Assessing the Academic Performance of Turkish Universities in 2023: A MEREC-WEDBA Hybrid Methodology Approach

ABSTRACT


Introduction
Education plays a pivotal role in fostering social development, with universities representing the apex of the formal education continuum.The pivotal role of university education in equipping students for professional life incentivizes them to seek enrollment in institutions of higher quality.Thus, prospective students often contemplate the question of which university offers superior educational opportunities.This dynamic prompts universities to strive for enhanced standings within the academic landscape.In response, universities engage in rigorous assessments of their current performance levels to inform strategic planning aimed at bolstering their academic achievements.This scholarly discourse has engendered numerous investigations focused on university rankings [1].
Scholarly literature features numerous investigations examining university rankings through various lenses.Jabjaimoh et al., [2] identifies five prominent studies on university rankings.Foremost among these is the "US News and World Report Best Global University Ranking (USNWR)" which regularly publishes rankings based on thirteen indicators [3].The second is the "Academic Ranking of World Universities (ARWU)" established by Shanghai Ranking Consultancy in 2009, also utilizing six indicators [4].Thirdly, the "University Ranking by Academic Performance (URAP)" issued by the Middle East Technical University Informatics Institute employs six indicators for its rankings [5].Fourthly, the "Quacquarelli Symonds World University Ranking (QS)" provides rankings across various disciplines, utilizing six indicators [6].Finally, the "Times Higher Education World University Ranking (THE)" considers five indicators in its ranking methodology [7].
While there exist similarities in university rankings organized by various institutions, distinctions emerge in measurement criteria (indicators), alongside differing perspectives on the significance levels of these criteria.Fauzi et al. [8] critically analyze university ranking reports, highlighting the advantages and drawbacks associated with each ranking criterion.Notably, Fauzi et al. [8] underscore that the determination of criterion importance levels often lacks empirical grounding, instead relying on fixed rates across reports.Within academic discourse, performance evaluation challenges are addressed through multi-criteria decision-making (MCDM) methodologies based on criterion weights.These methodologies are also observed in the context of university ranking determination [9][10][11][12][13][14].
This study aims to ascertain the ranking of Turkish universities in 2023 utilizing MCDM methods.To achieve this objective, the decision was made to utilize the university scores and criteria from 2023 as reported by THE, given their currency and relevance.The Method Based on The Removal Effects of Criteria (MEREC) method was chosen to determine criterion weights, while the Weighted Euclidean Distance Based Approach (WEDBA) method was selected for university ranking.The rationale for adopting these hybrid methods lies in their recent development within academic literature [15].In this context, three primary research questions are posited for this study: i. Research Question 1: To what extent can the ranking of Turkish universities be determined utilizing THE data?ii.Research Question 2: Is the hybrid MEREC and WEDBA method capable of generating university rankings effectively?iii.Research Question 3: Can actionable recommendations for universities to enhance their performance be derived by contrasting the newly formulated Turkish university rankings with those established by THE?

Literature Review
To assess the contributions of this research topic and the applied methodologies within the existing literature, a structured literature review is conducted in three stages.Firstly, a comprehensive examination of studies focusing on university rankings is undertaken to establish a contextual framework aligned with this research.Secondly, a critical review of scientific research concerning the MEREC method, a criterion weighting approach, is presented to elucidate the method's theoretical and empirical underpinnings.Thirdly, an in-depth analysis of the WEDBA method, utilized for alternative ranking, is conducted to ascertain its positioning within the literature concerning alternative ranking methodologies.
Diverse methodologies are evident in the examination of university rankings within academic research.Critiques have emerged against the methodologies employed in university ranking reports.Fauzi et al. [8] have provided a critical analysis of world university ranking reports, outlining both their merits and shortcomings.Noteworthy observations include the appreciation for subject and regional diversity in the QS reports juxtaposed with concerns regarding the 40% weight assigned to academic reputation.Similarly, while the inclusion of teaching metrics in THE reports is lauded, challenges related to normalizing subject disparities are noted.ARWU reports are commended for their emphasis on research aspects, although criticisms are directed at their reliance on Nobel prizes for ranking.The Leiden reports face criticism for limited university diversity, while Webometrics reports are scrutinized for their emphasis on marketability characteristics of universities, which is viewed as a drawback.
The literature review conducted within the domain of university ranking research yielded several significant findings.Aliyev et al. [12] investigated five United Kingdom universities using the fuzzy analytic hierarchy process (F-AHP) method to rank performance based on teaching, research, citations, and international outlook criteria.Teber and Karakaş [13] examined ten Turkish universities with Alternative Energy Resources Technology Programs, employing the AHP, simple multi-attribute rating technique (SMART), and technique for order preference by similarity to ideal solution (TOPSIS) methods across nine criteria.Salimi and Rezai [10] utilized the F-AHP method to rank universities based on criteria such as networking and knowledge exchange ability, general attractiveness, research ability, and commercialization ability.Ömürbek et al. [16] presented rankings of ten Turkish universities using AHP, TOPSIS, and VIKOR methods, with emphasis on criteria like international publications and project numbers.Ömürbek et al. [17] ranked fifty-three state universities in Türkiye using entropy and multi-attribute utility theory (MAUT) methods, with the number of databases being a pivotal criterion.Ömürbek and Karataş [9] applied the entropy method to rank entrepreneurial and innovative universities, focusing on criteria such as scientific and technological research competence, intellectual property rights, and entrepreneurship culture.Demir [14] utilized CRITIC and WEDBA methods for academic performance ranking in Turkish universities, emphasizing criteria like article score, citation score, and PhD score.Parlar and Palancı [18] evaluated country performances based on THE university ranking data using the Borda method and CRITIC and Entropy methods.Organ and Kaçaroğlu [19] ranked universities' performance using Entropy and TOPSIS methods across criteria like student numbers, URAP score, and library area.Ozdagoglu et al. [20] employed CoCoSo and MARCOS methods for performance ranking based on URAP data, while Birol and Ulutaş [11] used entropy and COPRAS methods for performance ranking based on financial criteria.These studies are summarized in Table 1.
A comprehensive review of studies adopting the MEREC method for criterion weighting reveals diverse applications across various domains.Ecer and Zolfani [21] conducted an economic freedom performance ranking of OPEC countries using the MEREC method for criterion weighting and the double normalization-based multiple aggregation (DNMA) method for ranking.Ecer and Aycin [22] employed the MEREC method to rank G-7 countries based on innovation performance using multiple MCDM methods.Ulutaş et al. [23] utilized the MEREC method in the pallet truck selection problem, employing the simple weighted sum product (WISP-S) method for ordering alternatives.Simic et al. [24] leveraged MEREC and MARCOS methods for sustainable policy selection to mitigate urban transport's impact on climate change.Haq et al. [25] applied the MEREC and MARCOS methods using single valued neutrosophic numbers for sustainable material selection.Entropy, MAUT and SAW "Scientific and technological research competence, intellectual property rights, cooperation and interaction, culture of entrepreneurship and innovation, economic contribution and commercialization (5 criteria)" Birol and Ulutaş [11] Entropy and CORPAS "New year's allowance amount, year-end allowance amount and spending amount (3 criteria)" Aliyev et al. [12] F-AHP "Teaching, research, citations, and international outlook (4 criteria)" Teber and Karakaş [16] AHP, SMART and TOPSIS "Campus facilities and social life opportunities, a ranking of the university, the technological background and laboratories, the ratio of preference, the number and the title of faculties, the number of ERASMUS students, foreign language education, the number of program vacancy, the percentage of the province's power plants (9 criteria)" Parlar and Palancı [18] BORDA, CRITIC and Entropy "Research, citation, industry revenues and international outlook (4 criteria)"

Organ and
Kaçaroğlu [19] Entropy and TOPSIS "Total number of students, university ranking by academic performance score, number of lecturers, number of printed books, library area, area per student, project support amount and full scholarship score (8 criteria)" Ozdagoglu et al. [20] CoCoSo and MARCOS "Number of articles, number of citations, number of scientific documents, number of Phd and number of academicians per student (5 criteria)" Demir [16] CRITIC and WEDBA "Article score, citation score, scientific document score, Phd.score and academician score per student (5 criteria)" Rani et al. [26] utilized MEREC and additive ratio assessment (ARAS) methods for food waste treatment technology selection.Shanmugasundar et al. [27] employed the MEREC method for optimal Spray-Painting Robot selection.Mishra et al. [28] utilized MEREC for criterion weighting in low-carbon tourism strategies selection, ranking strategies with the MULTIMOORA method.Ivanović et al. [29] applied the MEREC method for criterion weighting in truck mixer concrete pump selection, ranking alternatives with the DN-MARCOS method.Additionally, Toslak et al. [30] employed MEREC and WEDBA methods to evaluate a logistics company's performance, marking the hybrid application of these methods for the first time.A comprehensive literature review detailing the applications of the MEREC method is presented in Table 2.
A thorough examination of studies utilizing the WEDBA method for alternative ordering reveals diverse applications across various domains.Jain and Ajmera [31] utilized the WEDBA method to rank the flexibility performance of flexible manufacturing systems.Ulutaş [32] employed the WEDBA method for the selection of tackers following criterion weighting with the preference selection index (PSI) method.Al-Hawari et al. [33] introduced the F-WEDBA method, integrating fuzzy numbers into the WEDBA framework.Basar and Tolga [34] utilized the WEDBA method to evaluate smart systems.Işık [35] employed the WEDBA method for financial performance analysis following criterion weighting with the Analytic Hierarchy Process (AHP) and Criteria Importance Through Intercriteria Correlation (CRITIC) methods.Durak and Tolga [36] applied the WEDBA method in the selection of Robotic Process Automation solutions.Şimşek [37] utilized the WEDBA method to assess the financial performance of banks.Garg [38] calculated criterion weights using the AHP method for e-learning website selection, subsequently applying the WEDBA method for ranking.Ecemiş and Coşkun [39] used the PSI method for criterion weighting and the WEDBA method for ranking regions in evaluating information and communication technologies.A comprehensive literature review detailing the applications of the WEDBA method is provided in Table 3.   [35] AHP, CRITIC WEDBA "Financial performance analysis ranking" Şimşek [37] AHP WEDBA "Financial performance ranking of banks" Ecemiş and Coşkun [39] PSI WEDBA "Ranking of the use of information and communication technologies" This study is structured into five sections.Section-2 provides a detailed exposition of the MEREC and WEDBA methods, delineating their procedural steps.In Section-3, an empirical application is conducted to ascertain university rankings.Subsequently, Section-4 presents the results, grounded in the empirical findings.Section-5 offers actionable recommendations for universities and researchers, alongside a delineation of the study's limitations.

Methodology 2.1 Criteria Selection
To ascertain the rankings of Turkish universities in 2023, the study adopts five criteria outlined in THE (Times Higher Education) reports as the research criteria.These criteria encompass "Teaching (C1), Research (C2), Citations (C3), Industry Income (C4), and International Outlook (C5)." i.The Teaching criterion encompasses aspects related to the learning environment within universities.Scores derived from metrics such as reputation surveys, staff to student ratios, doctorate to bachelor's ratios, doctorates awarded to academic staff ratios, and institutional income are utilized to determine university scores for this criterion.
ii.The Research criterion focuses on evaluating universities' research capacities.Scores for this criterion are determined based on factors such as reputation surveys, research income, and research productivity.iii.The Citations criterion evaluates universities based on their citation levels.Scores for this criterion are derived from the citation averages of universities on a global scale.iv.
The Industry Income criterion assesses the income generated by universities through their industrial activities.v.
The International Outlook criterion gauges universities' international presence, including staff and student diversity.Scores for this criterion are determined using metrics such as international to domestic student ratios, international to domestic staff ratios, and international collaboration.Further details regarding the research criteria and information about the research dataset are provided in Table 4 for reference.

Method based on the Removal Effects of Criteria (MEREC) Method
The MEREC method, designed for criterion weighting, calculates based on the alterations in performance values among alternatives.This method was originally formulated by Keshavarz-Ghorabaee et al. [40] and has gained substantial traction in recent academic discourse.It has been implemented across various studies in literature, showcasing its widespread adoption.The method comprises six distinct steps, as delineated by Keshavarz-Ghorabaee et al. [40] and further expounded upon by Ulutaş et al. [23].These steps are systematically elucidated in the following section: Step 1-1: The decision matrix, denoted by  alternatives () and  criteria (), is represented in Eq. (1).
Step 1-2: The decision matrix undergoes normalization as per Eq. ( 2).Step 1-3: The comprehensive performance values of the alternatives (  ) are computed according to Eq. (3). )) Step 1-4: The alterations in the performance values of the alternatives (  ′ ) are determined using Eq. ( 4), achieved through the subtraction of each criterion.

Weighted Euclidean Distance Based Approach (WEDBA) Method
The WEDBA method, a prominent MCDM technique, is employed for ranking the alternatives.This method consists of six sequential steps, each meticulously detailed as per the works of Rao and Singh [41] and Jain and Ajmera [31]: Step 2-1: The decision matrix is formulated according to Step 1-1 of the MEREC method.
Step 2-2: The decision matrix is standardized using Eq. ( 7) in the normalization process.
Step 2-4: The ideal and anti-ideal values are determined using Eq. ( 10) and Eq. ( 11) in the calculation process.
Step 2-5: The weighted Euclidean distances among the alternatives are computed using Eq. ( 12) and Eq. ( 13) as part of the calculation process.
Step 2-6: The index scores for each alternative are determined using Eq. ( 14).The alternative exhibiting the highest index score is considered the optimal choice or best alternative within the context of the analysis.

Application
In this study focusing on the performance ranking of Turkish universities in 2023, five criteria were employed: "Teaching (C1), Research (C2), Citations (C3), Industry Income (C4), and International Outlook (C5)," as outlined in the methodology section.The scores for these criteria were sourced from THE (Times Higher Education) reports, specifically THE 2023 report that encompassed sixty-one Turkish universities.Consequently, all sixty-one universities (A1, A2, ..., A61) featured in THE 2023 report were included in this study and are listed in the criteria selection section within the methodology.
The research application unfolded in two distinct stages.Initially, criterion weights were determined using the MEREC method in the first stage.Subsequently, in the second stage, the universities were ranked employing the WEDBA method.The chronological sequence and detailed steps of this application are systematically elucidated throughout the methodology section.
Step 1-1: The decision matrix, comprising sixty-one alternatives and five criteria, is presented in Table 5.These data were sourced from THE 2023 report, serving as the foundation for the analysis in this study.Step 1-2: The decision matrix is standardized using Eq. ( 2) and is displayed in Table 6 as part of the normalization process.Step 1-3: The   values are computed according to Eq. ( 3) and are presented in Table 7 as part of the calculation process.Step 1-4: The   ′ values are determined using Eq. ( 4) and are displayed in Table 8 as part of the calculation procedure.Step 1-5: The   values are computed using Eq. ( 5) and are presented in Table 9 as part of the calculation process.Step 1-6: The criterion weights   were determined using Eq. ( 6) and are displayed in Table 10 as part of the calculation process.Step 2-1: In the application of the WEDBA method, the decision matrix presented in Table 5 is utilized.
Step 2-2: The decision matrix undergoes normalization using Eq. ( 7) and is depicted in Table 11 as part of the normalization process.Step 2-3: The normalized decision matrix is standardized using Eq. ( 8) and is presented in Table 12 as part of the standardization process.Step 2-4: The ideal and anti-ideal values are computed using Eq. ( 10) and Eq. ( 11), respectively.These values are displayed in Table 13 as part of the calculation process.Step 2-5: The weighted Euclidean distances among the alternatives are calculated using Eq. ( 12) and Eq. ( 13) as part of the calculation process.These distances are presented in Table 14 for reference.Step 2-6: The index scores for each alternative are computed using Eq. ( 14).The resulting university rankings are displayed in Table 15 as part of the analysis.

Table 15
The rankings of the universities. Alt.

IS i Rankings
Alt.

IS i Rankings
Alt.

Results
University rankings serve as a valuable tool for students, offering insights into a university's standing relative to others and the performance levels across various criteria.Extensive literature exists focusing on university rankings, with evaluations conducted using diverse criteria.In this study, the discussion centers on sixty-one Turkish universities included in THE university ranking.The primary objective is to recalibrate the criteria weights based on THE ranking data and subsequently organize the 2023 Turkish university rankings using these revised weights.Notably, the MEREC method was employed for criterion weighting, while the WEDBA method facilitated the ranking of universities.This study adopts a unique approach in obtaining criteria weights, diverging from the established THE criteria weights.
As per THE reports, the weight attributed to the Teaching (C1) criterion stands at 30%.However, in this research, the weight assigned to the Teaching criterion was determined to be 13.57%.Similarly, while the Research (C2) criterion carries a weight of 30% in THE reports, it was found to be 14.64% in this study.Conversely, the weight of the Citations (C3) criterion in THE reports is 30%, but it was calculated as 50.77% in this research.The Industry Income (C4) criterion, originally set at 2.5% in THE reports, was determined to be 6.77% in this study.Moreover, the International Outlook (C5) criterion, initially at 7.5% according to THE reports, was calculated as 14.25% in this research.
Based on these findings and the dataset pertaining to Turkish universities, there is a notable decrease in the weights assigned to the Teaching and Research criteria.Conversely, there is an increase in the weights allocated to the Citations, Industry Income, and International Outlook criteria.These adjustments reflect the nuanced performance and emphasis of Turkish universities across these criteria, as indicated by the study's methodology.
The 2023 performance ranking of Turkish universities was established through the WEDBA.Table 16 provides an overview of the rankings derived from WEDBA findings alongside THE rankings.The outcomes derived from this analysis are presented as follows: i.According to the 2023 THE report, the top three universities in Türkiye are Çankaya University, Koç University, and Sabancı University.However, in the context of this research, the three universities ranking highest are Çankaya University, Fırat University, and Bahçeşehir University.ii.When comparing the rankings between THE and the research findings, several universities experienced a decline in their rankings.

Table 1
Literature review for university ranking.

Table 2
Literature review for MEREC method.

Table 3
Literature review for WEDBA method.

Table 4
Selected criteria and universities.

Table 5
The decision matrix.

Table 7
The   values.

Table 8
The   ′ values.

Table 9
The   values.

Table 12
The standardized decision matrix.

Table 13
Ideal and anti-ideal values.

Table 14
The weighted Euclidean distances.

Table 16
Comparison of Turkish Universities ranking between MEREC-WEDBA and THE.