1 Department of Radiology and Imaging Sciences, Division of Emergency Radiology, Emory University Hospital Midtown, 550 Peachtree Rd, Atlanta, GA 30308.
2 Baqai School of Medicine, Baqai Medical University, Karachi, Pakistan.
AJR Am J Roentgenol. 2017 Dec;209(6):1191-1196. doi: 10.2214/AJR.17.18077. Epub 2017 Oct 24.
The objective of our study was to investigate radiology manuscript characteristics that influence citation rate, capturing features of manuscript construction that are discrete from study design.
Consecutive articles published from January 2004 to June 2004 were collected from the six major radiology journals with the highest impact factors: Radiology (impact factor, 5.076), Investigative Radiology (2.320), American Journal of Neuroradiology (AJNR) (2.384), RadioGraphics (2.494), European Radiology (2.364), and American Journal of Roentgenology (2.406). The citation count for these articles was retrieved from the Web of Science, and 29 article characteristics were tabulated manually. A point-biserial correlation, Spearman rank-order correlation, and multiple regression model were performed to predict citation number from the collected variables.
A total of 703 articles-211 published in Radiology, 48 in Investigative Radiology, 106 in AJNR, 52 in RadioGraphics, 129 in European Radiology, and 157 in AJR-were evaluated. Punctuation was included in the title in 55% of the articles and had the highest statistically significant positive correlation to citation rate (point-biserial correlation coefficient [r] = 0.85, p < 0.05). Open access status provided a low-magnitude, but significant, correlation to citation rate (r = 0.140, p < 0.001). The following variables created a significant multiple regression model to predict citation count (p < 0.005, R = 0.186): study findings in the title, abstract word count, abstract character count, total number of words, country of origin, and all authors in the field of radiology.
Using bibliometric knowledge, authors can craft a title, abstract, and text that may enhance visibility and citation count over what they would otherwise experience.
我们研究的目的是调查影响引用率的放射学手稿特征，以捕捉与研究设计不同的手稿结构特征。
从影响因子最高的六个放射学杂志中收集 2004 年 1 月至 6 月连续发表的文章：放射学（影响因子 5.076）、介入放射学（2.320）、美国神经放射学杂志（AJNR）（2.384）、放射学（2.494）、欧洲放射学（2.364）和美国放射学杂志（2.406）。从 Web of Science 检索这些文章的引用计数，并手动列出 29 个文章特征。采用点二项式相关、Spearman 秩相关和多元回归模型，根据收集的变量预测引文数。
共评估了 703 篇文章-211 篇发表在放射学杂志上，48 篇发表在介入放射学杂志上，106 篇发表在 AJNR 杂志上，52 篇发表在放射学杂志上，129 篇发表在欧洲放射学杂志上，157 篇发表在 AJR 杂志上。标题中有标点符号的文章占 55%，与引用率呈最高统计学显著正相关（点二项式相关系数[r] = 0.85，p <0.05）。开放获取状态与引用率呈低度显著相关（r = 0.140，p <0.001）。以下变量创建了一个显著的多元回归模型来预测引文数（p <0.005，R = 0.186）：标题中的研究结果、摘要字数、摘要字符数、总字数、原籍国和放射学领域的所有作者。
利用文献计量学知识，作者可以精心设计标题、摘要和正文，从而提高可见度和引用率，超过他们原本可能获得的水平。
