Next Article in Journal
PVNet: A Used Vehicle Pedestrian Detection Tracking and Counting Method
Next Article in Special Issue
Does Servant Leadership Decrease Bad Behaviors? The Mediating Role of Psychological Safety and the Moderating Effect of Corporate Social Responsibility
Previous Article in Journal
Study on the Impact of Internet Usage, Aging on Farm Household Income
Previous Article in Special Issue
The Relationship between Job Insecurity and Safety Behavior: The Buffering Role of Leadership Ethics
 
 
Article
Peer-Review Record

Mitigating the Impact of Work Overload on Cybersecurity Behavior: The Moderating Influence of Corporate Ethics—A Mediated Moderation Analysis

Sustainability 2023, 15(19), 14327; https://doi.org/10.3390/su151914327
by Yunsook Hong 1, Min-Jik Kim 2,* and Taewoo Roh 3,*
Reviewer 1:
Reviewer 2:
Sustainability 2023, 15(19), 14327; https://doi.org/10.3390/su151914327
Submission received: 16 August 2023 / Revised: 12 September 2023 / Accepted: 18 September 2023 / Published: 28 September 2023

Round 1

Reviewer 1 Report

1st Round Review Comments

September 1, 2023

 

1. The title of the paper is unclear and needs to be revised. An alternative version could be "Mitigating the Impact of Work Overload on Cybersecurity Behavior: The Moderating Influence of Corporate Ethics—A Mediated Moderation Analysis".

2. The abstract is unclear and needs to be revised. An alternative could be:

Abstract: This present study examines how work overload affects cybersecurity behavior, considering job stress as a mediator and corporate ethics as a moderator. Using a time-lagged survey design, 377 South Korean employees were surveyed. The results of SEM analysis (Structural equation model) show that work overload leads to higher job stress, which in turn worsens cybersecurity behavior. High levels of corporate ethics buffer this negative relationship. The research fills gaps in existing literature by linking work overload, job stress, and corporate ethics to cybersecurity behavior, offering valuable insights for both theory and practice. Future research directions are also discussed.

 

3.The term 'Increased work overload' appears multiple times in the research hypothesis and other sections, which is a redundant expression. For example, the authors said, "Increased work overload will decrease the degree of employee cybersecurity behavior" on line 175, making readers somewhat unclear. A more streamlined alternative could be, 'Work overload negatively affects employee cybersecurity behavior.'

 

4. Figure 1 lacks H4.

 

5. The author, in the second paragraph on page 8, mentions the specific timing of the survey participation using vague terms like 'the first time, the second time, and the third time. This is not accurate. Given that this is a scientific research paper, using vague time descriptions like 'the first time, the second time, and the third time' could compromise the study's reliability and accuracy. Even if exact dates aren't provided, a basic timeframe like 'July 2023' should be disclosed to enhance transparency and credibility.

 

6.Table 1 should have three columns: column 1 (Variable), column 2 (Category), and column 3 (Percent instead of “Percentage”). Additionally, education should be changed to “Educational Level”.

 

7. The decimals of the values should be aligned in Table 2.

 

8. All statements in Table 2 such as T1, T2, and T3 should be removed.

 

9. Position_T1 should be removed from Table 2.

 

10. S.D. should be changed to “SD”.

The author's English writing skills need significant improvement.

Author Response

Dear Reviewer 1,

Most of all, we are sincerely grateful for the opportunity to revise and resubmit our manuscript and are highly appreciative of the guidance and constructive suggestions provided. We have followed all the comments and suggestions closely in revising the manuscript, and we provide a comment-by-comment account of how we address them in this revision. Would you please check our response letter for you in the form of PDF file that is attached in this system?

Author Response File: Author Response.pdf

Reviewer 2 Report

This study explores the impact of work overload on network security behavior, and the design of the paper has strong practical significance. The research hypotheses are proposed by constructing a theoretical framework through relatively abundant literature research. Empirical research such as structural equation modeling was carried out with more than 300 samples of Korean companies as objects. The design is relatively reasonable, and a simulation design close to reality has also been carried out. The research results are supported by more reliable empirical results. The research results have certain time significance for enterprise practice. Considering the influence of corporate ethics on the regulation of employees' stress and overload behaviors on employees' cybersecurity behaviors also has important reference significance for companies' management practices in the current high-level cybersecurity scenario. Overall, this thesis is reasonable and presentation is clear. It could be acceptable.

Author Response

Dear Reviewer 2,

Most of all, we are sincerely grateful for the opportunity to revise and resubmit our manuscript and are highly appreciative of the guidance and constructive suggestions provided. We have followed all the comments and suggestions closely in revising the manuscript, and we provide a comment-by-comment account of how we address them in this revision. Would you please check our response letter for you in the form of PDF file that is attached in this system?

Author Response File: Author Response.pdf

Reviewer 3 Report

After reading and analyzing the article, I make a few observations:

- Pertinent title, objective;

- Summary meets expectations, but the methodology for carrying out the study could have been more detailed;

- Introduction: presents the context of the study, the gaps it aims to fill, its objectives and justification.

- Theory and Hypotheses: well contextualized and presented, however, only 26% of the bibliography used is from the last 5 years, of the 88 materials indicated, 23 are from 2019 to 2023 (2023-1 and 2022-2).

- Method: the methodology is well explained, it has the necessary information on how the study was carried out;

- Results: the results are well presented, the necessary tests and different statistical calculations were performed;

- Discussion: I felt a lack of discussion with the literature, there were few moments in which this occurred, it was in this topic that the limitations of the study were presented;

- Conclusion: simple text, rescues what is in the introduction, does not present new data from the study and/or any contribution.

Author Response

Dear Reviewer 3,

Most of all, we are sincerely grateful for the opportunity to revise and resubmit our manuscript and are highly appreciative of the guidance and constructive suggestions provided. We have followed all the comments and suggestions closely in revising the manuscript, and we provide a comment-by-comment account of how we address them in this revision. Would you please check our response letter for you in the form of PDF file that is attached in this system?

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Reviewer Comments

September 8, 2023

 

Introduction Section:

·       Please remove text from lines 94 to 135 and incorporate this content into the concluding sections like discussion, theoretical relevance, and practical implications. This will cut down on redundancies. More specifically,

·       Connecting with Prior Literature: You've noted links with existing literature in the discussion. Enhance these by contrasting them with your findings to underscore your study's added value.

·       Theoretical and Practical Insights: Infuse "Theoretical Implications" and "Practical Implications" with key points from prior sections. For instance, emphasize how integrating work overload, stress, and corporate ethics gives a richer understanding for both realms.

·       Contribution Highlight: In your summary, emphasize how your research fills the gap, especially in how work overload impacts cybersecurity behaviors, underscoring its relevance in today's digital landscape.

·       Wrap-up: In conclusion, beyond just summarizing findings, convey how your research pushes theoretical and practical boundaries and sets the stage for future studies.

 

Method Section:

·       The reference cited in line 367 is incorrect. The present study performed traditional SEM analysis using AMOS software, which is not the PLS SEM.

·       The column “Percentage” in Table 1 is still incorrect.

·       Text below the subheadings 3.2 should be reframed as: Initially, levels of work overload and corporate ethics of the participants were assessed, respectively. At the second time point, levels of job stress were evaluated. At the final time point, cybersecurity behaviors of the participants were measured. All variables were assessed using multi-item, five-point Likert scales.

·       Table 2 should be recreated because some key variables were incorrectly included and coded. For example, the variable “gender” should be coded as “0 for male” and “1 for female”. Another instance is the incorrect inclusion of the variable “Education”. This variable should be removed.

·       The authors do not have reason to report Cronbach’s alpha coefficients in the present study because you never tested whether the tau-equivalence assumption was satisfied.

·       The authors should check if common method biases exist.

·       Before running the SEM model, the authors should check outliers, normality etc. of the items used in the present study.

·       How did the authors convert the latent variables into observed variables? Factor score?

The authors need to improve English writing skills.

Author Response

Dear Reviewer 1,

We are sincerely grateful for the opportunity to revise and resubmit our manuscript and are highly appreciative of the guidance and constructive suggestions provided. We have followed all the comments and suggestions closely in revising the manuscript in this second revision. Would you please check our second round response letter for you in the form of PDF file that is attached in this system?

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Reviewer Comments

September 10, 2023

1-The reviewer highlighted issues with the authors' calculation of the alpha coefficient. However, the authors responded with flawed arguments that the reviewer deemed unacceptable. Something commonly used by scholars does not necessarily illustrate the correctness of this method. In fact, although Cronbach a is the most commonly used single-administration test score reliability coefficient, it is still commonly misconceived and widely misused (Cho & Kim, 2015; Dunn, Baguley, & Brunsden, 2013; Green & Yang, 2009a; Raykov, 2012; Sijtsma, 2009a; Yang & Green, 2011).If the authors are genuinely committed to academic rigor and truth-seeking, rather than merely aiming to publish, they should seriously consult Cho, E. (2016). Making reliability reliable: A systematic approach to reliability coefficients. This paper, published in a top-tier journal on quantitative management methodology, has been cited 443 times. Organizational Research Methods, 19(4), 651-682.

 2-Why didn't the authors correct the wrong term 'Percentage' in Table 1? It should be 'Percent'.

3-The title of Table 1 should be "Sample Composition". Also, the decimal points in the 'Percentage' column are still not aligned.

4-In Table 2, the decimal points in the 'SD' column are still not aligned. Moreover, the number of decimal places in the table is inconsistent. Some are rounded to two decimal places, while others have three. For magnitudes like 0.00X, this means the correlation is almost zero. For magnitudes like 0.00X, this means the correlation is almost zero. Whether it's significant or not is already not important.

5-The authors do not statistics at all. Otherwise, they would not convert the latent factors into the corresponding observed variables with the method they mentioned. How do you justify the weight coefficient is the same among indicators?

6-In academic writing, authors should avoid pronouns like 'we' or 'they'. The passive voice is typically preferred in scientific literature.

7-The subheading "3.3" should be revised to "Data Analytic Strategy".

8-The sample size needed computation depends on the complexity of the model and the desired level of power, which should be transparent to the readers. Regarding the justification of the sample size in the present study, the evidence provided is insufficient. The authors merely state, "using G*Power statistical analysis and ensuring a minimum sample size of 10 cases per variable, as suggested by Barclay, Higgins, & Thompson [75]." It appears that the authors lack a deep understanding of the rationales of statistics and have simply gone through a common procedure, citing a reference as many applied researchers do. This approach is very unprofessional. For a more in-depth understanding, please consult the resource at https://www.danielsoper.com/statcalc/calculator.aspx?id=89. The reasons behind these calculations can be found by clicking the "References" icon under "Related resources.

9-How did the authors set the metric for the latent variables used in the present study?

10- In the section discussing the measurement model, the authors address the topic of discriminant validity. While it's essential to first confirm that each indicator properly loads on its intended construct and that the overall model fits the data well, this is just the beginning. Establishing both convergent and discriminant validity goes beyond the results of a Confirmatory Factor Analysis (CFA).

To verify convergent validity, as per Fornell and Larcker, one must compute the Average Variance Extracted (AVE) for each construct. This involves determining the R² value of each indicator within a construct, summing these values, and then dividing by the total number of indicators. An AVE value greater than .50 signifies that the indicators are converging effectively on the construct.

Discriminant validity, on the other hand, necessitates an understanding of the shared variance between constructs. This is achieved by correlating the constructs directly. Each construct is represented by a composite variable, created by summing its indicators' scores and then dividing by the number of indicators. When you derive correlations between these composite scores, squaring the resulting values should yield numbers less than the AVE of each construct. This outcome would bolster the assertion of discriminant validity.

11- The author should use a long table to report the results of multiple CFA analyses, which should include the CFA results of multiple latent constructs. If the author really doesn't know how to present or create this table, please refer to page 89 of your reference [75].

12-The authors should ensure that the measurement models have the quality of measurement invariance. However, this section is missing.

The authors need to improve their English writing skills.

Author Response

Dear Reviewer 1,

We are sincerely grateful for the opportunity to revise and resubmit our manuscript and are highly appreciative of the guidance and constructive suggestions provided. We have followed all the comments and suggestions closely in revising the manuscript in this second revision. Would you please check our second round response letter for you in the form of PDF file that is attached in this system?

Author Response File: Author Response.pdf

Back to TopTop