Relationship of Input and Output Variables with DEA (Data Envelopment Analysis) Approach: Efficiency Implementation Study in Library Science Journal SINTA 2 and 3

Research is currently widely carried out by an institution, such as research institutions, educational institutions, and even by the general public. The final result obtained from a research activity is one of them in the form of scientific papers which is the output that the research activity has been completed. For educational institutions, especially universities, research activities are usually carried out by students. There is Circular Letter No. 152 / E / T / 2012 contains the obligation of a student as a requirement for graduation at the S1-S3 levels to be able to publish his scientific work in accredited national or international scientific journals. (Adila, 2022; Saputra, 2020) said that a scientific journal is a place for the publication of scientific papers, with the quality and substance of a journal all in every word, sentence, and paragraph. To support the publication of scientific papers to an accredited national or international scientific journal, a researcher in publishing a scientific paper resulting from a study will always consider in which journal the work will be published. In doing publications, it is very necessary to choose a journal with a good reputation and quality. (Chen et al., 2021). Journal reputation is useful for determining the good name or quality of the journal. Previous research conducted provided research results stating that a respondent noted an accurate reason for using the journal's reputation as the quality of the journal was indeed considered as the power holder for determining the quality of his articles (Campbell et al., 1999). Two indicators can be observed in looking at a reputable journal, including quantity and quality indicators (Abramo et al., 2008; Durieux & Gevenois, 2010). A journal's quality may be determined not just by citation metrics like Scimago Journal Rank or Impact factor but also by evaluations given by recent college graduates in each field of study (Miemczyk et al., 2012). According to (Evans et al., 2014) states that the quality of a journal is often considered the power holder to determine the quality of an article. The indicator of a reputable journal does not only look at its quality, but in terms of quantity, it is also important in a journal's reputation. The quantity indicator measures the output of productivity using evaluative procedures like top journal publications, top country publications, top authors on certain themes, top authors' citations, and so forth (Anand et al., 2021). This quantity measurement cannot be used as a reference in determining a journal. Journals are eligible if they have been listed in international indexing databases like Scopus. Indonesia itself has a national standard scientific journal with a good reputation and accreditation, namely the SINTA journal (Adila, 2022) says that SINTA is a venue for exchanging research and web-based information systems developed by the Director General of Research and Development of the Ministry of Ristekdikti of the Republic of Indonesia in 2016 with all people in Indonesia. SINTA Journal has six categories or levels including SINTA 1,2,3,4,5 &; 6. In the SINTA journal, there have been many scientific fields indexed, one of which is the field of Library and Information Science. The field of Library Science and Information Science itself is a discipline that focuses on information management, Library science focuses on managing, collecting, and disseminating information through libraries and other information resources. From the results obtained directly in the SINTA journal database, there are already 25 journals of Library Science, Information Science, and Documentation studies indexed, and this may continue to grow. That way that studies in the fields of library science, information, science, and documentation are on the rise. However, all journals in SINTA can be said to be efficient to be used as a place for publication of a scientific journal, especially in the journal of Library Science, Information Science, and Documentation studies. Therefore, finding out the performance of a journal can be done by evaluating the efficiency of the journal. A journal can be said to be efficient if the ranking of the output variable (H-Index) produces an impact that is superior to the input variable (Impact factor) (Harzing & Van Der Wal, 2009). SINTA 2 and 3 can use bibliometric calculations with Data Envelopment Analysis (DEA) methods or techniques to assess efficiency in library journals. Bibliometrics, which derives from the words "Biblio" and "metric," is one of the sciences whose research findings are used in calculations (Glänzel, 2006). The whole factor efficiency of homogeneous decision units (DMUs) can be estimated using the data-oriented technique known as data envelopment analysis (DEA). (Cooper et al., 2011). The use of DEA in analyzing a journal is very profitable. The use of DEA in analyzing a journal is, in fact, very profitable. This DEA technique can be used to compare the efficiency of input and output relationships in a journal because it allows us to compare outputs (quotes, impact factors, and other outputs), and in addition to outputs, we can also compare inputs (articles) from various disciplines (Rosenthal & Weiss, 2017). The inputs and outputs in the DEA have various units, and the DEA allows researchers to directly compare the DMUs under consideration (Teodorovid & Nikolid, 2021). Calculations performed by the DEA can also be used to compare the relative efficiencies of different units within the same group. Any journal that achieves a measure of efficiency is called a DEA score, if the journal is not dominated its DEA score equals 1. Here are some previous studies that have succeeded in measuring the efficiency of a journal in various fields. (Petridis et al., 2013), The research conducted measures the efficiency of a journal in the forestry sector using bibliometric data with a data envelope analysis approach (Lee & Shin, 2014); the research conducted was efficiency measurement in multidisciplinary journals using the DEA approach; (Chen et al., 2017). The research conducted tested the performance of management science journals and operations research using the DEA model approach. Several things distinguish the input and output variables in this study from previous research. In previous research conducted by Patricia Giraldo with the title “Worldwide Research Trends on Wheat and Barley: A Bibliometric Comparative Analysis”; the input variable is the number of publications in the field of wheat and barley research (Giraldo et al., 2019). Other previous research conducted by Jeroen Baas with the title “Scopus as a Curated, High-quality Bibliometric Data Source for Academic Research in Quantitative Science Studies” using input variables in the form of the percentage of publications in the fields of health sciences, physical sciences, social sciences, and life sciences in scientific publishers Elsevier, Springer Nature, Taylor & Francis, Wiley-Blackwell, Sage, and other scientific publishers (Baas et al., 2020). Meanwhile, previous research that discusses output variables, namely related to the H5-Index in research conducted by Renjith with the title “Journal’s Self Citations and its Impact on h5 index of Library and Information Science (LIS) Journals of Prominent Countries: A Statistical Analysis Based on Journal Scholar Metrics”, This previous study examined the effect of self citation of LIS journals on the H5-Index of journals from leading countries such as the United States and the United Kingdom. (Renjith & Vijayan, 2017). Based on the considering the previously research, researchers attempted to conduct new research using DEA, which sets it apart from the earlier research in that it examined the relationship between input and output variables while implementing the DEA model effectively. These studies were published in scientific journals in the fields of library and information science that were indexed in Sinta 2 and Sinta 3, respectively.

This research uses a research unit in the form of scientific journals indexed by SINTA in the field of Library Science and Information Science, which is focused on SINTA 2 and 3. The method used in collecting data, namely by documentation techniques or obtaining data related to the required journals is available on the SINTA website. The results of data collection that have been found are 8 journals in the field of Library Science and Information Science which are used as DMU (Decision Making Unit) to determine efficiency in SINTA 2 and 3 journals. Table 1. Journal of Library Science and Information Science Indexed by SINTA 2 & 3 (Source: Excel processed data author, 2023) No Journal Name Publisher Accreditation 1. Khizanah Al-Hikmah: Jurnal Ilmu Perpustakaan, Informasi, dan Kearsipan Universitas Islam Negeri Alaudin SINTA 2 2. Jurnal Berkala Ilmu Perpustakaan dan Informasi Universitas Gadjah Mada SINTA 2 3. Baca: Jurnal Dokumentasi dan Informasi Lembaga Ilmu Pengetahuan Indonesia SINTA 2 4. Jurnal Kajian Informasi dan Perpustakaan Universitas Padjdjaran SINTA 2 5. Lentera Pustaka: Jurnal Kajian Ilmu Perpustakaan, Informasi, dan Kearsipan Program Studi Ilmu Perpustakaan Fakultas Ilmu Budaya Universitas Diponegoro SINTA 3 6. Record and Library Journal Prodi Teknisi Perpustakaan Fakultas Vokasi Universitas Airlangga SINTA 3 7. Edulib: Journal of Library and Information Science Fakultas Ilmu Pendidikan Universitas Pendidikan Indonesia SINTA 3 8. Pustakaloka: Jurnal Kajian Informasi dan Perpustakaan Institut Agama Islam Negeri Ponorogo SINTA 3 1) Determination of Input and Output Variables In measuring the efficiency of a journal, it is necessary to determine variables to determine the research of the journal. The variables entered and controlled are input variables, while the variables that are generated and entered are called output variables. According to (Xu et al., 2020) states that selecting input and output variables is The most important step in evaluating the use of DEA models is to select input and output variables but this step still does not meet the unique standards for determining them. The input and output variables used in this study are as follows. Table 2. Input Variables and Output Variables (Source: Excel processed data author, 2023) Input Variables Number of Authors Number of Publications Output Variables Impact Factor H5-Index 2) Data Collection Techniques The analysis method used in data collection focuses on SINTA 2 and 3 accredited journals in the fields of library and information science in 5 years, 2018-2022, but the data sources of the input and output variables are different. In the process of collecting the author's data, it will go through a process of collecting data in each volume and publishing each year then processed with Microsoft Excel formatting tools to obtain information on how many of the authors wrote in the learned journal. The variable data such as impact factor, H5-index, number of authors, and number of publications in each SINTA journal. Table 3. Input and Output Variables Period 2018-2022 (Source: Excel processed data author, 2023) Journal Impact Factor H5- Index Number of Publications Number of Authors Khizanah Al-Hikmah: Jurnal Ilmu Perpustakaan, Informasi, dan Kearsipan. 1,00 18 89 217 Jurnal Berkala Ilmu Perpustakaan dan Informasi 0,68 14 64 159 Baca: Jurnal Doumentasi dan Informasi 6,47 11 92 201 Jurnal Kajian Informasi dan Perpustakaan 2,68 19 70 170 Lentera Pustaka: Jurnal Kajian Ilmu Perpustakaan, Informasi, dan Kearsipan 0,96 9 64 132 Record and Library Journal 0,30 12 120 254 Edulib: Journal of Library and Information Science 4,49 9 89 215 Pustakaloka: Jurnal Kajian Informasi dan Perpustakaan 8,27 14 78 136 3) Data Analysis Techniques The data collected was analyzed using software, namely the DEAP application. The DEAP application is a program that works based on the concept of DEA constraints to calculate technical and cost efficiency. There are two DEAP program models, namely the CRS and CRS model (Fatimah & Mahmudah, 2017). The DEAP application software makes calculations easier. To determine the level of efficiency and decision-making ability for accreditation of SINTA 2 and 3 journals in the field of library science and informatics. Adjust with the DEA correction used, this is solved in equation 1Maximize: 𝐸𝑏 = Σ𝑟 𝑅 = 1 𝑢𝑟𝑏 𝑦𝑟𝑏 Σ𝑖 𝐼 = 1 𝑣𝑖𝑏𝑥𝑖𝑏 Constraint function: Σ𝑟 𝑅 = 1 𝑢𝑟𝑏 𝑦𝑟𝑗 Σ𝑖 𝐼 = 1 𝑣𝑖𝑏 𝑥𝑖𝑗 ≤ 1, ∀𝑗, 𝑗 = 1,2,3, ⋯ , 𝑁 And urb, vib ≥ for each r,I (where r = 1,2,3,...,R and I = 1,2,3,...I) Description: EB = efficiency on unit b yrj = quantity of output r produced unit j =1,2,..., N xij = quantity of input I produced unit j = 1,2,...,N urb = the weight given to the output R on the basis of unit B vib = weight given to input I on the basis of unit b e = very small positive number (Source: Siti Fatimah and Umi Mahmudah, 2017) The weights you and v are larger variables and constrains just like the numbers e (small positive) which aim to avoid arbitrary undefined inputs and outputs when determining efficiency. In trying to make the efficiency of each unit using linear program techniques. A DMU is declared efficient if its efficiency score is 1. Meanwhile, DMU is inefficient if the efficiency score is less than 1. In obtaining the solution of equation 1 it is necessary to convert it into the form of Linear Programming in equation 2. Maximize Σ 𝑅 𝑟=1 𝑢𝑟𝑏 𝑦𝑟𝑏 Constraint function Σ 𝐼 𝑖=1 𝑣𝑖𝑏 𝑥𝑖𝑏 = 1 Σ 𝑅 𝑟=1 𝑢𝑟𝑏 𝑦𝑟𝑗 – Σ 𝐼 𝑖=1 𝑣𝑖𝑏 𝑥𝑖𝑗 ≤ 0 Atau Σ 𝑅 𝑟=1 𝑢𝑟𝑏 𝑦𝑟𝑗 ≤ Σ 𝐼 𝑖=1 𝑣𝑖𝑏 𝑥𝑖𝑗 (Source: Siti Fatimah and Umi Mahmudah, 2017)

This DEA research is an approach used to obtain results or efficiency scores of library science, information science, and documentation study journals indexed by Sinta 2 and 3 in the period of 5 years, 2018–2022, using the DEAP 2.1 application as a tool in carrying out calculations. If the efficiency score of an efficient DMU will be given an efficiency score of 1 or TE (technical efficiency), Meanwhile, inefficient DMUs will be given an efficiency score of less than 1, so that all efficient DMUs have the same efficiency score. Table 4: Efficiency Calculation Results (Source: Excel processed data author, 2023) Journal Impact Factor H5- Index Number of Publicati ons Number of Authors Technical Efficiency Khizanah Al-Hikmah: Jurnal Ilmu Perpustakaan, Informasi, dan Kearsipan 1,00 18 89 217 0.949 Jurnal Berkala Ilmu Perpustakaan dan Informasi 0,68 14 64 159 1 Baca: Jurnal Dokumentasi dan Informasi 6,47 11 92 201 1 Jurnal Kajian Informasi dan Perpustakaan 2,68 19 70 170 1 Lentera Pustaka: Jurnal Ilmu Perpustakaan, Informasi, dan Kearsipan 0,96 9 64 132 0.61 Record and Library Journal 0,30 12 120 254 0.423 Edulib: Journal of Library 4,49 9 89 215 0.563 and Information Science Pustakaloka: Jurnal Kajian Informasi dan Perpustakaan 8,27 14 78 136 1 Based on the results of calculations carried out in the journals of library science, information science, and documentation studies indexed by Sinta 2 and 3 using the DEAP Version 2.1 application, it was found that 4 journals were declared efficient. The statement can be seen from the table above that in Sinta 2, there are 3 journals, namely the Journal Berkala Ilmu Perpustakaan dan Informasi, Jornal Baca: Jurnal Dokumentasi dan Informasi, and Jurnal Kajian Informasi dan Perpustakaan. And in Sinta 3, there is one journal that is declared efficient, namely the journal Pustakaloka: Jurnal Kajian Informasi dan Perpustakaan. The journal can be said to be efficient with the results of TE (technical efficiency), namely 1. While the other 4 journals were declared inefficient because they got TE results below 1. Based on research using the DEAP version 2.1 application, researchers calculated three calculations using the DEAP application. In the first calculation, there are 3 journals on each count through the DEAP application, namely Khizanah Al-Hikmah Jurnal Ilmu Perpustakaan, Informasi, dan Kearsipan. The average efficiency value of the three journals is 0.983, but there are only two efficient journals, namely the Jurnal Berkala Ilmu Perpustakaan dan Informasi with an efficiency value of 1.00 and Baca: Jurnal Dokumentasi dan Informasi with a value of 1.00, while the Khizanah Al-Hikmah Journal obtained a value of 0.949. Khizanah Al-Hikmah Journal is in the inefficient category but has a very high value. The difference is very little to reach a value of 1.00. (Petridis et al., 2013). In the second calculation, researchers counted three journals using the DEAP application, namely the Journal of Information and Library Studies, Lentera Pustaka: Jurnal Kajian Ilmu Perpustakaan, Informasi, dan Kearsipan. The average value of the efficiency of the three journals is 0.678, but there is only one journal that has an efficient value, namely the Journal of Library and Information Science Studies with an efficiency value of 1.00, while the Lentera Pustaka Journal has an efficient value of 0.610 and the Record and Library Journal gets a value of 0.423. In the last calculation using the DEAP version 2.1 application, researchers calculated the last 2 journals, namely the journals Edulib: Journal of Library and Information Science and Pustakaloka: Jurnal Kajian Informasi dan Perpustakaan The average efficiency value of the two journals is 0.782. Jurnal Edulib obtained an efficiency of 0.563, while Jurnal Pustakaloka obtained an efficiency value of 1.00.

Publishing a work in a reputable or accredited journal can enable it to get high citations. This is in accordance with the results of the study that the Sinta journal has a high efficiency value, so it has the opportunity to get high citations. Discussions related to reputable journals have also been presented in previous studies. The number of article citations is an important strategy to achieve nationally and internationally accredited journals. However, getting a high number of citations is also influenced by the accreditation of the journal that publishes the article. (Ridho Aulianto et al., 2019). One factor that indicates the success of a journal in providing usefulness of published works is the number of authors who cite from the journal. The better the accreditation of a journal, the greater the number of citations. (Ridho Aulianto et al., 2019) In research conducted for lecturers in publishing their work in reputable journals, the level of publishing scientific work is very productive until those who cite in their scientific work are high until the h-index reaches 14, and the highest level of citation from one of the authors is up to 854 citations. (Maya & Sarbini, 2023). The high number of citations indicates that the authors have made significant contributions to academia through publications in reputable journals, such as the journal BACA, which has a citation rate of 30 (Aulianto & Nashihuddin, 2020). The quality of an article's content is also affected by the number of citations it receives in the journal. A study revealed that articles with a high number of citations have better content than articles with a large number of pages but poorer content. (Nugroho, 2022)

This study aims to identify journals that are efficient and inefficient. This study uses the Data Envelopment Analysis (DEA) methodology to examine the relationship between input and output relationships in library science journals Sinta 2 and Sinta 3. This study has inputs in the form of the number of authors and the number of publications, and outputs in the form of impact factors and H5-Index. The goal of this study is to comprehend how input factors associated to library science journals, such the quantity of authors and publications, influence the results, like the impact factor and H-5 Index. By comparing each entity to those that are best at utilising their inputs to produce similar outputs, the DEA framework is used to quantify entity efficiency. Using the DEAP application, it is to determine that a journal with a calculation value of 1 is efficient. Sinta 2 consists of the journal Berkala Ilmu Perpustakaan dan Informasi, Baca: Jurnal Dokumentasi dan Informasi, Jurnal Kajian Informasi dan Perpustakaan. Sinta 3 consists of Pustakaloka: Jurnal Kajian Informasi dan Perpustakaan. For future research, the authors suggest that researchers submit journals to several journals that have an efficient value calculation of 1. Journals have a high impact factor, so it is possible to be cited in the work is also high.