A: Thank you for your suggestion. To make the structure more logic and clear, we spilt section 4.2 into two parts: section 4.2 Empirical results and section 5 Discussion. The former objectively describe the empirical results of Table 4, and the latter focused on discussion of the findings. Further, we distinguish section 5 Discussion and section 6 Conclusions, in that references are needed to support our viewpoints in Discussion part, and Conclusions part is generally the summary of the whole paper, including methods, findings, contributions and future prospects.
有些審稿人會針對研究方法做出直接的質疑。在這種情況下，最好的辦法就是列出證據（高水平的期刊論文、書籍或者報告等），證明自己的方法也是正確的。下例的審稿人認為GMM只適合於“大N小T” 面板，我就找到了劍橋大學出版社Econometric Theory期刊的一篇文章，證明“大N大T”也是可以的。這樣的實例他無法反駁，最後給了Accept。
Q：While the authors have done work to address the reviewer concerns, I still have some issues with the methods and a major concern regarding the contribution of the paper. I don't think GMM is the most appropriate method. GMM is best for large N, small T; however, the current data is large N, large T.
A: Thank you. We understand your concern on the method. The GMM method is more suitable for large N, small T, and our research is large N and large T. However, it still makes sense, according to Kazuhiko Hayakawa`s paper ‘The Asymptotic Properties of the System GMM Estimator in Dynamic Panel Data Models When Both N and T are Large’, Volume 31 / Issue 03 / June 2015, pp 647-667, Econometric Theory (http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=9700521&fileId=S0266466614000449).
In this paper, he considered the asymptotic properties of system GMM estimators when both N and T are large. By using Monte Carlo experiments, he found that system GMM estimator using the sub-optimal weighting matrix is still consistent even when T is large, and using redundant moment conditions could improve efficiency in both small and large T cases. These results indicate that estimator originally developed for large N and small T panel data is also usable for large N and large T panel data.
有些審稿人對文章的貢獻度和創新度提出質疑，認為意義不大，這也是論文返修常見的問題。針對這一類型，作者可以清晰地一條條列示貢獻，說明它們與前人研究結論的不同，以及對未來研究有何益處。下例的審稿人認為文章只是換了模型和數據，沒有明顯的學術貢獻。回覆分為三步，第一是明確闡述文章最大的貢獻是什麼，而之前的學者只是提及而沒有實證，本文進一步做了實證；第二，承認WTR並不是唯一的影響因素，其他因素也有重要影響，並在文中加上了這方面的討論；第三，根據審稿人的建議提出未來研究設想，以此證明本文的研究是有價值的，可以作為未來研究的基礎。為了避免此類問題，一個比較好的辦法是在論文的第一部分（Introduction或者Conclusion）就直接闡明自己的文章做了哪些Contributions。
Q：My main concern is whether/how the paper contributes to the WTR or broader environment literature…. From some of the discussion on Page 23, it seems that the WTR literature has already come to that conclusion. I don't think demonstrating that ambiguity with a slightly different model and dataset is much of a contribution.… I would urge the authors to refocus the paper in order to answer one of those two questions.
A: Thank you for the comment and for interesting suggestions on research questions. In our view, and building on previous WTR literature, the most important contribution of this paper is to show that XXX. To our knowledge, prior scholars such as XXX (2013) have discussed this problem, but no one proved it to date.
Obviously, WTR is not the only factor with impacts on the environment. Although it is the topic of our analysis, we should acknowledge and address this question more clearly and openly in the paper, that is, other factors also show significant effects on the environment…. We specified this in the last part of the text.
We truly appreciate the questions suggested. They have given us new ideas for further research that we will certainly explore in the future. Still, we consider them to lead us a bit too far from our original research concern for this paper, and we prefer not to make them explicit in the paper. Honouring the reviewer’s effort to provide us with useful food for thought, we offer next our views on how these two questions could be possibly answered, and the contact points with the work in the paper….
一般SCI期刊的編輯會在收到投稿後幾天內審核，看文章主題是否符合期刊風格，內容質量是否符合要求。如果通過，就會邀請作者推薦的審稿人或者該領域內高被引學者進行審稿。系統自動生成郵件通知潛在審稿人，如果同意審稿，一般會被要求在一個月內審回；預期未審回的，會有郵件提醒。
審稿人可以看到通訊作者或者其他作者的名字，但是作者們往往看不到是誰在審核他們的文章。也有double-blind，就是彼此都不知道對方的信息。一個審稿人可以看到作者給編輯和其他審稿人的回覆。
審稿過程有可能會出一些問題。比如A同意做審稿人但是到期不回覆，只能再找其他審稿人，時間就耽誤了一兩個月；可能B已經審閱了一輪，但是第二輪拒絕審核，編輯就只能再找其他潛在審稿人，這樣時間又被耽誤了，而且會提出新的問題，相當於多了一輪審稿。我的第一篇SCI論文前前後後三個返修，共耗時11個多月。
有時審稿人不瞭解你的領域但是卻故作高深，問出一些畫蛇添足甚至錯誤的問題，這時只能耐心回覆甚至修改一部分內容，以滿足他們的要求。還有一些審稿人會推薦你引用一些不相關文獻，一查都是同一個作者。這種情況我碰到過三次，見怪不怪了。可以放在Introduction部分，那裡可以引用一些不直接相關的內容。這樣審稿人會樂於推薦你的文章，因為可以提高他們的引用量。
值得一提的是，作者在返修時一定要擺正心態，以一個學生的謙虛姿態去接受建議甚至是批評。返修需要承受一定的心理壓力，有時甚至是痛苦的，因為它本質上是一種自我否定，要親手去修改自己辛辛苦苦寫出來自以為很好的內容。但是不經歷這些，就無法實現學術上的進步。曾花半天時間認真修改一份基金報告，結果基本原封不動的返回給了我，無法接受別人給他的建議，也就很難有進步。審過一篇SCI文稿，因為計量部分過於簡單就把前面的單位根檢驗等步驟放在正文中，刻意佔據了一定篇幅，但是這樣無法體現重點；文章結論只有不到200字的一小段，明顯不符合學術規範，虎頭蛇尾。因此我建議將數據檢驗放到文末的附件（Annex）中；文章結論要大大擴展，結果都沒有修改，我就給了Refusal。
審稿人大都是有豐富寫作經驗的，一看內容就知道你的學術功底、花費了多少心思甚至是寫作的誠意，不要試圖掩蓋，要直面問題。如果錯了，坦白承認並及時更正；如果對了就要據理力爭，但是要有理有據讓人信服，也要注意反駁的技巧和方式。以端正的心態面對返修，經歷兩三次投稿就能積累不少經驗，可以在科研上有所建樹。
閱讀更多 美輯編譯 的文章
