Can work value orientations explain the gender wage gap in Austria?

Figures & data

Table 1. Descriptive statistics of the items for work value orientations and results of an exploratory factor analysis.

Item	Mean	Std. deviation	Factor 1	Factor 2
Job security (x1)	4.60	0.64	0.21	0.58
High income (x2)	3.85	0.95	−0.22	0.77
Good opportunities for advancement (x3)	3.63	1.14	0.21	0.72
An interesting job (x4)	4.45	0.83	0.70	0.24
A job that allows for independent working (x5)	4.32	0.97	0.70	0.09
A job that allows someone to help other people (x6)	3.85	1.11	0.84	0.04
A job that is useful to society (x7)	4.02	1.08	0.85	0.07
A job that involves personal contact with other people (x8)	4.10	1.11	0.80	0.01
Eigenvalues of rotated factors			3.163	1.516
Explained variance			39.5	19.0
Cronbach’s alpha			0.844	0.478

Source: SSÖ (2016), only employed persons without a migration background, n = 1,115.

Remarks: Scale values range from 1 to 5, with 5 = very important, 4 = important, 3 = neither important nor unimportant, 2 = not important, 1 = not important at all.

Table 2. Independent and control variables.

Group	Variables	Categories
Independent variable
Gender	Gender	1 = female, 0 = male
Control variables
Other socialization-related variables	Age	1 = 16 to 30 years old; 2 = 31 to 49 years old; 3 = 50 years or older
	Parent’s highest educational level	Education of father or – if higher – education of mother 1 = low; 2 = middle; 3 = high
	Parent’s highest socioeconomic status	Socioeconomic status of father or – if higher – status of mother. Ganzeboom’s international index of socioeconomic status was used (Ganzeboom 2010)
	Respondent’s education	1 = low; 2 = middle; 3 = high
Career-related variables	Interruption of occupation	1 = never; 2 = less than 4 years; 3 = 4 years or longer
Household-related variables	Partnership	0 = no; 1 = yes
	Child 0 to 5 years old	0 = no; 1 = yes
	Child 6 to 17 years old	0 = no; 1 = yes
	Size of place of residence	1 = up to 5,000 inhabitants; 2 = 5,001 to 30,000; 3 = 30,001 to 300,000; 4 = Vienna (approx. 2,000,0000)
Job-related variables	Working time	hours
	Economic branch	12 categories (a)
	Occupation (ISCO)	10 different categories that correspond to the first ISCO level

Note: (a) The 12 branches are: 1 = Production, 2 = Trade, 3 = Agriculture, 4 = Tourism, 5 = Construction, 6 = Communication/Money/Insurance/Housing, 7 = Traffic/Transport, 8 = Professionals/Science/Technical Services, 9 = Public Administration, 10 = Education, 11 = Health and Social Services, 12 = Other Services.

Table 3. Models for data analysis.

Testing	Model	Variables of interest	Control variables
H1 and H3	1	Independent variable = gender	Age, parent’s highest educational level and socioeconomic status
		Dependent variable = work value orientations
	2	Independent variable = gender	Model 1 and educational level of respondent, household- and career-related variables
		Dependent variable = work value orientations
	3	Independent variable = gender	Model 2 and job-related variables
		Dependent variable = work value orientations
H2 and H4	4	Independent variables = work value orientations	Gender, age, parent’s highest educational level and socioeconomic status and educational level of respondent, household- and career-related variables
		Dependent variable = income
	5	Independent variables = work value orientations	Model 4 and job-related variables
		Dependent variable = income
H1 and H3	3*	Independent variable = gender and interactions of gender with all other control variables	Model 3
		Dependent variable = work value orientations
H2 and H4	5*	Independent variables = work value orientations and interactions of gender with work value orientations and all other control variables	Model 5
		Dependent variable = income

Table 4. Intrinsic and extrinsic work value orientations by gender (estimated marginal means).

Analysis	Female	Male	Difference	p-value (Two-tailed)	p-value (One-tailed)
	Intrinsic work values
Bivariate	4.24	4.05	0.19	0.000	0.0000
Model 1	4.23	4.05	0.18	0.001	0.0005
Model 2	4.07	3.88	0.19	0.003	0.0015
Model 3	3.99	3.86	0.13	0.067	0.0335
	Extrinsic work values
Bivariate	4.00	4.10	−0.10	0.020	0.0100
Model 1	4.00	4.09	−0.09	0.021	0.0105
Model 2	3.88	3.90	−0.02	0.607	0.3035
Model 3	3.89	3.82	0.07	0.208	0.1040

Source: SSÖ (2016), only employed persons without a migration background, n = 905.

Remarks: Scale values for intrinsic and extrinsic work values range from 1 to 5, with 5 = very important and 1 = not important at all. Models 1–3 are described in Table 3.

Table 5. Effects of intrinsic and extrinsic work value orientations on monthly net income.

Analysis	Intrinsic work value orientation			Extrinsic work value orientation
	Effect	p-value (Two-tailed)	p-value (One-tailed)	Effect	p-value (Two-tailed)	p-value (One-tailed)
		Gross monthly income
Trivariate	225.40	0.000	0.0000	2.05	0.977	0.4885
Model 4	160.90	0.003	0.0015	94.17	0.153	0.0765
Model 5	128.23	0.008	0.0040	64.37	0.263	0.1315

Source: SSÖ (2016), only employed persons without a migration background, n = 619.

Remarks: Scale values for intrinsic and extrinsic work values range from 1 to 5, with 5 = very important and 1 = not important at all. Trivariate = intrinsic and extrinsic work values as independent variables, and income as the dependent variable. Models 4–5 are described in Table 3. Effects are estimated marginal means.

Table 6. Gender wage gap explained by different set of variables.

Model	Absolute in EUR	p-value	in %	Absolute in EUR	p-value	in %
	Gross monthly income
Men	2,630.8	0.000	100.0%	2630.8	0.000	100.0%
Women	1,806.2	0.000	68.7%	1806.2	0.000	68.7%
	Wage gap in gross monthly income
Without control variables	−824.6	0.000	100.0%	−824.6	0.000	100.0%
After including control variables of Model 1	−811.9	0.000	98.5%	−811.9	0.000	98.5%
After including control variables of Model 2	−537.8	0.000	65.2%	−537.8	0.000	65.2%
After including control variables of Model 3	−349.2	0.000	42.3%	Included after work value orientations
Work value orientations	−379.7	0.000	46.1%	−594.6	0.000	72.1%
After including control variables of model 3	Included before work value orientations			−379.7	0.000	46.1%
	Contribution of work value orientations to wage gap
Effects of work value orientations	30.5	0.011	3.7%	56.8	0.003	6.9%

Source: SSÖ (2016), only employed persons without a migration background, n = 619.

Remarks: The control variables of Models 1 to 3 are defined in Table 3. Pw-value for two-tailed testing was examined because the hypotheses were undirected. Control variables may increase or decrease the gender wage gap. Effects are estimated marginal means.