0 = otherwise

0.127

0

0.333

Membership candidate

1 = if the target country is a candidate for EU membership, 0 = otherwise

0.028

0

0.166

Nonmember

1 = if the target country is a non-EU member state, 0 = otherwise

0.449

0

0.498

Average estimation year

Average estimation year

2001

2002

7.193

Urban region

1 = if the target region is urban, 0 = otherwise

0.110

0

0.314

Rural region

1 = if the target region is rural, 0 = otherwise

0.001

0

0.039

Young age

1 = if the sample is limited to young workers, 0 = otherwise

0.009

0

0.094

Middle age

1 = if the sample is limited to middle-aged workers, 0 = otherwise

0.012

0

0.109

Older age

1 = if the sample is limited to older workers, 0 = otherwise

0.009

0

0.094

State enterprise

1 = if the target firm is a state enterprise, 0 = otherwise

0.049

0

0.217

Private firm

1 = if the target firm is a private firm, 0 = otherwise

0.091

0

0.288

Annual

1 = if annual wage is employed for empirical analysis, 0 = otherwise

0.082

0

0.275

Weekly

1 = if weekly wage is employed for empirical analysis, 0 = otherwise

0.039

0

0.193

Hourly

1 = if hourly wage is employed for empirical analysis, 0 = otherwise

0.467

0

0.499

Panel data

1 = if panel data is employed for empirical analysis, 0 = otherwise

0.067

0

0.250

Original household survey

1 = if the results of an original household survey are used as the data source, 0 = otherwise

0.481

0

0.500

Non-OLS

1 = if an estimator rather than OLS is used for estimation, 0 = otherwise

0.118

0

0.323

IV/2SLS/3SLS

1 = if IV, 2SLS, or 3SLS estimator is used for estimation, 0 = otherwise

0.001

0

0.039

Control for selection bias

1 = if the sample selection bias of employment is controlled for, 0 = otherwise

0.031

0

0.174

Occupation

1 = if the estimation simultaneously controls for occupation, 0 = otherwise

0.351

0

0.478

Health

1 = if the estimation simultaneously controls for the health of workers, 0 = otherwise

0.012

0

0.109

Firm size

1 = if the estimation simultaneously controls for the size of firms to which workers belong, 0 = otherwise

0.354

0

0.478

Trade union

1 = if the estimation simultaneously controls for the presence of trade unions, 0 = otherwise

0.045

0

0.207

Location fixed effects

1 = if the estimation simultaneously controls for location fixed effects, 0 = otherwise

0.427

0

0.495

Industry fixed effects

1 = if the estimation simultaneously controls for industry fixed effects, 0 = otherwise

0.373

0

0.484

Time fixed effects

1 = if the estimation simultaneously controls for time fixed effects, 0 = otherwise

0.051

0

0.220

Quality level

20-point scale of the quality level of the study

16.233

18

3.839

SE

Standard error of partial correlation coefficient

0.020

0.0183

0.017

  1. This table lists meta-independent variables used in regression estimation in Tables 4 and 5 and thier descriptive statistics. We evaluated the quality levels of studies included in our meta-analysis as follows: As the basic source of information for evaluating the quality levels of journal articles, we used the rankings of economic journals published on February 1, 2018 by IDEAS (http://ideas.repec.org/), a bibliographic database dedicated to economics and available freely on the Internet. IDEAS provided the most comprehensive rankings of 2159 economic journals as of February 2018. We conducted a cluster analysis by using the comprehensive evaluation scores provided by IDEAS to divide the 2159 journals into 20 clusters and then graded (weight) each journal on a scale from 20 (a group of journals belonging to the highest cluster) to 1 (a group of journals belonging to the lowest cluster). For journals not indexed by IDEAS, we referred to impact factors (Thomson Reuters) and other journal rankings that allowed us to compare them against the journals indexed by IDEAS, and then graded them according to the scores given to the equivalent journals listed in IDEAS. For academic books or book chapters, we gave an initial score of 1 and upgraded them to a median score of 10 when any of the following conditions were satisfied: (1) it is clearly specified that the book or article in question has been peer reviewed; (2) the book or article in question has been published by a major academic publisher that is assessed by outside experts; and, (3) the quality level of the research in question is clearly high