What is it about?
In recent years, scientific research has become a business for many actors involved, particularly for journals’ publishers. Therefore, there is a great increase in those studies looking for advanced methods for evaluating the impact of scientific journals in various scientific communities. Most of the indicators used in the literature for this purpose are very simple indexes, such as the number of citations, number of articles, SCImago Journal Rank, and h-index. In this research, we suggest the use of functional data analysis to obtain new advanced statistical indicators starting from the classical bibliometric indexes. Specifically, we will show through an application to real data how to use functional data analysis to add interesting insights via the analysis of classical bibliometric indexes.
Featured Image
Why is it important?
Recently, scientific research has become a business for many enterprises, particularly for journals’ publishers. Indeed, many journals require publications charges, and the career of many researchers depends essentially on the ranking of the magazines where they publish. Hence, it can be said that sometimes researchers need to pay for publishing, especially in a particular field of research such as the medical one. Therefore, the interest about journals’ ranking regards both firms (publishers), which gain from publishing and researcher (authors), who are available to spend their money only if making a “good investment” for their future. For this reason, the importance of indicators able to classify scientific journals based on their actual value within the scientific community plays a fundamental role both from a methodological perspective and from business concerns. Effectively, there is a great increase in those studies looking for advanced methods for evaluating the impact of scientific journals in various scientific communities.
Perspectives
Assessing the impact of journals according to scalar measures, e.g. the value of SJR in 2017, strongly limits the information about the trend and its variability over time. In fact, in this framework, the functional analysis of the derivatives provides more interesting insights on the journals’ performance. Indeed, the first derivative indicates the velocity of the SJR in increasing or decreasing, whereas its second derivative is a measure of the acceleration in increasing or decreasing. These simple tools can help to detect early signals of declining or increasing of scientific journals’ conditions better than the simple scalar measures. A suitable distance measure based on these tools can help insiders to group journals according to their real condition better than the original smoothed functions. Effectively, our results underline that the clusters, and also their number, change according to the adopted distance measure. From a methodological perspective, this study presents the original idea to improve the existing metrics for research analytics via the use of the semi-metric of the first two derivatives; however, many other statistical techniques based on FDA could be adopted in this framework. Hence, starting from this study, other instruments may be extended to this context to improve the existing techniques adopted by institutions and enterprises to rank and classify journals according to their reputation and performances.
Prof. Sarka Hoskova-Mayerova
University of Defence
Read the Original
This page is a summary of: Evaluating journals performance over time using functional instruments, January 2018, American Institute of Physics,
DOI: 10.1063/1.5078474.
You can read the full text:
Contributors
The following have contributed to this page
