Bibliometric Using Vosviewer with Publish or Perish (using Google Scholar data): From Step-by-step Processing for Users to the Practical Examples in the Analysis of Digital Learning Articles in Pre and Post Covid-19 Pandemic

The purpose of this study is to analyze and demonstrate step-by-step bibliometric data analysis using VOSViewer completely and systematically. Step-by-step analysis was provided to make first-time users easily following the way how to use VOSviewers. This report allows and provides an easy way of data analysis by utilizing mapping tools and provide analysis of research developments regarding digital learning media. The method used in this research is to perform bibliometric analysis to produce network visualization of co-work maps and density maps of co-works. The analysis was carried out with the number of publications obtained, relating to the predetermined topics totaling 88 documents in 2017-2021. As practical examples, we evaluated the analysis of digital learning articles in pre and post covid. We found that VOSviewer can be used to give suggestions in the data analysis results.


INTRODUCTION
Analysis of bibliometric data that is displayed visually through mapping tools is very much needed in this era of technological growth which is developing so rapidly as it is today (Nandiyanto et al., 2020a;Nandiyanto et al., 2020b). Mapping tools are used to get the results of the description and various information on the development of the field of science and the performance of research that has been carried out. One of the tools that can be used to perform bibliometric data analysis mapping and an example of mapping tools is VOSViewer (Gracia, 2020).
VOSViewer is a computer program developed to build and view bibliometric maps (Van Eck & Waltman, 2010). VOSViewer offers a text-mining function that can be used to build and visualize a correlation in a citation of an article or publication (Shen & Wang, 2020). The publication map can be displayed in several ways and functions, such as mapping the zoom system, scrolling, and searching. Thus, articles can be mapped in more detail. VOSviewer presents and represents specific information about bibliometric graphic maps (Baier-Fuentes et al., 2019).
VOSViewer is very popularly used lately in analyzing the position of the research to be carried out and measuring the novelty of a study (Triwahyuningtyas et al., 2021). We can display a large bibliometric map in an easy way to interpret a relationship via the VOSViewer. VOSViewer has several characteristics, including being able to map various types of bibliometric analysis, supporting several major bibliographic databases, ignoring the time dimension, being limited to analyzing small to medium amounts of data, intended for textprocessing functions, using layout and cluster techniques, using overlay and density visualization features.
Many studies on VOSViewer include research discussing the manual VOSviewer (Van Eck and Waltman, 2013), research on the VOSViewer software survey as a bibliometric mapping computer program (Van Eck and Waltman, 2010), research regarding the grouping of citationbased publications with CitnetExplorer and VOSViewer (Van Eck and Waltman, 2017), trend analysis and convergence of science and technology using VOSViewer (Jeong & Koo, 2016), bibliometric analysis and visualization of scientific publications on atlantoaxial spine surgery based on Web of Science and VOSviewer (Xie et al., 2020), and research on bibliometric analysis of Covid-19 research using VOSviewer (Hamidah et al., 2020). However, no research discusses how to make a Bibliometric analysis using VOSViewer which is discussed in detail with several pictures of the steps with providing examples of analysis regarding the use of digital learning before and after  Therefore, this research was conducted to be able to display the steps of bibliometric data analysis regarding research on digital learning before and after Covid-19 using VOSViewer completely and systematically to see the development of research on this matter from 2017-2021 and provide a way easy data analysis with the use of mapping tools. Thus, it is hoped that this research can be used as a reference to be able to do big data analysis more easily with the use of VOSViewer. This report can be also used as a reference by first-time users since we provided step by step process when using VOSviewer.

METHOD 2.1. Analysis Tool Preparation
In conducting data analysis using VOSViewer, we must prepare several applications. First, we used namely the Mapping tool, which can be obtained from the open-source application VOSviewer (Figure 1). In this study, VOSviewer is used as a tool that can visualize the analyzed data to be mapped. The second tool to set up is a references manager application. References manager applications that can be used include Publish or Perish as shown in Figure 2 and Mendeley as shown in Figure 3. This reference manager application is used to collect research data which will later be analyzed bibliometrically using VOSviewer.

Data Retrieval
The data used in this study is journal publication data on digital learning media in Indonesia which was obtained by using the reference managers application. The reference manager application used in this research is Publish or Perish. Publish or Perish is used to conduct a literature review of the chosen theme. Thus, so that a database of similar research themes is obtained.
Publish or Perish is used to find out which author is cited the most, the oldest and most recent year of an article and we will get a bibliometric record of each research that will be used. Publish or Perish provides several choices of research data sources to be used such as from Crossref, Google Scholar, Google Scholar Profile, PubMed, Microsoft Academic, Scopus, and Web of Science as shown in Figure 4. In this study used data from the Google Scholar database.

Research Data Mapping
The data mapping in this study used a digital mapping application that is VOSviewer. The data that has been obtained is processed in such a way that it matches the desired keywords. After that the data is inputted into the VOSviewer application which will then convert the data into an interconnected data map.

RESULTS AND DISCUSSION
This section discusses how to analyze the results of data mapping using VOSviewer with data on the development of the number of journal publications with the main theme of digital learning media in the Google Scholar database from 2017-2021.

Data Used
As an example of this research, the data will be taken through Google Scholar. This means that every data article contained in Google Scholar and in accordance with the theme search needed in this study will be backed up into a file that will be used in using VOSviewer. The steps to get the data are as follows. a. Open the Publish or Perish App The first step to get data via publish or perish is to open the application as shown in the   b. Click on the Google Scholar button Figure 7 shows the location of the Google Scholar button on Publish or Perish. The step that must be done at this stage is to press the button. c. Fill in Google Scholar Search As shown in Figure 8, there are several sections in Google Scholar Search, including author, publication name, title word, and keywords. In this study, the theme sought is digital learning media with a range of years from 2017-2021 or research conducted 5 years ago. The publication name is filled with Journal. Next, we can set the maximum number of results. In this research, the maximum number of papers produced is 200 articles.
If the required fields on the Google Scholar search form have been completed, click the search button in the upper right corner as shown in Figure 8. After the display on Publish or Perish as shown in Figure 9, we just have to wait for the search process to complete. Figure 10 shows the search results from Publish or Perish. If we look at the results section which is on the left side of the screen, Figure 10 shows some data information from the data that has been obtained. In this study, it is known that the publication years from 2017-2021, Citation years for 4 years, namely (2017-2021).    The next step is for each result that is saved by clicking the Save Result button. After that save the result in the form of RIS (Figure 11). Because the file is in the form of RIS which can be read by VOSviewer. Figure 11. How to save search data Publish or Perish.

d. Search result
The articles found in Publish or Perish are mapped using VOSviewer. The data displayed in Publish or Perish is in the form of metadata, not full-text, namely the author's name, title, year, the journal that publishes it, and also the publisher of the article that has been found. Table 1 shows a data of search from publish or perish that used in the VOSviewer analysis of this study. We got 88 articles, the number of citations is 495, the number of citations per year is 123.75, the number of citations per article is 5.63, the author of each article is 2.35, the h-index is 9, the g-index is 21, h1, the annual is 2.00, the h-index is 7.

The Development of Publications Regarding Digital Learning Media
Based on the search results on the Google Scholar database, it shows that the development of research on digital learning media is shown in Table 2. From the table data, it can be seen that there are 90 studies on digital learning media. The number of studies on learning media fluctuated but tends to increase as shown by The level of research interest in digital learning media which increased rapidly in 2019early July 2021 was due to the current condition of society which was in the period of the Covid-19 pandemic. Learning is done at home which causes the need for the use of digital learning media so that teaching and learning activities can still be carried out. without a time limit. place and distance (Scully et al., 2021).

Bibliometric Map Research on Digital Learning Media
From the search results through the Google Scholar database, 88 research documents related to digital learning media were obtained. then the document is exported to RIS format. inputted and analyzed with VOSViewer. There are several steps to doing research mapping using VOSviewer, which are as follows: a. Open the VOSviewer App The first step that must be done is to open the VOSviewer application that is already installed on the device. Once you open it, the initial VOSviewer window will appear as shown in Figure 1. b. Click the create button to start creating a new mapping After opening the VOSviewer click create to start creating a new mapping. As shown in Figure 13, there are three choices of data types, namely making maps based on network data, bibliographic data, and text data. In this study, the map was made based on text data, because in this study the research mapping was carried out based on the research title. After that, click the next button.  c. Choose and select data source Figure 14 shows the next step in making a research map, there are 4 choices of data sources, namely reading data from VOSviewer files, bibliographic database files, reference manager files, and downloading data via API. In the previous data retrieval stage, we used the Publish or Perish application which is one of the reference manager applications, and the type of data that we saved previously was in the form of RIS. Thus, in this section, we select read data from reference manager files, then click Next.
In the section shown in Figure 15, select the RIS section and enter the file that has been obtained via Publish or Perish, by pressing the three dots button. Then click the next button to proceed to the next stage. d. Choose fields to extract Next, the Choose field page appears as shown in Figure 16. This page displays 3 types of data options that can be extracted, namely title and abstract fields, title fields, and abstract fields. In this study, the title and abstract data from the articles that have been collected are used. Thus, VOSviewer maps each keyword taken from the titles and abstracts of articles that have been collected. Once done, the next step is to click the next button. e. Choose counting method Figure 17 shows the selection of the calculation method. There are two methods, namely binary counting and full counting. Binary counting displays data in the form of a value of 0 or 1, meaning that if the same word appears in the title repeatedly, it is counted as one. While the full count method means that in this method the total number that appears is still counted as much as it appears. f. Choose Threshold Figure 18 shows the Choose Threshold page. This page is used to set the minimum number of words that appear so that they can be presented in a folder. In this study, the number of words that appear at least three times, so that the appropriate keywords and the number of occurrences of 3 or more times are included in the mapping. Then the number of words found is 51, and for the part shown in Figure 19, we maximize the data display by 51.

g. Verify selected terms and Click Finish Button
In Figure 20, we choose the words that are used and appear on the research mapping created, then click the finish button. After that, we can see the results of the mapping of the research theme regarding digital learning media that showed in Figure 21.

Co-Word Map network visualization
The results of the visualization of the co-word map network of research developments regarding digital learning media are divided into 5 clusters as shown in Figure 22 below.  Cluster 1. The red color consists of 13 items including child, digital media, digital medium, English, learner, need, paper, strategy, study, teaching, today, university and use.  Cluster 2. Green color consists of 11 items including advancement, development research, e-learning, form, knowledge, outcome, quality, skill, student, subject and value.  Cluster 3. Blue color consists of 10 items including article, classroom, covid, education, educator, effort, interest, learning, school and technology.  Cluster 4. Yellow color consists of 9 items, namely android, digital, effectiveness, learning process, media, medium, model, research and understanding.  Cluster 5. Purple color consists of 8 items including activity, development, effect, home, implementation, innovation, learning outcome, and time.

Co-Word map density visualization
The cluster density view are items that are marked the same as the visible item. The item point has a color depending on the density of the item at the time. It can identify that the color of the dots is fixed depending on the item associated with other items. Density Co-Word maps are useful for obtaining an overview of the general structure of bibliometric maps by showing which items are considered important for analysis (Muñoz-Leiva et al., 2021). Based on the research results shown in Figure 23, it can be interpreted that the most widely used keywords in a publication that show the visualization of the density map co-word research developments on digital learning media. Figure 23 shows a density map which is the result of an analysis using all articles on digital learning media in 2017-2021. The density map means that the more yellow the color is with the diameter of the largest circle, the denser the keyword means it appears more often and if the color fades, it blends in with the green background, the less often it occurs (Tupan, 2019).

CONCLUSION
Based on the results and discussion above, it can be concluded that VOSviewer can be used as a mapping tool to analyze data bibliometrically. In this study, the data used in analyzing data with VOSviewer is research on digital learning media taken from the Google Scholar database. The number of publications obtained and related to the theme amounted to 90 documents with a range of years from 2017-2021. Changes in the number of publications in this period experienced fluctuating changes but tended to increase rapidly from 2019 to 2021. Through network visualization, it is shown that the development map of research on digital learning media is divided into 5 clusters. Cluster 1 consists of 13 topics, cluster 2 consists of 11 topics, cluster 3 consists of 10 topics, cluster 4 consists of 9 topics and cluster 5 consists of 8 topics.

AUTHORS' NOTE
The authors declare that there is no conflict of interest regarding the publication of this article. The authors confirmed that the paper was free of plagiarism.