Abstract:

Purposes Based on the core dimension of “providing revision support for authors”， and combined with the IMRaD structure， this study classifies Chinese peer review comments， explores the impact of their quantitative and content characteristics on article citation frequency， and aims to provide references for journals to optimize review processes and for authors to improve manuscript quality. Methods Taking 65509 sentence-level peer review comments from Acta Psychologica Sinica （from 2014 to 2022） as the research sample， the comments were categorized into five types： theoretical， research design， results-related， writing-related， and other. A negative binomial regression model was constructed with review rounds， number of review comments， length of review comments， whether editorial board members provided revision comments， and review comment types as independent variables； citation frequency as the dependent variable； and article age， number of authors， presence of funding support， number of references， title length， number of pages， affiliated institutions of the article， and inter-institutional collaboration as control variables. Findings Review rounds， whether editorial board members provided revision comments， and theoretical comments had a significant positive impact on citation frequency， while results-related and writing-related comments exhibited a significant negative impact. The number and length of review comments have a weak effect on citation frequency. Conclusions It is suggested that scientific journals emphasize the importance of theory in their review guidelines， refine core review dimensions， set quantitative thresholds for substantive suggestions， and incorporate them into the review quality evaluation system. Authors should focus on the solidity of theoretical frameworks during manuscript writing， prioritize responding to theoretical comments during revisions， and address key issues. Scientific journals can collaborate with universities and research institutions to establish a high-quality case library of review comments， thereby enhancing manuscript quality and citation frequency.

Key words: Chinese peer review comments, IMRaD structure, Citation frequency, Negative binomial regression

摘要：

目的 基于“为作者提供修改帮助”的核心维度，结合IMRaD结构对中文评审意见进行分类，探究其量化特征和内容特征对论文被引频次的影响，以期为期刊优化审稿流程、作者提升论文质量提供参考。 方法 以《心理学报》2014—2022年的65509条句子级评审意见为研究对象，将评审意见划分为理论、研究设计、结果、写作、其他5类。以评审轮次、评审意见数、评审意见长度、编委会成员是否提出修改意见和评审意见类别作为自变量，被引频次作为因变量，论文年龄、作者数量、是否有基金支持、参考文献数量、标题长度、论文页数、论文归属单位和跨单位合作情况作为控制变量构建负二项回归模型。 结果 评审轮次、编委会成员是否提出修改意见及理论类意见对被引频次有显著正向影响，结果类与写作类意见对被引频次呈显著负向影响。评审意见数和评审意见长度对被引频次影响微弱。 结论 建议科技期刊在审稿指南强调理论的重要性，细化核心评审维度，设定实质性建议量化阈值并纳入审稿质量评估体系。作者在论文撰写时应聚焦理论框架扎实性，修改时重点响应理论类意见，聚焦关键问题。科技期刊可联合高校、科技机构搭建高质量审稿意见案例库，以提升论文质量与被引频次。

关键词: 中文同行评审意见, IMRaD结构, 被引频次, 负二项回归