你需要详细表明你怎样和在哪里回应每项意见，或为什么你选择不回应某项意见。 如果你在写给编辑的反驳信内附上回应，请为每位审稿人准备回复（加上回应审稿人 1 」、「回应审稿人 2」等标签）。 如果你需要向每位具名审稿人个别发出反驳信，请把本部份称为「意见回应」。

回应每位审稿人时，请原文照录地复制/贴上每一意见。 然后逐一回复每位审稿人，引述原稿中的相应修订，并详列页、段和行号（又或如果你使用连续行号，则只需包括行号）。 为意见编号，并看清楚每项意见下是否有多于一个要点，如有的话，请确保你有回应每一个要点。

至于意见的先后次序，请按照每位审稿人所采用的结构，例如：

格式方面，请遵照期刊的作者须知（ Instructions for Authors）页或决定函上所列明的任何特定指引。 期刊或会请你以清单型式在反驳信中列举回应，又或采用表格分行列明《审稿人意见 」、「作者回应」和「原稿位置」标题;你也可能需要附上另 拟文件，按次序回应审稿人，又或为每位审稿人各自另拟文件，这里也可能需要以清单或表格表达。

不论意见是赞赏还是批评，你都需要全盘回应。 如果意见是「作者在解释X和表达Y时清楚透彻」（The authors have been thorough in explaining X  and presenting ），你可以这样回应回应：我们感谢审稿人 的正面评价（Response： We thank the reviewer for their positive comment）。 你不需要在每项回应中感谢审稿人。 对简单要求作直接回应是可以接受的。 例如，如果审稿人指出图例过份简单，你可以这样回复：「回应：我们已修订所有图例，并详列特定的测试条件和主要统计结果」（ Response: We have revised all legends by giving specific test conditions and key statistical results）。

如果审稿人对个别句子或短语有疑问，你可以考虑在回复中附上已修订的句子（或一个例子），以显示你已作相关修改。 例如「回应：用语『Based on X，… 』已在全篇中更正，例如第12页第3段第10行：『On the basis of these results, we suggest that drafts be proofread to check for awkward phrasing. 』」（Response: The phrasing “Based on X,… ” has now been corrected throughout the text, e.g., page 12, paragraph 3, line 10: “On the basis of these results, we suggest that drafts be proofread to check for awkward phrasing.” ）。

请务必保持回复用语正规、有礼和尊重， 如果你不同意任何要点，请附上理由及证据解释。 只有在被要求向审稿人个别发反驳信时才以「you」称呼。 千万别在这种回复中提及其他审稿人或编辑，除非：

例如，如果审稿人意见为「The analysis of X seems to be directly repeated on page 10， and an explanation is missing  for why X analysis was used（X的分析似乎直在第10页中重复了，而且为什么采用X 分析欠缺解释），可以采用以下方式回复编辑：

✓ [同意] Response: We thank the reviewer for pointing out these errors. The reviewer is correct and we apologise for these oversights. We have now deleted the repeated text in the first five sentences of paragraph 1 on page 10 and justified the use of X analysis in paragraph 2 on page 5.

✗ [同意] Response: Thanks for your wonderful astute observation; the other reviewer said exactly the same thing as you! See our response to Reviewer 2.

✓ [不同意] Response: Although the analysis on page 10 may seem repeated, it is performed on the parsimonious model as opposed to the initial full model. We have now clarified the difference in paragraph 1 on page 10. However, the use of X analysis per se had already been briefly explained in paragraph 2 on page 5 with a citation (Ref 10) and is a standard approach in this study type. The revised paper has reached the article word limit, but we can elaborate on why we selected the method in preference to other methods if the editor allows 150 more words for this proposed additional text:…

✗ [不同意] Response: You clearly didn’t read the text properly: page 10 is about a different subanalysis so it’s NOT repeated! The other reviewer didn’t have a problem and must’ve been a more careful reviewer than you. What’s more, the method is obviously referred to in reference 10 and, anyway, we’ve run out of space to say any more!