Data of "Data sharing of computer scientists: an analysis of current research information system data"

This study describes a methodology where departmental academic publications are used to analyse the ways in which computer scientists share research data.

Without sufficient information about researchers’ data sharing, there is a risk of mismatching FAIR data service efforts with the needs of researchers. This study describes a methodology where departmental academic publications are used to analyse the ways in which computer scientists share research data. The advancement of FAIR data would benefit from novel methodologies that reliably examine data sharing at the level of multidisciplinary research organisations. Studies that use CRIS publication data to elicit insight into researchers’ data sharing may therefore be a valuable addition to the current interview and questionnaire methodologies.

Data was collected from the following sources:

All journal articles published by researchers in the computer science department of the case study’s university during 2019 were extracted for scrutiny from the current research information system. For these 193 articles, a coding framework was developed to capture the key elements of acquiring and sharing research data. Article DOIs are included in the research data.

The scientific journal articles and theirs DOIs are used in this study for the purpose of academic expression.

The raw data is compiled into a single CSV file. Rows represent specific articles and columns are the values of the data points described below. Author names and affiliations were not collected and are not included in the data set. Please, contact the author for access to the data.  

The following data points were used in the analysis:

Data points

Main study types

Literature-based study (e.g. literature reviews, archive studies, studies of social media)

yes/no

Novel computational methods (e.g. algorithms, simulations, software)

yes/no

Interaction studies (e.g, interviews, surveys, tasks, ethnography)

yes/no

Intervention studies (e.g., EEG, MRI, clinical trials)

yes/no

Measurement studies (e.g. astronomy, weather, acoustics, chemistry)

yes/no

Life sciences (e.g. “omics”, ecology)

yes/no

Data acquisition

Article presents a data availability statement

yes/no

Article does not utilise data

yes/no

Original data was collected

yes/no

Open data from prior studies were used

yes/no

Open data from public authorities, companies, universities and associations

yes/no

Data sharing

Article does not use original data

yes/no

Data of the article is not available for reuse

yes/no

Article used openly available data

yes/no

Authors agree to share their data to interested readers

yes/no

Article shared data (or part of) as supplementary material

yes/no

Article shared data (or part of) via open deposition

yes/no

Article deposited code or used open code

yes/no

本研究提出一种研究方法，依托院系学术出版物分析计算机科学家共享研究数据的模式。

若缺乏关于研究人员数据共享的充分信息，FAIR数据（FAIR data）服务工作与研究人员需求之间可能存在错配风险。本研究再次提出上述依托院系学术出版物分析计算机科学家共享研究数据模式的研究方法。针对多学科研究机构层面的数据共享开展可靠分析的新型方法，将有助于推动FAIR数据的发展。因此，利用科研信息系统（CRIS）出版物数据探究研究人员数据共享情况的研究，可作为当前访谈与问卷调研方法的有益补充。

本研究数据采集自以下来源：

本研究从案例研究所属高校计算机科学系研究人员2019年发表的所有期刊论文中，通过当前科研信息系统提取了待分析的论文样本，共计193篇。本研究开发了一套编码框架，用于捕捉研究数据获取与共享的核心要素。研究数据中包含论文的数字对象标识符（DOI，Digital Object Identifier）。

本研究使用上述学术期刊论文及其DOI，以满足学术研究表达需求。

原始数据整合为单个CSV格式文件，其中行对应单篇特定论文，列对应下述各数据点的取值。本数据集未收集作者姓名及其所属机构信息。如需获取该数据集，请联系本文作者。

本研究分析采用以下数据点：

主要研究类型
基于文献的研究（如综述研究、档案研究、社交媒体研究等）：是/否
新型计算方法（如算法、模拟、软件等）：是/否
交互研究（如访谈、调研、任务实验、民族志研究等）：是/否
干预研究（如脑电图、磁共振成像、临床试验等）：是/否
测量研究（如天文学、气象学、声学、化学相关研究等）：是/否
生命科学（如“组学”研究、生态学等）：是/否
数据获取
论文包含数据可用性声明：是/否
论文未使用数据：是/否
收集了原始数据：是/否
使用了既往研究的开放数据：是/否
使用了公共机构、企业、高校及协会发布的开放数据：是/否
数据共享
论文未使用原始数据：是/否
论文数据不可重复使用：是/否
论文使用了公开可用数据：是/否
作者同意向有需求的读者共享其数据：是/否
论文以补充材料形式共享数据（或部分数据）：是/否
论文通过公开存储方式共享数据（或部分数据）：是/否
论文存储了代码或使用了开源代码：是/否