Decision Tree Analysis of Sustainable and Ethical Food Preferences of Undergraduate Students of Gastronomy and Culinary Arts
Round 1
Reviewer 1 Report
Thanks for letting me review this.
Philosophically, I am not sure if patrons who eat out want to be lectured (or goaded) by a chef regarding what they should eat. Ultimately, it appears that culinary establishments will be guided by commercial considerations as their main decision-making guide regarding what's on their menu. To this extent, it is not clear that the intention of a chef in training will trickle through to reality, once they leave school and enter the commercial world.
Aside from that, there should be a table that lists individual variables with clear meanings (PV13, PV14 etc, what does this stand for?), along with basic statistical information (means, standard deviations etc). I realise a different article is referenced that presumably explains that, but I think readers cannot be expected to look this up just to understand what these variables names mean.
With regards to the classification model, I wonder if the number of responses is sufficient to have a testing cohort of 20 percent, which appears to be merely ca.66 responses. This seems a little thin for the purpose of validation?
To be honest, it appears to me that this whole exercise may be easier accomplished by using a simple regression model, perhaps adding in a few interaction terms to determine the finer influence of potential mediators. The authors don’t make a strong enough case for the use of their classification model.
Overall, this is a nice little project though it seems only of theoretical interest, as its practical relevance is somewhat questionable (see opening remarks).
Author Response
Response to Reviewer 1 Comments
We would like to thank the three reviewers for sharing their constructive comments. I found all the comments to be useful and believe they have contributed to improving the quality of this paper. Responses to Reviewer 1’s specific comments are provided below.
Point 1: Philosophically, I am not sure if patrons who eat out want to be lectured (or goaded) by a chef regarding what they should eat. Ultimately, it appears that culinary establishments will be guided by commercial considerations as their main decision-making guide regarding what's on their menu. To this extent, it is not clear that the intention of a chef in training will trickle through to reality, once they leave school and enter the commercial world.
Response 1: Commercial concerns are an important factor in determining the menus of businesses. However, chefs have the potential to influence customer preferences, and therefore restaurant menus, by presenting delicious and visually appealing products using sustainable ingredients. It is shown that using healthier and more sustainable ingredients can provide a competitive advantage, especially in fast food and fine dining restaurants (Lu & Gursoy, 2017). Marketing and gaining competitiveness through sustainable menus can enable to reduce commercial concerns. In addition, as chefs can be more flexible especially in independent restaurants, they can include more local and sustainable products in their menus. Telling the stories of products can also create a competitive advantage in another way by influencing customers and positively affecting their intention to revisit (Sharma, Moon &Strohbehn, 2014). By incorporating underutilized and traditional products into their dishes, chefs can help show how important diversity is in creating a more sustainable and equitable food system (Bioversity International, 2017). Recently, chefs have been involved in sustainable, ethical and socially beneficial projects. (eg, https://www.foodforsoul.it). Moreover, important organizations such as the United Nations are collaborating with chefs to achieve sustainable development goals through food (https://sdg2advocacyhub.org/intro_CM). This suggests that chefs have the potential to influence society and choices about sustainable and ethical food. The fact that gastronomy has become an increasingly popular concept leads to the increasing popularity and prestige of the culinary profession. (Hughes, 2003). This popularity leads to more media coverage of chefs. TV shows and celebrity chefs can inform, educate and inspire audience to make more environmentally friendly and sustainable food choices. They can provide guidance on how to produce less food waste or how to utilize leftovers and waste products (Bölükbaş, Sarıkaya & Yazıcıoğlu,2021). Chefs reach many people through their social media accounts and can influence people and raise awareness on various issues through their posts. It is thought that the awareness of sustainable and ethical nutrition in society will affect consumer preferences over time and chefs will plan restaurant menus accordingly.
Please see L370-395.
References:
Bioversity International, 2017. Mainstreaming Agrobiodiversity in Sustainable Food Systems: Scienti¬c Foundations for an Agrobiodiversity Index. Bioversity International, Rome, Italy.
Bölükbaş, R., Sarıkaya, G.S., & Yazıcıoğlu, İ. (2021). Analysis of food waste and sustainability behavior in Turkish television cooking shows. International Journal of Gastronomy and Food Science, 24, 100336. https://doi.org/10.1016/j.ijgfs.2021.100336
Hughes, M. H. (2003). Culinary Professional Training: Measurement of Nutrition Knowledge Among Culinary Students Enrolled in a Southeastern Culinary Arts Institute. Unpublished Ph.Doctorial Thesis, University of Alabama.
Lu, L., & Gursoy, D. (2017). Does offering an organic food menu help restaurants excel in competition? An examination of diners’ decision-making. International Journal of Hospitality Management, 63, 72-81. https://doi.org/10.1016/j.ijhm.2017.03.004.
Sharma, A., Moon, J., & Strohbehn, C. (2014). Restaurant's decision to purchase local foods: Influence of value chain activities. International Journal of Hospitality Management, 39,130-143. https://doi.org/10.1016/j.ijhm.2014.01.009.
Point 2: Aside from that, there should be a table that lists individual variables with clear meanings (PV13, PV14 etc, what does this stand for?), along with basic statistical information (means, standard deviations etc). I realise a different article is referenced that presumably explains that, but I think readers cannot be expected to look this up just to understand what these variables names mean.
Response 2: A brief explanation of the relevant variables was written under each table to make it easier for readers to understand. The relevant explanations have been revised again. In addition, explanations of items have been added to the supplementary materials section.
Point 3:With regards to the classification model, I wonder if the number of responses is sufficient to have a testing cohort of 20 percent, which appears to be merely ca.66 responses. This seems a little thin for the purpose of validation?
Response 3: “The foregoing data splitting methods can be implemented once we specify a splitting ratio. A commonly used ratio is 80:20, which means 80% of the data is for training and 20% for testing. Other ratios such as 70:30, 60:40, and even 50:50 are also used in practice. There does not seem to be clear guidance on what ratio is best or optimal for a given dataset. The 80:20 split draws its justification from the well-known Pareto principle, but that is again just a thumb rule used by practitioners.” Joseph, V. R. (2022). Optimal ratio for data splitting. Statistical Analysis and Data Mining: The ASA Data Science Journal, 1-8.
Please see L297-298
Point 4: To be honest, it appears to me that this whole exercise may be easier accomplished by using a simple regression model, perhaps adding in a few interaction terms to determine the finer influence of potential mediators. The authors don’t make a strong enough case for the use of their classification model.
Response 4: There are pros and cons to using a classification model versus a regression model. In general, classification models are used when the target variable is categorical, regression models are used when the target variable is continuous. Advantage of using a classification model is that it can be easier to interpret, especially if you have a small number of categories in your target variable, a classification model could be simpler to understand than a regression model. On the other hand, regression models can be more powerful, especially when you have a large number of categories in your target variable or when you have a continuous target variable. Both classification and regression models can be useful tools for analyzing data, and it's important to choose the one that is most appropriate for your specific problem. In the end, it really comes down to the specific characteristics of your data and the research questions you are trying to answer.
Please see L202-206.
Point 5: Overall, this is a nice little project though it seems only of theoretical interest, as its practical relevance is somewhat questionable (see opening remarks).
Response 5: Opening remarks unavailable.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript describes latent variable and decision tree analysis of attitudes toward sustainability of future chefs. Methodologically the analysis appears to be sound, but presentation and understandability of the analysis and results needs improvement.
Specific comments:
L15 - I think the word threatening is a little strong
L46 - ref. 5 is not the original study for this data. please reference the original study whenever possible.
L99 - food systems, like all human activities, contribute to these environmental problems, but are not the sole cause. please revise language
L101 - ref. 22 is not the original study for this data.
L153-162 - much more information on the survey instrument and administering the survey is needed. how were the students selected? Were they all from the same university? Demographic data were collected but not presented? Please include the survey instrument as supplementary material.
L163 - in the data analysis, was the Likert scale values handled as categories or continuous values?
L174 - please define CFI, MFI, RMSEA
Table 1, 2, and 3 - data in these tables are not referenced in the text and the data are not discussed in the text
L183 - please define TLI
L191 - are you sure the reliability coefficient is exactly 0.85 for all 3 groups of predictors
Table 4 - not sure this table is necessary as these can easily be presented in the text
Table 6 - there is only 1 decision tree predicting 2 groups. there is no need to show data for both high and low because it is the same. Only show data for the total model. Suggest including model evaluation parameters of AUC, accuracy, sensitivity, specificity and Matthews correlation coefficient.
L240-249 - this is an explanation of how a decision tree works rather than discuss of findings/results. please remove
L315-325 - much of what is in this paragraph has already been discussed in the Conclusion
Author Response
Response to Reviewer 2 Comments
We would like to thank the three reviewers for sharing their constructive comments. I found all the comments to be useful and believe they have contributed to improving the quality of this paper. Responses to Reviewer 2’s specific comments are provided below.
Point 1: This manuscript describes latent variable and decision tree analysis of attitudes toward sustainability of future chefs. Methodologically the analysis appears to be sound, but presentation and understandability of the analysis and results needs improvement.
Response 1: Necessary improvements have been made to ensure presentation and understandability.
Point 2: L15 - I think the word threatening is a little strong
Response 2:. “Threatening” is changed to “having effects on”. Please see L15.
Point 3: L46 - ref. 5 is not the original study for this data. please reference the original study whenever possible
Response 3: The original studies have been added in the paper.
Please see L46 – ref.5 (revised)-6
Point 4: L99 - food systems, like all human activities, contribute to these environmental problems, but are not the sole cause. please revise language
Response 4: The sentence is changed as “Production and consumption of food, like all human activities, contribute to some environmental problems such as climate change, water pollution, water scarcity, soil degradation, and biodiversity loss.” Please see L98-101.
Point 5: L101 - ref. 22 is not the original study for this data.
Response 5: The original study has been added in the paper. Please see L104 – ref. 23
Point 6: L153-162 - much more information on the survey instrument and administering the survey is needed. how were the students selected? Were they all from the same university? Demographic data were collected but not presented? Please include the survey instrument as supplementary material.
Response 6: Necessary changes were made and shown in the paper (L156-184). Demographic data has been added to the paper (L278-294). The questionnaire form is included as supplementary material (please see Appendix 1).
Point 7: L163 - in the data analysis, was the Likert scale values handled as categories or continuous values?
Response 7: An index value is calculated by summing up item scores for each dimension, which means Likert type scale is transformed into a continuous scale. The calculated index values representing each dimension are used in the analysis. Please see L195-197.
Point 8: L174 - please define CFI, MFI, RMSEA
Response 8: Remarks have been added in the paper. Please see L214-221.
Point 9: Table 1, 2, and 3 - data in these tables are not referenced in the text and the data are not discussed in the text
Response 9: Tables are referenced in the text.
Point 10: L183 - please define TLI
Response 10: Explained in L217
Point 11: L191 - are you sure the reliability coefficient is exactly 0.85 for all 3 groups of predictors
Response 11: A typo has been made. Thank you for noticing. Corrections have been made in the relevant parts.
Point 12: Table 4 - not sure this table is necessary as these can easily be presented in the text
Response 12: Table 4 has been removed.
Point 13: Table 6 - there is only 1 decision tree predicting 2 groups. there is no need to show data for both high and low because it is the same. Only show data for the total model. Suggest including model evaluation parameters of AUC, accuracy, sensitivity, specificity and Matthews correlation coefficient.
Response 13: Table 6 has been edited as you indicated. (In this version this table was named as table 5)
Point 14: L240-249 - this is an explanation of how a decision tree works rather than discuss of findings/results. please remove
Response 14: This explanation is included at the end of the "3.2. Data Analysis section" to make the decision tree analysis more understandable.
Point 15: L315-325 - much of what is in this paragraph has already been discussed in the Conclusion Response 15: Conclusion section has been revised.
Author Response File: Author Response.pdf
Reviewer 3 Report
This research predicts the sustainable and ethical food preferences of gastronomy and culinary arts (GCA) undergraduate program students, who have the potential to become the chefs of the future, and determine the importance of the independent variables in the decision tree classification and the effects of these independent variables. This research is novel and meaningful.
However, the authors should improve English language and style. And some main details should be improved. you should check and assure that all the cited references relevant to the research, And the results should be clearly presented. the article should be adequately referenced.
Author Response
Response to Reviewer 3 Comments
We would like to thank the three reviewers for sharing their constructive comments. I found all the comments to be useful and believe they have contributed to improving the quality of this paper. Responses to Reviewer 3’s specific comments are provided below.
Point 1: This research predicts the sustainable and ethical food preferences of gastronomy and culinary arts (GCA) undergraduate program students, who have the potential to become the chefs of the future, and determine the importance of the independent variables in the decision tree classification and the effects of these independent variables. This research is novel and meaningful.
Response 1: Thank you for your kind comment.
Point 2: However, the authors should improve English language and style. And some main details should be improved. you should check and assure that all the cited references relevant to the research, And the results should be clearly presented. the article should be adequately referenced.
Response 2:. The paper has been revised in line with your suggestions. The revisions are shown in the paper.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Additions added to manuscript and additional explanations have improved my understanding of this paper.
The supplementary material provided (e.g. description of variables) in the latest version helps to understand the model better.
Still find the explanation for low sample size lacking (There is no problem with the 20/80 split, the problem is the low validation number!)
Anyways, sufficient improvements have been made and I support publication.
Well done, authors!
Author Response
Dear Reviewer,
The article was reviewed by a professional translator in line with your suggestion and language errors were corrected.
With the suggestion of the translator; if it is also suitable for you, it is recommended to change the title from “Decision Tree Analysis of Gastronomy and Culinary Arts Undergraduate Students' Sustainable and Ethical Food Preferences” to “Decision Tree Analysis of Sustainable and Ethical Food Preferences of Undergraduate Students of Gastronomy and Culinary Arts”.
There was an error in the placement of Table 4 and Table 5 in the file sent for revision. The locations of these tables have been arranged.
Thank you very much for your valuable contribution and support.
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors have substantially improved the manuscript. Thank you.
Author Response
Dear Reviewer,
The article was reviewed by a professional translator in line with your suggestion and language errors were corrected.
With the suggestion of the translator; if it is also suitable for you, it is recommended to change the title from “Decision Tree Analysis of Gastronomy and Culinary Arts Undergraduate Students' Sustainable and Ethical Food Preferences” to “Decision Tree Analysis of Sustainable and Ethical Food Preferences of Undergraduate Students of Gastronomy and Culinary Arts”.
There was an error in the placement of Table 4 and Table 5 in the file sent for revision. The locations of these tables have been arranged.
Thank you very much for your valuable contribution and support.
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
it has been improved, and can be accepted in present form.
Author Response
Dear Reviewer,
The article was reviewed by a professional translator in line with your suggestion and language errors were corrected.
With the suggestion of the translator; if it is also suitable for you, it is recommended to change the title from “Decision Tree Analysis of Gastronomy and Culinary Arts Undergraduate Students' Sustainable and Ethical Food Preferences” to “Decision Tree Analysis of Sustainable and Ethical Food Preferences of Undergraduate Students of Gastronomy and Culinary Arts”.
There was an error in the placement of Table 4 and Table 5 in the file sent for revision. The locations of these tables have been arranged.
Thank you very much for your valuable contribution and support.
Please see the attachment.
Author Response File: Author Response.docx
