Decent Work as Determinant of Work Engagement on Dependent Self-Employed
Round 1
Reviewer 1 Report
The work is well structured but without ambition aims too high.
Some suggestions for the authors:
Paragraphs 3.2 and 3.3 are named identically
Illustration 1 is too wide to be left inside the work.
Pay attention to this table-illustration 1 because the table heads are not specified.
Author Response
We have incorporated your suggestions in the revised manuscript.
Thank you so much for your comments and the time you gave for the improvement of our manuscript.
Author Response File: 
Author Response.docx
Reviewer 2 Report
1. The identification assumptions of the models should be stated in the paper.
2. Are all controls that are included in the models predetermined or not?
3. Do the data contain (survey) weights or not? Why they have not been used in the estimations?
4. The paper reports a set of regression results that include the standard controls. An important issue is that employees are not randomly assigned into workplaces. Failure to account for sorting of employees will bias any estimated effects for the measures of well-being at work (including work engagement). The size of the bias is not known. This problem can be addressed using information on employees’ wage and work histories (https://doi.org/10.1016/j.jebo.2012.09.005). The issue should be discussed in the revised version.
5. The estimates may suffer from the omitted-variable bias (see also point 4 above). For example, personality traits are most likely correlated with all measures that are used in the analyses. This should be noted in the revised version.
6. The empirical analyses present average effects for all workers in the data. The paper does not consider the potential heterogeneity in the estimated effects. The relationships can differ significantly e.g. by age/gender. The sample size (N=42963) would an analysis of policy-relevant of the estimated relationships that are reported in the paper.
Author Response
In our research, we considered the entire sample of the European Working Conditions Survey (EWCS). The sample of this survey is a multistage, stratified, random sample of the working population in each country. Data were collected by the European Foundation for Living and Working Conditions. The main characteristics of the methodology are the following (for more information please see the technical report on https://www.eurofound.europa.eu/sites/default/files/ef_survey/field_ef_documents/6th_ewcs_-_technical_report.pdf):
Coverage: 28 EU Member States, Norway, Switzerland, Albania, the Former Yugoslav Republic of Macedonia, Montenegro, Serbia and Turkey.
Fieldwork period: February–September 2015 in the EU28, Norway and Switzerland; September– December 2015 in Albania, the Former Yugoslav Republic of Macedonia, Montenegro, Serbia and Turkey.
Target population: All residents of the countries mentioned above aged 15 or older (16 or older in Bulgaria, Norway, Spain and the UK) and in employment at the time of the survey. People were considered to be in employment if they had worked for pay or profit for at least an hour in the week preceding the interview (ILO definition).
Sample: Multi-stage, stratified, random samples of the working population in each country. Depending on the availability of high-quality registers, sampling was carried out using individual-level, household-level and address-level registers, or through enumeration using a random-walk approach. Country-level samples were stratified by region and degree of urbanisation. In each stratum, primary sampling units (PSUs) were randomly selected proportional to size. Subsequently, a random sample of households was drawn in each PSU. Finally, unless individual-level registers were used, in each household the selected respondent was the person in work who would have their birthday next.
Sample size: In most countries, the target sample size was 1,000. To reflect the larger workforce in larger countries, the target was increased to 1,200 in Poland, 1,300 in Spain, 1,400 in Italy, 1,500 in France, 1,600 in the UK and 2,000 in Germany and Turkey. The total sample size for the sixth EWCS in all 35 countries is 43,850 interviews.
Type of interview: Face to face, at the respondent’s home; average duration of 45 minutes.
We have clarified in the manuscript that the sample is a multistage, stratified, random sample of the working population in each country.
With regards to the issue of the weights, we know that EWCS considers three types of weights: a) design weights (people in households with fewer workers have a greater chance of being selected into the sample than people in households with more workers, so design weights were constructed to correct for this); b) post-stratification weights (because certain groups are overrepresented –and others underrepresented–, the design weights were calibrated by comparing the EWCS with Eurostat’s Labour Force Survey with regard to respondents’ gender, age, region, occupation and sector of economic activity); and c) supranational weights: (to ensure that larger countries weigh heavier in the EU-level results, supranational weights need to be applied when performing analyses on the European level).
These weights have not been used in the estimations of the models.
According to your valuable suggestion, relying on the arguments provided by Böckerman, Bryson, and Ilmakunnas (2012), we have explained in more detail the positive influence that involving employees and providing them with autonomy at work has on work engagement. This change has been included in the sixth paragraph of the “Discussion and conclusion” section in the new version of the manuscript.
We are very grateful for your comment regarding the control variables. Following your advice, we have explained in the “Limitations and future directions” section that we have not included the “work stories” variable in the research, what could lead to some bias in the data. Although we have not considered “work histories” in the present research, we will consider its inclusion in future studies.
