What Contributes to the Gender Gap? A Blinder–Oaxaca Decomposition Analysis of Hidden Workers in Australia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This article aims to contribute to that discussion by highlighting the gendered impact on hidden workers and to show how the explained and unexplained components of the hidden workers differ between the gender groups. Hidden workers are defined as those being underemployed, discouraged workers or unemployed. Despite the extensive literature concerning gender issues and gender gaps, the gender gap among the hidden workers remains unexplored area. Authors use an Oaxaca–Blinder decomposition approach, which is often used to analyse the differences in outcomes between groups. They find out that the gender gap among hidden workers is widened by care responsibilities, locally available socio-economic resources and narrowed by education, individual health endowment and social capital.

 The topic of hidden workers is highly relevant for the research, as the gendered impact has been overlooked and not as extensively analysed as other phenomena (such unemployment and inactivity). This article addresses the mentioned gap in the existing literature. An additional significant advantage of the article lies in its data source, which enables a highly precise identification of individuals classified as hidden worker. Moreover, the findings from this study show that much of the gender gap among hidden workers in Australia is socially and politically produced and potentially avoidable through public policy. Policy measures focused to the factors that proved to have the biggest impact on the gender gap among hidden workers. The findings from this study show that much of the gender gap among hidden workers in Australia is socially and politically produced and potentially avoidable through public policy. Greater attention from researchers and policy makers on hidden workers should be devoted to reducing this gender gap to prevent further social implications.

 The article is well-structured, beginning with a comprehensive literature review in this field, followed by a very clear and precise formulation of hypotheses that are subsequently tested. The methodology is highly appropriate for testing the stated hypotheses, as it enables the distinction between the contributions of individual or household characteristics and the differing returns on those characteristics across the observed groups. The article is suitable for publication with minor revisions.

Despite the very good overall rating, the authors need to fix some issues before the article is published:

11. Have you checked if the correlation between the variables married and childrt causes any troubles?

22. SEIFA index is mentioned. The index should be shortly described and the references should be provided, otherwise it’s very confusing for non-Australian audience.

33. The valiable nohelp is not discussed at all. It is only stated that it is based on 4 indicators. How is it constructed? Is it the average of 4 indicators? Is it an index? If yes, how is it formed? Has it been developed and used for the previous research? Is the methodology already recognized / approved from a scientific point of view?

44. Table 1: there are few mistakes, therefore please check:

a.      the label of married

b.      the label of childrt – is the age limit 14 or 15?

c.      The label of seifa

55. The line 348: is45-54 age group correct? Shouldn’t be 45-64?

66. Table 3 requires improvement in its layout for clarity and readability:

a.      the variable names are on two lines,

b.      sign (+ or -) and number are on different lines

c.      The part of the table on the page 10 is not clear: What do the variable names stand for? What do columns refer to? Is this the extended part of the table from page 9? what the resource under the table refers to?

77. Line 398: What does the number 12.4 refer to? Should’t be there 12.9?

TThe article is suitable for publication with minor revisions. 

Author Response

Thank you very much for the detailed comments of the reviewer. This has significantly improved the manuscript.

1. Have you checked if the correlation between the variables married and childrt causes any troubles?

à Yes, the Pearson correlation between the variables married and childrt is 0.2317. Furthermore, we have performed multicollinearity test which satisfies no strong collinearity among the x variables.

2. SEIFA index is mentioned. The index should be shortly described and the references should be provided, otherwise it’s very confusing for non-Australian audience.

à The explanation of the SEIFA index has been added on lines 232-234 with a reference from Australian Bureau of Statistics.

3. The valiable nohelp is not discussed at all. It is only stated that it is based on 4 indicators. How is it constructed? Is it the average of 4 indicators? Is it an index? If yes, how is it formed? Has it been developed and used for the previous research? Is the methodology already recognized / approved from a scientific point of view?

à The variable ‘nohelp’ has been further described under the table with an * on lines 241-244.

4. Table 1: there are few mistakes, therefore please check:
5. the label of married : The variable married includes legally married and de facto married.
6. the label of childrt – below 15
7. The label of seifa

à The appropriate changes have been made in Table 1.

5. The line 348: is45-54 age group correct? Shouldn’t be 45-64?

à Thanks for spotting the errors. The changes have been made.

6. Table 3 requires improvement in its layout for clarity and readability:
7. the variable names are on two lines,
8. sign (+ or -) and number are on different lines
9. The part of the table on the page 10 is not clear: What do the variable names stand for? What do columns refer to? Is this the extended part of the table from page 9? what the resource under the table refers to?

à Thanks for the suggestions. I have made the font smaller to make it clear but we can insert the table horizontally if the editorial team suggests. The detailed information on the state variables have been included under the table 3.

7. Line 398: What does the number 12.4 refer to? Should’t be there 12.9?

à It should be 12.9. The figure has been changed.

Reviewer 2 Report

Comments and Suggestions for Authors

The article touches on the impact of gender on hidden workers, which is important given the gender inequality that exists around the world.

To improve the manuscript I have a few suggestions.

Section 2. Literature review can be divided into subsections discussing further issues, i.e. demographics, human capital, etc.

The literature review presents only the factors considered in the study. I suggest expanding the literature review and discussing other factors that influence gender gaps e.g. social and cultural exclusions, patriarchy, unpaid domestic work, violence - see, for example, the literature review in: Perzyńska, J., & Guzowska, M. K. (2024). The Feminisation of Poverty in European Union Countries—Myth or Reality? Sustainability, 16(17), 7594.

I suggest moving the formulation of hypotheses from the literature review to the material and methods section.

In line 79 the word Age at the beginning of the sentence is unnecessary.

Line 125 and 128 and further in the text - should be Hypothesis 2.

Section 3. Method also presents material - I suggest changing the title to Material and methods.

How are the values of the eduqual variable interpreted? There are only 8 grades in the ISCED classification. 

In formula 1, there should be H with subscript i; clarify what Xi and beta mean.

Yun’s 2004 paper is missing in the bibliography.

In formula in line 228 there is an error; please clarify the designations of formulas 2 and 3 (what is F?) and number them separately.

In table 2 there are Age 16-29, 30-44, in the text: 16-30, 20-44 - please correct this.

Explain ACT and NT (from line 354).

Explain the letters E and C (from table 3).

In table 4 the age groups are written as inequalities and before that with a hyphen - can this be standardized in all tables?

The discussion section lacks a comparison of the results obtained with other studies - this should be supplemented.

I suggest swapping the order of chapter 6 and 7.

Typos (e.g dercile, smaple) and punctuation should be corrected.

Author Response

The authors deeply appreciate the comments made by the reviewer and accommodated all the suggestions.

Section 2. Literature review can be divided into subsections discussing further issues, i.e. demographics, human capital, etc.

• Thank you for the suggestion. The literature review now includes the expanded variables such as decision making power (lines 112-120), poverty (lines 121-130) and social and cultural norms (lines 147-155).

 

I suggest moving the formulation of hypotheses from the literature review to the material and methods section.

• The formulation of hypotheses have been moved to the materials and methods section.

In line 79 the word Age at the beginning of the sentence is unnecessary.

• The word Age has been removed.

Line 125 and 128 and further in the text - should be Hypothesis 2.

• I have made those changes.

Section 3. Method also presents material - I suggest changing the title to Material and methods.

• The title has been renamed to Materials and methods.

How are the values of the eduqual variable interpreted? There are only 8 grades in the ISCED classification. 

• The education questions in HILDA uses the Australian Standard Classification of Education (ASCED) which has 9 grades, from year 11 or below (1) to postgraduate degree (9). The higher the eduqual value, the higher one’s education level.

In formula 1, there should be H with subscript i; clarify what Xi and beta mean.

• The sentence clarifying the formulation has been added under the equation (1).

Yun’s 2004 paper is missing in the bibliography.

• The paper has been added to the references.

In formula in line 228 there is an error; please clarify the designations of formulas 2 and 3 (what is F?) and number them separately.

• The designations of formulas have been changed. F has been explained under the equation (2).

In table 2 there are Age 16-29, 30-44, in the text: 16-30, 20-44 - please correct this.

• The figures have been changed.

Explain ACT and NT (from line 354).

• The states have been explained in the parenthesis.

Explain the letters E and C (from table 3).

• The E (endowment effect) and C (coefficient effect) has been explained in the beginning of the section 4.2.

In table 4 the age groups are written as inequalities and before that with a hyphen - can this be standardized in all tables?

• The table 4 age groups have been changed for consistency.

The discussion section lacks a comparison of the results obtained with other studies - this should be supplemented.

• The discussion section has added comparisons with the other studies in lines 448-452 and in lines 460-462.

I suggest swapping the order of chapter 6 and 7.

• The order of chapters have been swapped.

Typos (e.g dercile, smaple) and punctuation should be corrected.

• The typos have been corrected.