The Impacts of Social Media on Accounting and Auditing: Evidence from Greek Listed Firms
Round 1
Reviewer 1 Report
INTRODUCTION
The relevance of the study is not sufficiently described: why is it interesting to investigate the relationship between social media and accounting and auditing factors?
I suggest numbering the three research questions.
In the introduction, I would emphasize the role of impression management theory in the phenomenon studied.
THEORETICAL FRAMEWORK
Why did you decide to refer to Twitter, Facebook and LinkedIn? Try to motivate your choice.
METHODOLOGY
Research questions must be motivated and explained in light of the theoretical framework introduced in the previous paragraph. Furthermore, I would actually formulate them in the form of a question.
RESULTS
Lines 399 to 407: I would use the word "relationship" rather than a "correlation" between variables. The same also applies to the comments in the subsequent tables.
I also suggest you add a paragraph dedicated to a more in-depth discussion of the results, also in the light of impression management theory.
CONCLUSIONS
Please add more details on the practical implications of your study.
FORM
The paper needs a strong form and language revision.
Author Response
Many thanks for your comments. We attempted to explain more the significance of studying the relationship between social media accounting and auditing by changing the spelling and the references in introduction paragraph. Furthermore, we numbered research’s questions. We made clearer the impression management techniques we used throughout the research. Moreover, we motivated our choice to use Twitter, Facebook and LinkedIn in our research. So, we modified the following sentence: “Twitter, Facebook and LinkedIn were chosen as the studying social media, because they are already developed, and they are capable of providing validated and useful corporate information. More specifically, Twitter and LinkedIn are quite popular in corporate environment and Facebook was one of the top social media digital channels in Greece in 2018 [10]”. Concerning methodology’s feedback, we made changes in basic research questions and between Literature Review and Methodology in order the content to be more articulate; and we replaced in results the suggested words and we added the proposed paragraph.
Reviewer 2 Report
In the beginning, I would like to congratulate the selection of the research problem and the very nice research design.
Please try to clarify the selected sentences:
- line 208 (p.5) There is "The population of this study consists of 183 being listed on ATHEX, as it was on 208 21/07/2019" - maybe "consists of 183 companies"? The same applies to lines 382-390 (p. 10)
- lines 317/318 (p. 8) - the model could be presented more clearly (symbols and explanations).
- line 365 (Table 2) - please provide the source of data - the same applies to other tables.
The Author/s could provide more emphasis on robustness tests and post-estimation diagnosis.
Author Response
Thank you for your feedback. We clarified more the suggested sentences. Regarding the tables 2-16, they presented the results from econometrical models; for this reason, we did not provide the source of the data. We emphasized more on robustness tests in implications of our research and recommendations of future studies.
Reviewer 3 Report
Please see the comments in the uploaded file - the paper, as already stated, needs major revisions to language used and editing. You should also reconsider your research design in my opinion, since some of your assertions are too strong. For instance, the nexus between profitability and social media should be better explained and justified, if it really exists.
Moreover, please take into consideration (at least) the following references:
1. Debreceny, R. S. (2015). "Social Media, Social Networks, and Accounting". Journal of Information Systems. 29(2), 1-4.
2. Tettamanzi, P., Venturini, G. & Murgolo, M. (2022). "Sustainability and Financial Accounting: a Critical Review on the ESG Dynamics". Environmental Science and Pollution Research. 29(11), 16758-761.
3. Bae, S., Sung, C. & Kwon, O. (2021). "Accounting for Social Media Effects to Improve the Accuracy of Infection Models: Combatting the COVID-19 Pandemic and Infodemic". European Journal of Information Systems. 30(3), 342-355.
4. Drake, M., Moon, J., Twedt, B. & Warren, J. (2022). "Social Media Analysts and Sell-Side Analyst Research". Review of Accounting Studies.
Comments for author File: 
Comments.pdf
Author Response
Thank you very much for the helpful comments.
Regarding the structure and the language, we made major changes in content. We clarified that we used the definition of IAS 28 and IFRS 10, because some of the selected companies were published consolidated financial statements and we chose the accounting data from parent companies (separated financial statements). Moreover, we made inquires in social media accounts from parent companies, related parties and subsidiaries. Also, we presented the tables 2-13, because they referred in different econometrical models. We clarified the reason we used the fiscal year 2018 as “We analyzed the findings for a whole fiscal year, 2018 and we choose companies with common traits in a mutual economic environment; in other words, we only chose Greek companies being listed on ATHEX. It is mentioned that we restricted in 2018, because the following fiscal years 2019-2021 have affected by COVID-19 or Russia-Ukraine war effects in Financial Statements Data or Financial Statements Notes “events after the end of the reporting period”. Also, we took into account the referred papers, which were quite helpful for our research.
Reviewer 4 Report
Referee Report
Manuscript Number: sustainability-1809494
Article title: "The impact on social media on Accounting and Auditing: Evidence from Greek listed firms"
Manuscript Type:
Comments
The paper intends to investigate and report the relation among social media and accounting and auditing.
The authors used information (social media posts, accounting information and audit opinion information) on 175 companies being listed on ATHEX (on 21/07/2019). They applied Kruskal Wallis and OLS models.
I congratulate the authors on the enormous work they have already done, but I still think that the article needs improvement. I leave below some comments and suggestions:
- I think the authors should review the title, I think it is poorly translated/drafted.
- The abstract can be improved. There is no reference to the methodology used. On the other hand, 183 companies are not used, but 175, since the authors excluded some of the base sample.
- On page 1, line 29, the authors refer to "... Nowadays plenty of companies depend on social media....", but it is necessary to mention for what specifically.
- In the Introduction, it remains unjustified because it is important to use information from Greek companies. There is no rationale for this, which have these companies in special to be important to carry out this study.
- The authors need to review the English of the article, it has to be substantially improved. In addition to presenting several spelling errors.
- The literature review should be updated.
- The authors need to justify why they choose those social networks, in view of the purpose they propose, in particular Linkedin (essentially is an employment network) whose purpose as a network has nothing to do with the goal of the article.
- The working hypotheses should be presented by the authors throughout the literature review. The 1st paragraph of the methodology section makes no sense, it is preferable to present the hypotheses.
- The last sentence of line 222 on page 5 needs to be improved, certainly what the authors intend to mention is that it is the final sample.
- All equations have to be numbered and their presentation improved to be presented in a scientific way.
- The last paragraph on page 9 is quite confusing.
- Table 5 need to show the significance of the correlations. On the other hand, the authors need to present the VIF values of the variables, because there are several values above the 45% of correlations.
- The authors need to substantially improve the discussion of the results. What they do is only the exposure of the signals obtained in the models. They should also compare the results obtained with the literature presented.
- Finally, the conclusion needs to be improved. What are the implications of the results for literature, for the practical and in political terms?
Good luck with the paper while you keep working on it. I hope my comments can help you strengthen the paper.
Author Response
Thank you very much for your feedback. We explained the populations of our study in abstract and methodology/sample paragraph. We also made major changes in grammar and presentation of our study. Furthermore, we justified the use of LinkedIn in our study (“More specifically, Twitter and LinkedIn are quite popular in corporate environment and Facebook was one of the top social media digital channels in Greece in 2018 [10].”). Last but not least, we made substantial changes in conclusion as “Considering the sample and time period ….provide rational results with the effective use of data analysis tools and computer science in new studies.”
We would like once again to thank you for your comments and suggestions, without which our work wouldn’t have been improved.
Round 2
Reviewer 1 Report
Well done. The paper is much improved.
I only have two final concerns:
1) concerning the research questions formulation, I suggest you again clearly explain how you have defined them in the theoretical paragraph. A good idea could be to add a specific paragraph on this;
2) please add more details on the practical implications of your study in the conclusion section.
Good luck!
Author Response
Many thanks for your comments. We attempted to explain more the significance of studying the relationship between social media accounting and auditing by changing the spelling and the references in introduction paragraph. Furthermore, we numbered research’s questions. We made clearer the impression management techniques we used throughout the research. Moreover, we motivated our choice to use Twitter, Facebook and LinkedIn in our research. So, we modified the following sentence: “Twitter, Facebook and LinkedIn were chosen as the studying social media, because they are already developed, and they are capable of providing validated and useful corporate information. More specifically, Twitter and LinkedIn are quite popular in corporate environment and Facebook was one of the top social media digital channels in Greece in 2018 [10]”. Concerning methodology’s feedback, we made changes in basic research questions and between Literature Review and Methodology in order the content to be more articulate; and we replaced in results the suggested words and we added the proposed paragraph.
Well done. The paper is much improved.
I only have two final concerns:
- concerning the research questions formulation, I suggest you again clearly explain how you have defined them in the theoretical paragraph. A good idea could be to add a specific paragraph on this;
2) please add more details on the practical implications of your study in the conclusion section.
Thank you again for your feedback. We changed the first paragraph in methodology section with research’s questions and we added the questions in results section in order to can be understood. Concerning the practical implications we added one more paragraph in conclusion with practical implications, please refer to “The study can have a valuable…distribution expenses”.
Reviewer 3 Report
The major concerns are two:
- the language and editing of the article, which need an extensive review (words misspelt and extra typing are still present);
- the methodological approach concerns, that we raised in the previous round of review, have not been properly tackled nor commented - in fact, I struggle to understand whether results are indeed due to some sort of causation or, in the contrary, the correlations identified are due to third factors, not properly considered in the research design.
Author Response
Thank you very much for the helpful comments.
Regarding the structure and the language, we made major changes in content. We clarified that we used the definition of IAS 28 and IFRS 10, because some of the selected companies were published consolidated financial statements and we chose the accounting data from parent companies (separated financial statements). Moreover, we made inquires in social media accounts from parent companies, related parties and subsidiaries. Also, we presented the tables 2-13, because they referred in different econometrical models. We clarified the reason we used the fiscal year 2018 as “We analyzed the findings for a whole fiscal year, 2018 and we choose companies with common traits in a mutual economic environment; in other words, we only chose Greek companies being listed on ATHEX. It is mentioned that we restricted in 2018, because the following fiscal years 2019-2021 have affected by COVID-19 or Russia-Ukraine war effects in Financial Statements Data or Financial Statements Notes “events after the end of the reporting period”. Also, we took into account the referred papers, which were quite helpful for our research.
The major concerns are two:
- the language and editing of the article, which need an extensive review (words misspelt and extra typing are still present);
- the methodological approach concerns, that we raised in the previous round of review, have not been properly tackled nor commented - in fact, I struggle to understand whether results are indeed due to some sort of causation or, in the contrary, the correlations identified are due to third factors, not properly considered in the research design.
Thank you again for your feedback. We made further corrections in grammar, spelling and vocabulary in order the text to be more articulate. Moreover, concerning we added a sentence concerning the sample of companies that we have studied. First of all, the sample is representative of Greek companies due to different type characteristics in social media accounts, Financial Reporting Rules and Audit Reports. Moreover, we added a paragraph in “results” section in which we calculated and analyzed VIF and the correlation of variables. Also, we added parts concerning the methodology, for instance the Kruskal Wallis Model: “With the application of non-parametric Kruskal-Wallis Model, we checked if there were any statistically significant differences between the groups of social media ac-counts and change of profit for 2017-2018” and the OLS model “in order to study the relation among accounting variables and social media posts and determine if the social media posts (dependent variable) depends on the independent variables of Intangible Assets, Total Assets, Cash, Earnings before Taxes and Total Liabilities”.
Reviewer 4 Report
Referee Report
Manuscript Number: sustainability-1809494 R1
Article title: "The impact of social media on Accounting and Auditing: Evidence from Greek listed firms"
Manuscript Type:
Comments
Unfortunately, when analyzing the new version of the article, I note that little has been done by the authors to follow my comments.
For example, the discussion of the results on pages 11 to 13 that the authors pointed out as being new is exactly the same as it was before and so there was no improvement in the discussion of results.
So, I return to expose below much of the comments I had previously submitted:
- The abstract can be improved. There is no reference to the methodology used.
- In the Introduction, it remains unjustified why it is important to use information from Greek companies. There is no rationale for this, which have these companies in special to be important to carry out this study.
- The authors need to review the English of the article, it continues to need to be improved.
- The literature review should be updated.
- The working hypotheses should be presented by the authors throughout the literature review. The 1st paragraph of the methodology section makes no sense, it is preferable to present the hypotheses.
- With this version we realized that the empirical application is carried out only with 75 companies. I wonder how the results allow conclusions? What is the representativeness of this sample? The authors need to explain this part well.
- All equations have to be numbered and their presentation improved to be presented in a scientific way.
- The last paragraph on page 9 is quite confusing.
- The authors need to present the results of the regressions that are summarized in table 5.
- Table 6 need to show the significance of the correlations. On the other hand, the authors need to present the VIF values of the variables, because there are several values above the 45% of correlations.
- The authors need to substantially improve the discussion of the results. What they do is only the exposure of the signals obtained in the models. They should also compare the results obtained with the literature presented.
- On page 14, the last paragraph of section 4 needs to be developed. To what extent does this work extend that of Yang and Liu (2017).
- Finally, the conclusion needs to be improved. What are the implications of the results for literature, for the practical and in political terms? What the authors added in lines 582 to 591 on page 15 is not enough and confused.
Good luck with the paper while you keep working on it. I hope my comments can help you strengthen the paper.
Author Response
Thank you very much for your feedback. We explained the populations of our study in abstract and methodology/sample paragraph. We also made major changes in grammar and presentation of our study. Furthermore, we justified the use of LinkedIn in our study (“More specifically, Twitter and LinkedIn are quite popular in corporate environment and Facebook was one of the top social media digital channels in Greece in 2018 [10].”). Last but not least, we made substantial changes in conclusion as “Considering the sample and time period ….provide rational results with the effective use of data analysis tools and computer science in new studies.”
We would like once again to thank you for your comments and suggestions, without which our work wouldn’t have been improved.
Unfortunately, when analyzing the new version of the article, I note that little has been done by the authors to follow my comments.
For example, the discussion of the results on pages 11 to 13 that the authors pointed out as being new is exactly the same as it was before and so there was no improvement in the discussion of results.
- So, I return to expose below much of the comments I had previously submitted.
Thank you for your feedback. We added our comments as indicated below. We made changes in abstract concerning the methodology.
- The abstract can be improved. There is no reference to the methodology used.
Thank you for your feedback. We added our comments as indicated below. We made changes in abstract concerning the methodology.
- In the Introduction, it remains unjustified why it is important to use information from Greek companies. There is no rationale for this, which have these companies in special to be important to carry out this study.
We added the following section as an explanation: “We used Greek Listed Companies in our research as many of them have not a quite developed social media strategy and their reports came from different audit companies (not only Big4 firms), but they also present different audit opinions (not only unmodified opinion) and there are companies which have not the ability to operate as a going concern. Also, Listed Greek Companies have similar Reporting of Financial Statements, because they are listed in Greek Stock Exchange. Thus, because of the characteristics we mentioned above, the results will be constructive for future studies”.
- The authors need to review the English of the article, it continues to need to be improved.
We made grammatical and vocabulary changes in whole text, so it can be more articulate. Moreover, we
- The literature review should be updated.
We updated the literature. Moreover, we explained the literature is limited we the following paragraph: “Taking into consideration the relative background of studies of between social media, accounting and auditing, the literature is quite limited; for this reason, our study is focus on the collection of data of corporate social media accounts and the development of statistical and econometrical models in order to analyse the relationship between social media posts and strategies in relation with specific accounting and auditing variables”.
- The working hypotheses should be presented by the authors throughout the literature review. The 1st paragraph of the methodology section makes no sense, it is preferable to present the hypotheses.
We changed the paragraph, and we added the hypotheses’ questions.
- With this version we realized that the empirical application is carried out only with 75 companies. I wonder how the results allow conclusions? What is the representativeness of this sample? The authors need to explain this part well.
We justified that the selected sample is representative of population. The sample is representative of Greek companies due to different type characteristics in social media accounts, Financial Reporting Rules and Audit Reports; we added the following section: “The 75 companies of our sample presented different characteristics in social media ac-counts (active, inactive, of not social media accounts at all), they had different performances and auditing disclosures; so, our sample is a representative sample of 183 Greek Listed Companies.”
- All equations have to be numbered and their presentation improved to be presented in a scientific way.
The equations are changed. Thank you for your comment.
- The last paragraph on page 9 is quite confusing.
We changed the paragraph thank you for your comment.
- The authors need to present the results of the regressions that are summarized in table 5.
The Table 5 presents the results of our study which are mentioned in the conclusion section. Thank you for your feedback.
- Table 6 need to show the significance of the correlations. On the other hand, the authors need to present the VIF values of the variables, because there are several values above the 45% of correlations.
We analyzed the significance of correlation, and we added a paragraph concerning the VIF. Thank you for your comments.
- The authors need to substantially improve the discussion of the results. What they do is only the exposure of the signals obtained in the models. They should also compare the results obtained with the literature presented.
- On page 14, the last paragraph of section 4 needs to be developed. To what extent does this work extend that of Yang and Liu (2017).
- Finally, the conclusion needs to be improved. What are the implications of the results for literature, for the practical and in political terms? What the authors added in lines 582 to 591 on page 15 is not enough and confused.
Concerning the literature and the previous studies which we were extent we changed and added the following section in conclusion: “Widespread use of social media has equal potential … between social media posts and strategies in relation with specific accounting and auditing variables. Also, we added the following paragraph: “The study can have a valuable impact on corporate… development expenses and distribution expenses”. Thank you again for your valuable and constructive feedback.
Good luck with the paper while you keep working on it. I hope my comments can help you strengthen the paper.
Thanks you again for your suggestions
Round 3
Reviewer 4 Report
Referee Report
Manuscript Number: sustainability-1809494 R2
Article title: "The impact on social media on Accounting and Auditing: Evidence from Greek listed firms"
Manuscript Type:
Comments
I thank the authors for considering my suggestions in this review of the article.
There are still five aspects for which I think the authors should pay attention and correct. Like this:
- When I mentioned the authors need to write their working hypotheses and present them, I did not refer to the type of simple correction made in this version. Don't just go to page 2, lines 65 to 71, and put H1, H2 and H3. And on page 5, lines 218 to 222, add in that sentence the word hypotheses.
As I mentioned before, hypotheses should appear in the respective section of the literature review where the theoretical support for the hypothesis is presented. And it should present them as it is usual in the articles.
I ask the authors to make this change correctly.
- Regarding the final size of the sample and justification of whether it is a representative dimension. What I asked for was a literature-based rationale already published that ensured that the number of companies (75) was representative of the total number of companies and could allow for conclusions.
What the authors added on page 5, lines 242 to 245, is not an acceptable justification!
Please review this aspect.
- The paragraph on page 9 that I referred to as being confused is what appears on that page in lines 405 to 413. Please improve it.
- Please, and again, place in table 6 the significance levels of each pair of correlations that you present there.
- What is shown in Table 5? If you just want to analyse the significance of the templates you, then present in the following tables, then you should change the order of presentation of the tables, or eventually consider putting this table 5 attached. As it stands, it becomes confusing.
Good luck with the paper while you keep working on it. I hope my comments can help you strengthen the paper.
Author Response
I thank the authors for considering my suggestions in this review of the article.
There are still five aspects for which I think the authors should pay attention and correct. Like this:
- When I mentioned the authors need to write their working hypotheses and present them, I did not refer to the type of simple correction made in this version. Don't just go to page 2, lines 65 to 71, and put H1, H2 and H3. And on page 5, lines 218 to 222, add in that sentence the word hypotheses.
As I mentioned before, hypotheses should appear in the respective section of the literature review where the theoretical support for the hypothesis is presented. And it should present them as it is usual in the articles.
I ask the authors to make this change correctly.
Thank you for your clarifications. We added the hypotheses in an additional paragraph at literature review section (please refer to “2.4 Research Questions and Hypothesis”). Also, and we updated the section of literature review according to these questions.
Regarding the final size of the sample and justification of whether it is a representative dimension. What I asked for was a literature-based rationale already published that ensured that the number of companies (75) was representative of the total number of companies and could allow for conclusions.
What the authors added on page 5, lines 242 to 245, is not an acceptable justification!
Please review this aspect.
Thank you for your feedback. We had added a justification taking into consideration the audit methodology in sampling. However, we corrected and updated this part according to the existing literature (please refer to “Concerning the sample of 75 companies, Chi-square statistics did not show statistically significant differences (p> 0.05) between the sample of 75 companies and the total population of 183 firms regarding the Sales, the Employees’ number and the Total Assets [32].)”
The paragraph on page 9 that I referred to as being confused is what appears on that page in lines 405 to 413. Please improve it.
We changed the paragraph to be more articulate, please refer to “In order to analyze more the relationship between the companies which have social media accounts and their profits, we extent more Kruskal Wallis Model; we did not present the numerical option “1” separately. So, we set the number “1” for companies which had at least one social media account on Twitter, Facebook and LinkedIn. We set the number “0” for companies with inactive social media accounts and for companies with not social media accounts at all. The Table 4 presents that the average of companies “1”is higher than the “0” entities. Therefore, the change of profits for companies with social media accounts is bigger than the companies which have not any social media accounts.”
Please, and again, place in table 6 the significance levels of each pair of correlations that you present there.
Thank you for your clarifications. We used Pearson’s Model to analyse the correlation and we made some changes to present the significance level at the table.
- What is shown in Table 5? If you just want to analyse the significance of the templates you, then present in the following tables, then you should change the order of presentation of the tables, or eventually consider putting this table 5 attached. As it stands, it becomes confusing.
Thank you for your comments. Table 5 summarized the research’s questions. We deleted the table, and we made some changes in tables’ order and the comments of results.
