Should I Stay or Should I Go? Mapping the Key Drivers of Skilled Migration Using Fuzzy Multi-Criteria Decision Methodology

Abstract

The emigration of highly skilled individuals has become a critical concern for many countries amid increasing global labor mobility. This study employs the Improved Fuzzy Step-Wise Weight Assessment Ratio Analysis (IF-SWARA) method within a fuzzy multi-criteria decision-making (FMCDM) framework to identify and prioritize the key drivers of skilled migration. Drawing on opinions from sixteen Turkish emigrants currently residing abroad, the study captures firsthand perspectives on the structural factors influencing their migration decisions. The results indicate that the most influential factors are workplace conditions, living standards, and academic standards. These findings underscore the multifaceted nature of brain drain and highlight the necessity for comprehensive policy approaches that address both push and pull dynamics. By systematically ranking these determinants, the study contributes to the growing body of evidence-based research on international human capital flows.

1. Introduction

Migration, a fundamental aspect of human history, predates even the earliest forms of commerce and capital flows [1]. This term is defined as an individual’s permanent shift in place of residence. Migration occurs primarily due to economic or political differences across countries, particularly among highly trained individuals [2]. According to Lee, push and pull factors affect migration. Also, globalization has led to an increased demand for highly skilled individuals worldwide, necessitating the evaluation of talented and successful individuals under similar standards from an international perspective [3]. Brain drain, often referred to as human capital flight [4] (p. 2), was a term first coined in a report by the Royal Society of London in 1963 [5]. It describes the migration of highly educated and skilled individuals, such as scientists and technologists, from their home countries in pursuit of better opportunities abroad [6]. This form of migration is predominantly observed among individuals from developing nations relocating to developed countries [7]. Brain drain poses significant challenges to sending countries by depleting their human capital, thereby hindering both development and socioeconomic progress. In the post-COVID-19 era, a new dimension of this phenomenon, termed “virtual brain drain,” has emerged, referring to individuals working for foreign companies while remaining physically in their home country [8].

A widely used framework to explain the dynamics of brain drain is the push–pull model, which distinguishes between structural factors that drive individuals away from their home countries and those that attract them to specific destinations. While some scholars critique the push–pull framework for its simplicity and determinism [9,10], it remains valuable due to its intuitive and empirically grounded perspective that emphasizes the role of structural forces in shaping migration processes [11]. Push factors include unfavorable domestic conditions such as unemployment, wage stagnation, or political instability, while pull factors refer to opportunities abroad, including higher standards of living, career-related opportunities, and language skills [12]. These dynamics, amplified by persistent socio-economic challenges at home and the growing accessibility of global labor markets driven by globalization, have encouraged skilled workers to relocate to destinations offering better employment opportunities and career advancement prospects [13]. The movement of highly skilled individuals seeking better opportunities abroad for work, education, or retirement raises serious concerns regarding both the root causes and the far-reaching consequences for sending and receiving countries.

These drivers intensify skilled migration flows, particularly from developing to developed countries, by creating persistent structural imbalances in human capital development [14] and undermining the capacity of home countries to retain talent. As such, brain drain is not merely a consequence of individual decision-making, but a reflection of broader systemic deficiencies that shape the global distribution of skills and knowledge. Research indicates that the departure of skilled individuals disproportionately harms the development of sending countries, often exacerbating existing challenges [15]. Considering the economic implications, the outflow of human capital diminishes productivity and hampers innovation, ultimately weakening a nation’s ability to pursue sustainable development [16]. In domains where human capital is irreplaceable, such as healthcare, the emigration of skilled professionals often leads to acute workforce shortages, severely disrupting the delivery of basic services and intensifying public health challenges as demonstrated in various sub-Saharan African cases [17].

Comparable patterns of skilled migration have also been documented across various developing and emerging economies, each reflecting distinct national contexts while converging around common structural challenges. For instance, Malaysia has witnessed substantial outflows of tertiary-educated individuals—particularly engineers and STEM graduates—primarily driven by wage disparities and limited domestic career advancement opportunities [18]. Most low-income countries experience a significant shortage of physicians, and the emigration of this limited medical workforce intensifies difficulties in delivering sufficient healthcare services [19]. South Africa provides another illustrative case, where political transition and economic instability have contributed to negative net migration among skilled workers, with official statistics significantly underestimating actual emigration by 57.3% in 2003 [20].

This phenomenon is not limited to developing countries; it has also emerged in certain high-income economies grappling with prolonged economic uncertainty. Despite being classified as a developed country, Greece has experienced pronounced levels of skilled emigration, particularly following its sovereign debt crisis. Between 2008 and 2017, approximately 467,765 individuals (roughly 4.6% of the national population) emigrated, with an estimated 70% holding tertiary-level degrees [21]. These diverse examples underscore the systemic and cross-regional character of brain drain, reinforcing the need for multidimensional analytical frameworks and context-specific policy interventions aimed at mitigating its long-term socio-economic consequences.

These diverse examples underscore the systemic and cross-regional character of brain drain, reinforcing the need for multidimensional analytical frameworks and tailored policy responses. A particularly notable case is Turkey, where emigration of highly educated individuals has accelerated considerably over the past few years. Especially for Turkish citizens, the most popular destinations include the United States, Germany, and the United Kingdom, reflecting favorable labor market conditions in host countries [8]. Between 2015 and 2020, the Turkish migrant population residing in EU-28 countries surpassed 6 million, with a five-year growth rate of 6.06%, indicating a steep rise compared to previous periods [22]. Moreover, the Turkish Statistical Institute (TURKSTAT) reported an unprecedented outflow of nearly half a million individuals in 2022 alone, underscoring growing concerns about the sustainability of human capital retention in the face of persistent political, economic, and professional dissatisfaction. These trends reflect how both structural conditions and perceived opportunity differentials continue to fuel brain drain even in countries with significant investment in higher education and professional training.

Concerns about the number of and trends in highly skilled migration and a lack of statistical data on the number of individuals have made empirical research an invaluable resource in this area [6,23]. According to Massey et al., a multi-level approach and interdisciplinary tools are necessary to comprehend the talent diaspora thoroughly [24]. There have been limited studies on brain drain in Turkey [25,26,27], and some lack empirical evidence [28]. Additionally, in the literature on brain drain, Multi-Criteria Decision-Making (MCDM) techniques are uncommon [29].

FMCDM methods are valuable tools for tackling decision-making problems characterized by uncertain or vague criteria. These methods enhance traditional MCDM approaches to accommodate fuzzy information, enabling decision-makers to incorporate imprecise data into their analyses. By merging fuzzy set theory with MCDM techniques, FMCDM provides a systematic approach to managing uncertainty and differing perspectives in the decision-making processes [30]. These methods offer a structured framework for evaluating alternatives based on various criteria and sub-criteria when the data is imprecise or uncertain. They align with the opinions of experts and facilitate a comprehensive assessment of the decision space [31]. Through the integration of fuzzy logic and MCDM principles, these methods deliver a systematic strategy for decision-making across diverse domains, ensuring thorough assessments and informed choices. In light of these considerations and the existing complex dynamics of migration decisions, the IF-SWARA method has been chosen to evaluate the influencing criteria related to the brain drain motivations of Turkish citizens.

Using a hybrid approach that incorporates FMCDM, our study aims to illuminate the complex nature of the Turkish brain drain. It provides a detailed and nuanced analysis of the factors driving skilled workers or students from Turkey to pursue opportunities abroad. Although FMCDM approaches have rarely been applied to brain drain studies, this study introduces IF-SWARA to explore migration motivations in the Turkish context. This article discusses the causes of this migration trend, examining the numerous factors contributing to brain drain, ranging from political stability and improved academic opportunities to economic uncertainty. The narrative highlights not only the loss of talent but also reflects countries’ political and socioeconomic landscape, signaling a need for reflection and potential change. By exploring these drivers, this article seeks to highlight the implications for counties’ future and the possible strategies for retaining their most valuable asset—their people.

Overall, this paper focuses on the following research questions:

RQ1: What are the factors affecting Turkish brain drain phenomena?

RQ2: What are the most critical factors behind Turkish citizens’ brain drain motivations?

This study employs an FMCDM method, IF-SWARA, to address the research questions at hand. The IF-SWARA method provides a more reliable evaluation of the criteria within a subjective context. The methodology presented here will aid in the complex decision-making process such as assessing the motivations behind brain drain. Furthermore, this method will be introduced and applied for the first time to evaluate the factors influencing Turkish citizens’ motivations for brain drain, despite the limited sample size (n = 16), its novelty underscores the analytical and methodological contribution of this study.

The following section presents our analytical framework for examining the drivers of migration and introduces the notion of driver complexes, which describe the interactions among these factors in shaping migration patterns. This analysis is grounded in the MCDM approach, which has been extensively applied in migration research. Section 3 outlines the methodology and dataset, while Section 4 presents the results. Finally, the findings are discussed in depth to draw meaningful conclusions.

2. Push and Pull Factors

Globalization has considerably intensified the brain drain phenomenon, as international migration has become increasingly accessible and widespread in modern society. According to Ryazantsev et al., globalization has triggered structural transformations that facilitate the movement of highly skilled individuals seeking to optimize their economic prospects across national borders [32]. The growing interdependence of global economies encourages skilled professionals to gravitate toward wealthier nations, thereby perpetuating a self-reinforcing cycle of migration that undermines the economic and social development of their countries of origin.

The prevailing literature acknowledges that the decision to migrate results from the interplay of various identifiable factors; these factors are often categorized as “push factors” that drive migrants to leave their countries of origin and “pull factors” that attract them to destination countries [13]. Push factors impact migration, including conditions and reasons that drive people to leave their homelands. Population increase and the resulting unemployment and poverty, as well as political instability, conflicts, political events [33], and environmental drivers [34]—such as climate change, drought, land degradation, and sea-level rise—are major driving forces in developing countries [35]. These conditions can be categorized into economic, social, political, legal, and personal factors such as self-realization [36,37]. When a country fails to fulfill these aspects, leading to dissatisfaction among its citizens, the likelihood of experiencing a brain drain escalates.

On the other hand, pull factors refer to foreign countries’ attractions, such as higher salaries, better working conditions, opportunities for career advancement, and access to advanced technology, and better health conditions [38]. Furthermore, access to higher education, healthcare, and overall quality of life is an attractive pull factor for migrants [39]. Similarly to push factors, pull factors can be classified into several categories. Beyond the economic benefits and professional opportunities, social and cultural inclusivity greatly influences a nation’s ability to attract skilled workers. The promise of living openly and safely as oneself, free of the threat of violence and prejudice in countries with restrictions towards LGBTQ people, contributes to the appeal of specific destinations [40]. The dynamics of migration flows, including their volume and direction, are substantially influenced by the interplay between push and pull factors [41]. Before listing the criteria evaluated in this study, it is important to note that the initial categorization and selection were largely inspired by the framework proposed by Incekas and Kadaifci. Their classification of career, educational, political, and personal factors served as the foundational reference. However, the present study adapts and extends this structure through expert validation and cross-referencing with other empirical works (e.g., Castelli,; Karaduman & Çoban,), ensuring that the set of criteria reflects both literature-grounded and context-specific considerations.

2.1. Career-Related Factors

Career-related motivations significantly influence the migration decision of highly skilled individuals. From a push perspective, restricted employment opportunities (C1), inadequate workplace conditions (C2), and unfair salary policies (C3) lead to professionals’ dissatisfaction in their home countries [29,42]. In numerous developing contexts, the absence of defined career pathways and merit-based promotions diminishes motivation and obstructs long-term engagement [43]. In contrast, pull factors including higher wages, improved working conditions, side benefits (C4) [29,36,44], and access to advanced research and development facilities (C5) attract skilled workers to high-income economies [45]. Empirical evidence indicates that wage differences between source and destination countries significantly influence emigration decisions, in particular among STEM professionals [13,46]

2.2. Education-Related Factors

Educational aspirations play a crucial role in the trend of brain drain, especially among younger populations. Factors driving the situation encompass diminishing academic standards (E1), inadequate access to global academic networks (E2), i.e., cross-border education, and restricted opportunities for scholarships or grants (E3). In numerous developing countries, inadequate funding for higher education weakens the quality and accessibility of academic programs [47]. However, the allure of educational excellence internationally, reflected in advanced research ecosystems, global recognition and language skill advancement (E4), also motivates individuals to pursue academic mobility as the reputational advantages of holding a degree from prestigious institutions often enhance employment security and income potential in both host and home countries [48]. In this regard, language-related skills are equally crucial for successful adaptation, since the ability to manage emotional experiences in a foreign language is a significant accomplishment [49,50]. Technology-enabled platforms and access to worldwide knowledge production (i.e., technology services for education (E5)) enhance the attractiveness of host countries [43].

2.3. Governance and Political Climate

Governmental structures and political climates serve as strong push factors in emigration. Shifts in a nation’s institutional quality, particularly regarding political stability and economic climate of the country (G1), that influence market regulation and policy formulation, establish the framework within which skilled individuals assess and decide on emigration [51]. Mistrust in the political system and erosion of civil liberties, i.e., limitation/affirmation of fundamental rights (G3), are primary drivers of brain drain [52]. A country’s political dynamics (G2)—shaped by crises, conflict, coups, and violence, or even by the perception that its government may be overthrown by unconstitutional means [13]—erodes professionals’ trust in public institutions and infringes on basic rights, prompting skilled workers to seek opportunities abroad [7,29,42,44,53,54,55]. Conversely, pull factors include effective migrant integration programs (G4), inclusive social policies (G5), and strong democratic governance. Countries characterized by political stability and disaster resilience (G6) attract migrants for economic incentives as well as for personal safety and long-term security [40]. Van Hear et al., 2018 [11] stated that persistent rains, beginning in March 2010, caused significant flooding in southern and central Somalia by June, resulting in extensive damage to crops, animals, and property. At least 6000 more households were thus uprooted, many of them moving to IDP camps.

2.4. Personal Motives

Personal motives and sociocultural inclusivity significantly influence migration decisions in subtle yet impactful ways. Push dynamics encompass dissatisfaction with overall living conditions (P3), limitations on identity expression (P6), and insufficient network-building chances (P5). In societies with high rates of discrimination or limited lifestyle freedoms, emigration serves as a means for self-realization [54]. While such push factors compel individuals to seek alternatives abroad, it is the promise of significantly better living standards and societal well-being in host countries that completes the migratory equation. Particularly access to high-quality healthcare, education, and social services—serve as major pull factors attracting professionals to migrate to developed countries [44,48]. Developed countries generally offer higher living standards, reflected in lower crime rates, advanced public infrastructure, and more dependable services, thereby contributing to an improved overall quality of life. In addition, the political and economic stability of these countries offers a secure environment for raising families and achieving long-term career objectives [56]. Other pull factors such as cultural diversity (P4), religious tolerance (P1), and desire to reside overseas (P2) [29], position developed countries as attractive destinations [44]. These factors contribute to personal satisfaction, career sustainability, and social belonging.

2.5. The Migration and MCDM Nexus

Çetinkaya et al. proposed a scientific method for campsite selection near the Syrian border: evaluating sites based on various criteria using Geographic Information Systems (GIS) via ESRI ArcGIS 10.2 software, prioritizing indicators with a Fuzzy Analytic Hierarchy Process (FAHP), and ranking sites with the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method [57]. This approach suggests 15 new, more suitable locations for refugee camps. Denekos et al. introduced a methodology combining GIS and MCDM techniques to support siting refugee camps in Greece [58]. Another GIS-based study was conducted by Rashid to formulate a spatial internal migration model [59]. In the investigation of the internal migration potential of the regions in Serbia, Arandarenko et al. offered a multiple criteria decision-aiding approach by using Elimination and Choice Translating Reality (ELECTRE) Tri-C, imprecise SRF, and stochastic multicriteria acceptability analysis methods [60].

A study by Blouchoutzi et al. proposed a model using the Preference Ranking Organization METHod for Enrichment Evaluations (PROMETHEE), a multiple criteria decision analysis method, which is informed by the Migrant Integration Policy Index for integrating the foreign-born population in the European Union (EU) [61]. Another study by Martín & Indelicato introduces a method based on data envelopment analysis to measure citizens’ openness to immigration and refugees in the EU countries. Results show that countries such as the Nordic countries, Spain, and Portugal are found to be more open to immigration and refugees compared to other countries in the sample [62].

Khaw et al. proposed an integrated structural equation modeling and multiple-criteria decision-making model to address the intentions of immigrant entrepreneurs from Malaysia, Pakistan, Nigeria, and Singapore [63]. Chang & Lai analyzed the entrepreneurial process based on the opinions of 43 experts and 21 indicators through the fuzzy analytic hierarchy process, focusing on Taiwanese who operate in China [64]. Their findings revealed that marketing, business development, and administration play essential roles in influencing entrepreneurs’ decisions.

Guo & Wu used a hierarchical model and the AHP to identify the factors driving brain drain among 160 IT professionals [65]. Findings suggest that factors related to the company, such as industry environment, career development prospects, corporate culture, nature of work, and total rewards, are the most critical in influencing IT professionals’ migration. In contrast, personal factors like family status, age, educational background, seniority, lifestyle, and working conditions are deemed the least important. Incekas & Kadaifci presented a study employing the Analytic Network Process (ANP) in an MCDM model to explore what drives highly skilled individuals to emigrate, pinpointing economic, career, and political factors as primary reasons. The study’s insights, derived from twenty-six immigrants, emphasize the profound effects of brain drain on source countries’ economic, political, and socio-cultural aspects, proposing a model to help policymakers craft effective countermeasures.

Oreški et al. analyzed brain drain factors among 1323 students at the University of Zagreb, employing neural network analysis and the AHP to discern and prioritize push and pull factors influencing emigration decisions [66]. The study concluded that pull factors are more influential than push factors for most fields, indicating the need for developing countries to improve employment opportunities, quality of life, and business climates to prevent talent loss. Mukhtarov et al. developed a model to mitigate brain drain in emerging economies by integrating a balanced scorecard and multi-objective optimization based on ratio analysis using the step-wise weight assessment ratio analysis (BOFQ M-SWARA) method [67]. The evaluation was based on the opinions of three experts. They highlighted technical competency as key to retaining talent, advocating for R&D tax exemptions to encourage technological innovation and sustainable economic growth.

Considering the critical importance of understanding citizens’ motivations for migration and the phenomenon of brain drain, coupled with the limited number of studies, particularly those related to Turkish citizens, this research aims to address this notable gap in the existing literature. To achieve this, an FMCDM methodology has been proposed to assess the primary factors influencing brain drain motivations among citizens, with support from a case study involving Turkish nationals.

3. Methodology

This study introduces an FMCDM method to analyze the underlying causes of the migration trend among Turkish citizens. To achieve this, the IF-SWARA method, a fuzzy logic extension of the SWARA method proposed by Vrtagić et al. is employed to weight criteria through pairwise comparisons [68]. The primary consideration in selecting the methodology for this study is the extent to which fuzzy MCDM methods, which are capable of resolving intricate problems, can be adapted to address a complex issue involving numerous factors such as migration motivation.

The comprehensive framework of the utilized model is illustrated in Figure 1.

3.1. Improved Fuzzy Step-Wise Weight Assessment Ratio Analysis (IF-SWARA)

The SWARA method, developed by Keršulienė et al. is an MCDM technique used to calculate criteria weights based on experts’ opinions [69]. In real-world applications, exact numbers cannot represent most criteria values. As stated by Zadeh, fuzzy set theory is widely applied to manage imprecise, vague, and uncertain data in real-life scenarios [70]. In fuzzy set theory, linguistic terms are employed and converted into quantitative data to facilitate better evaluations.

Since conventional MCDM methods struggle to effectively address issues based on imprecise data and information, fuzzy MCDM methods have been developed to tackle the uncertainties in assessing the relative importance of criteria and the performance scores of alternatives regarding those criteria. From this standpoint, Mavi et al. introduced the Fuzzy SWARA method, which is grounded in fuzzy sets and seeks to mitigate many ambiguities inherent in the evaluation process [71].

Fuzzy SWARA is a hybrid model that integrates fuzzy logic and the SWARA technique to evaluate the relative importance of criteria and sub-criteria within a fuzzy environment. Vrtagić et al. indicates that the linguistic scale employed in Fuzzy SWARA for pairwise comparisons between criteria is inadequate and recommended introducing a new scale by enhancing the technique to address this shortcoming. In this context, the IF-SWARA was proposed as a novel method. The steps of the IF-SWARA method are defined below [66]:

Step 1. Sort the criteria according to their importance, considering the purpose of the problem.

Step 2. Determine the comparative importance of criteria. Experts allocate an importance score to criterion j, in relation to the previous criterion, based on the fuzzy linguistic scale as shown in

Table 1. Fuzzy Linguistic Scale for the Improved Fuzzy SWARA.

Linguistic Scale	Fuzzy Numbers
Absolutely less significant (ALS)	(1, 1, 1)
Dominantly less significant (DLS)	(0.50, 0.67, 1)
Much less significant (MLS)	(0.40, 0.50, 0.67)
Really less significant (RLS)	(0.33, 0.40, 0.50)
Less significant (LS)	(0.29, 0.33, 0.40)
Moderately less significant (MDLS)	(0.25, 0.29, 0.33)
Weakly less significant (WLS)	(0.22, 0.25, 0.29)
Equally significant (ES)	(0, 0, 0)

Source: [66].

.

Step 3. Calculate the coefficient values (${\tilde{k}}_j$) as follows Equation (1)

$${\tilde{k}}_j=\left\{\begin{array}{cc}1& j=1\\ {\tilde{s}}_j+1& j>1\end{array}\right.$$

(1)

Step 4. Calculate the fuzzy weight coefficients (${\tilde{k}}_j$) as follows Equation (2)

$${\tilde{q}}_j=\left\{\begin{array}{cc}1& j=1\\ {\displaystyle \frac{{\tilde{q}}_{j-1}}{{\tilde{k}}_j}}& j>1\end{array}\right.$$

(2)

Step 5. Calculate the relative weights of the evaluation criteria (${\tilde{w}}_j$) as follows Equation (3)

$${\tilde{w}}_j={\displaystyle \frac{{\tilde{q}}_j}{{\sum }_{j=1}^n{\tilde{q}}_j}}$$

(3)

Step 6. In the last step, the calculated fuzzy weights should be converted into crisp values using Equation (4). Here, $\tilde{A}$ is a triangular fuzzy number where “l”, “m”, and “u” represent the lower value, middle value, and upper value, respectively.

$$d\left(\tilde{A}\right)={\displaystyle \frac{\left(l+4m+u\right)}{6}}$$

(4)

Basic mathematical operations on triangular fuzzy numbers $A_1$ ($l_1,m_1,u_1$) and $A_2$ ($l_2,m_2,u_2$) that also used in IF SWARA are shown Equations (5)–(8).

$$A_1+A_2=(l_1+l_2,m_1{+m}_2,+u_1+u_2)$$

(5)

$$A_1-A_2=(l_1-u_2,m_1{-m}_2,u_1-l_2)$$

(6)

$$A_1\times A_2=(l_1·l_2,m_1{·m}_2,u_1·u_2)$$

(7)

$$A_1÷A_2=(l_1/u_2,m_1{/m}_2,u_1/l_2)$$

(8)

3.2. Problem Statement

Following the establishment of the criteria, sixteen highly skilled Turkish immigrants (who completed their education in their country) participated in the study between June and December 2024 and assessed their motivations for brain drain based on the aforementioned criteria (detailed in Appendix A). Although the number of participants is relatively low and respondents are not formal experts, they have direct experience with high-skilled emigration from Turkey. Their perspectives offer valuable insights into the motivational decision-making framework. The main aim of this study is not to generate statistically generalizable findings but rather to demonstrate the relevance of the IF-SWARA method in prioritizing the key factors contributing to brain drain from the perspective of individuals who have experienced it firsthand. Since the participants resided outside the country, individual online interviews were conducted with each participant. They were subsequently able to assess the criteria in accordance with the fuzzy MCDM methodology employed in the study.

It is widely acknowledged that fuzzy MCDM methods offer reliable results when resolving issues involving qualitative criteria. Participants articulate their opinions on these criteria using fuzzy linguistic scales, which are subsequently converted into fuzzy numbers to derive solutions. Consequently, it is crucial to highlight that this methodology diverges from studies that collect data via surveys with extensive sample sizes. Similarly to qualitative research, the methodology selected in this study emphasizes the appropriateness of participants for the research rather than concentrating on increasing the participant count.

Given the challenges of reaching dispersed, highly skilled immigrants and the often-sensitive nature of this topic, a snowball sampling strategy was employed. While this introduces limitations in terms of representation, the study offers valuable exploratory findings that can guide future research with more diverse samples.

Table 2. Details of the Decision Makers.

DM	Degree	Work Experience (Year)	Duration of Migration (Year)	DM	Degree	Work Experience (Year)	Duration of Migration (Year)
DM1	Master	10+	2–5	DM9	Bachelor	10+	6–10
DM2	Master	10+	10+	DM10	Bachelor	10+	2–5
DM3	Master	2–5	2–5	DM11	Master	0–2	6–10
DM4	Master	2–5	6–10	DM12	Bachelor	0–2	0–2
DM5	Master	2–5	0–2	DM13	Master	10+	6–10
DM6	PhD	10+	10+	DM14	PhD	6–10	2–5
DM7	Bachelor	6–10	2–5	DM15	PhD	2–5	0–2
DM8	Bachelor	2–5	2–5	DM16	Bachelor	6–10	2–5

presents certain demographic details of these highly skilled immigrants (designated as DM, or decision makers). These immigrants, who are Turkish citizens, possess at least a bachelor’s degree from their home country and are proficient in at least one foreign language, typically English. The study aims to investigate the motivations behind the brain drain experienced by these highly skilled immigrants.

Furthermore, Figure 2 illustrates gender, country of residence, and return intentions in the form of a Sankey diagram.

4. Results

During the initial implementation phase of the IF-SWARA, sixteen decision-makers determined the criteria’s relative importance values, ranking them from most to least important. Then, each decision-maker evaluated the main and sub-criteria using the linguistic terms in Table 1. After the evaluations, the criteria weights were calculated separately using Equations (1)–(8). The resulting criteria weights based on DM-1’s evaluations are shown in

Table 3. Final Weights of Main Criteria (DM-1).

Criteria	Sj Values			Coefficient Kj (Sj + 1)			Recalculated Weight Qj			Fuzzy Weight			Crisp Weight	Final Weight
C				1	1	1	1	1	1	0.304	0.313	0.326	0.314	0.314
E	0	0	0	1.00	1.00	1.00	1.000	1.000	1.000	0.304	0.313	0.326	0.314	0.314
G	0.33	0.4	0.5	1.33	1.40	1.50	0.667	0.714	0.752	0.203	0.224	0.245	0.224	0.224
P	0.40	0.5	0.67	1.40	1.50	1.67	0.399	0.476	0.537	0.121	0.149	0.175	0.149	0.149

and

Table 4. Final Weights of Sub-Criteria (DM-1).

Sub Criteria	Sj Values			Coefficient Kj (Sj + 1)			Recalculated Weight Qj			Fuzzy Weight			Aggregated Fuzzy Weight			Crisp Weight	Final Weight
C2				1	1	1	1	1	1	0.293	0.306	0.324	0.089	0.096	0.106	0.097	0.306
C3	0.22	0.25	0.29	1.22	1.25	1.29	0.775	0.800	0.820	0.227	0.245	0.265	0.069	0.077	0.087	0.077	0.245
C1	0.22	0.25	0.29	1.22	1.25	1.29	0.601	0.640	0.672	0.176	0.196	0.218	0.054	0.061	0.071	0.062	0.196
C5	0.33	0.4	0.5	1.33	1.4	1.5	0.401	0.457	0.505	0.117	0.140	0.164	0.036	0.044	0.053	0.044	0.140
C4	0.22	0.25	0.29	1.22	1.25	1.29	0.311	0.366	0.414	0.091	0.112	0.134	0.028	0.035	0.044	0.035	0.112
E1				1	1	1	1	1	1	0.265	0.274	0.288	0.081	0.086	0.094	0.086	0.274
E5	0	0	0	1	1	1	1.000	1.000	1.000	0.265	0.274	0.288	0.081	0.086	0.094	0.086	0.274
E2	0.29	0.33	0.4	1.29	1.33	1.4	0.714	0.752	0.775	0.189	0.206	0.223	0.058	0.065	0.073	0.065	0.206
E4	0.33	0.4	0.5	1.33	1.4	1.5	0.476	0.537	0.583	0.126	0.147	0.168	0.038	0.046	0.055	0.046	0.147
E3	0.40	0.5	0.67	1.4	1.5	1.67	0.285	0.358	0.416	0.076	0.098	0.120	0.023	0.031	0.039	0.031	0.098
G1				1	1	1	1	1	1	0.229	0.242	0.260	0.047	0.054	0.064	0.055	0.242
G3	0.33	0.4	0.5	1.33	1.4	1.5	0.667	0.714	0.752	0.153	0.173	0.196	0.031	0.039	0.048	0.039	0.173
G4	0	0	0	1	1	1	0.667	0.714	0.752	0.153	0.173	0.196	0.031	0.039	0.048	0.039	0.173
G6	0	0	0	1	1	1	0.667	0.714	0.752	0.153	0.173	0.196	0.031	0.039	0.048	0.039	0.173
G5	0.40	0.5	0.67	1.4	1.5	1.67	0.399	0.476	0.537	0.092	0.115	0.140	0.019	0.026	0.034	0.026	0.115
G2	0.33	0.4	0.5	1.33	1.4	1.5	0.444	0.510	0.565	0.102	0.124	0.147	0.021	0.028	0.036	0.028	0.124
P2				1	1	1	1	1	1	0.256	0.276	0.306	0.031	0.041	0.054	0.042	0.275
P3	0.33	0.4	0.5	1.33	1.4	1.5	0.667	0.714	0.752	0.171	0.197	0.230	0.021	0.029	0.040	0.030	0.197
P1	0.40	0.5	0.67	1.4	1.5	1.67	0.399	0.476	0.537	0.102	0.132	0.165	0.012	0.020	0.029	0.020	0.132
P4	0	0	0	1	1	1	0.399	0.476	0.537	0.102	0.132	0.165	0.012	0.020	0.029	0.020	0.132
P5	0	0	0	1	1	1	0.399	0.476	0.537	0.102	0.132	0.165	0.012	0.020	0.029	0.020	0.132
P6	0.00	0	0.00	1	1	1	0.399	0.476	0.537	0.102	0.132	0.165	0.012	0.020	0.029	0.020	0.132

.

After calculating the criteria weights for each decision-maker, the final step was to combine these weights using the geometric mean. The resulting criteria weights are shown in

Table 5. Final Weights.

	DM1	DM2	DM3	DM4	DM5	DM6	DM7	DM8	DM9	DM10	DM11	DM12	DM13	DM14	DM15	DM16	Mean	Weight	Rank
C1	0.062	0.029	0.030	0.083	0.030	0.066	0.156	0.094	0.051	0.052	0.045	0.026	0.085	0.036	0.071	0.0350	0.052	0.058	4
C2	0.096	0.046	0.092	0.047	0.083	0.050	0.093	0.028	0.071	0.092	0.059	0.060	0.046	0.080	0.089	0.0730	0.065	0.073	1
C3	0.077	0.037	0.073	0.083	0.059	0.037	0.067	0.063	0.021	0.069	0.045	0.060	0.064	0.054	0.034	0.0251	0.051	0.056	5
C4	0.035	0.037	0.055	0.031	0.042	0.028	0.040	0.042	0.031	0.039	0.032	0.043	0.034	0.026	0.051	0.0487	0.038	0.042	12
C5	0.044	0.018	0.041	0.063	0.020	0.099	0.033	0.063	0.016	0.026	0.079	0.019	0.085	0.019	0.051	0.0189	0.036	0.040	14
E1	0.086	0.058	0.039	0.073	0.047	0.136	0.087	0.076	0.054	0.028	0.039	0.031	0.033	0.085	0.045	0.0390	0.054	0.060	3
E2	0.065	0.058	0.039	0.073	0.063	0.070	0.052	0.046	0.036	0.028	0.052	0.031	0.020	0.085	0.075	0.0390	0.048	0.054	7
E3	0.031	0.033	0.013	0.033	0.020	0.038	0.019	0.017	0.018	0.028	0.021	0.019	0.020	0.057	0.075	0.0196	0.026	0.029	21
E4	0.046	0.058	0.028	0.073	0.034	0.052	0.032	0.022	0.022	0.028	0.016	0.039	0.033	0.038	0.056	0.0261	0.035	0.039	16
E5	0.086	0.044	0.020	0.055	0.012	0.097	0.087	0.033	0.013	0.028	0.028	0.019	0.042	0.057	0.045	0.0196	0.035	0.039	15
G1	0.054	0.080	0.070	0.060	0.038	0.025	0.041	0.012	0.087	0.047	0.037	0.095	0.064	0.038	0.042	0.0812	0.049	0.054	6
G2	0.028	0.043	0.070	0.023	0.059	0.018	0.018	0.017	0.044	0.019	0.037	0.063	0.043	0.023	0.032	0.0542	0.033	0.037	17
G3	0.039	0.080	0.087	0.045	0.059	0.018	0.030	0.042	0.145	0.063	0.052	0.063	0.043	0.023	0.042	0.0388	0.048	0.053	8
G4	0.039	0.060	0.045	0.034	0.059	0.034	0.068	0.026	0.058	0.026	0.028	0.063	0.031	0.023	0.023	0.0388	0.038	0.042	11
G5	0.026	0.043	0.037	0.034	0.034	0.034	0.030	0.012	0.032	0.036	0.021	0.042	0.021	0.023	0.016	0.0388	0.028	0.031	19
G6	0.039	0.029	0.056	0.034	0.047	0.012	0.022	0.021	0.035	0.018	0.020	0.038	0.022	0.014	0.015	0.0292	0.026	0.029	20
P1	0.020	0.043	0.014	0.019	0.024	0.016	0.015	0.157	0.022	0.021	0.095	0.035	0.075	0.028	0.027	0.0316	0.031	0.034	18
P2	0.041	0.043	0.023	0.013	0.082	0.024	0.025	0.026	0.059	0.117	0.046	0.059	0.075	0.104	0.068	0.0585	0.046	0.051	9
P3	0.029	0.043	0.058	0.052	0.082	0.032	0.025	0.094	0.099	0.088	0.127	0.059	0.054	0.069	0.041	0.1313	0.060	0.067	2
P4	0.020	0.043	0.058	0.025	0.055	0.059	0.025	0.057	0.040	0.066	0.068	0.059	0.054	0.046	0.027	0.0875	0.046	0.051	10
P5	0.020	0.034	0.039	0.035	0.036	0.044	0.025	0.035	0.029	0.047	0.033	0.059	0.023	0.046	0.054	0.0420	0.036	0.040	13
P6	0.020	0.043	0.014	0.009	0.015	0.012	0.011	0.019	0.017	0.032	0.022	0.021	0.032	0.028	0.020	0.0238	0.019	0.022	22

.

Based on Table 5, “workplace conditions (C2)”, “living conditions (P3)”, and “academic standards (E1)” are identified as the most significant criteria behind the motivations of the Turkish individuals. Conversely, “identity expression (P6)”, “scholarships or grants (E3)”, and “disaster reliance (G6)” are considered less critical criteria. Overall, career-related factors accounted for 26.9%, education-related factors for 22.1%, governance and political climate for 24.6%, and personal motives for 26.5%, highlighting a nearly balanced distribution across domains.

The findings, shown in the Sankey diagram, provide demographic and professional details about Turkish individuals living in different countries. The analysis emphasizes education, employment status, and plans for returning to Turkey. A large majority, 75%, hold either bachelor’s or master’s degrees, while a smaller percentage have doctorates. These Turkish expatriates are spread across several countries, including Germany (25%), the United Kingdom (20%), the Netherlands (15%), as well as Sweden, Poland, Switzerland, and the United States. They work in various fields such as automotive, information technology, textiles, and academia, showing a wide range of expertise. Notably, 70% of this group has more than five years of work experience, indicating a highly skilled workforce. Although their time abroad varies, only 10% plan to return to Turkey, suggesting a strong desire to stay in their host countries. This highlights important issues like brain drain and the global movement of skilled professionals.

5. Discussion

This paper aims to investigate the motivations behind the brain drain of highly educated individuals from Turkey, utilizing data collected from sixteen participants. A fuzzy multi-criteria decision-making model is employed to explore the factors that drive Turkey’s talented citizens to emigrate. The methodological rigor of the IF-SWARA approach offered a precise quantification of expert judgments, thereby strengthening the reliability of the findings. The twenty-two criteria influencing brain drain phenomena are classified into four primary categories: career, education, governance and politics, and personal motives. Workplace conditions (C2, 0.073), living conditions (P3, 0.067), and academic standards (E1, 0.06) were identified as the top-ranked migration drivers. These results support prior research highlighting economic dissatisfaction, job insecurity, and wage disparities as critical elements in the migration decisions of skilled professionals [4,6,7]. While the emphasis has been placed on top-ranked drivers, the remaining criteria such as social policies (G5, 0.031), scholarships and grants (E3, 0.029) or disaster reliance (G6, 0.029) also offer important insight. However, they were ranked lower by participants. This may reflect the specific profile of our sample, which largely consists of recent emigrants with highly educated and predominantly employed backgrounds.

The second major insight of this study is the misalignment between individual aspirations and structural realities. Respondents consistently ranked career and economic concerns higher than socio-cultural or contextual factors such as restricted opportunities for scholarships or grants, limitations on identity expression, or disaster vulnerability, suggesting that while personal circumstances may shape individual experiences, they are not perceived as principal determinants of brain drain at a systemic level. This result is consistent with migration trends observed in countries such as Malaysia, where limited upward mobility in the STEM fields accelerates talent loss [18], and South Africa, where economic volatility and institutional weaknesses have discouraged the retention of skilled professionals [72].

This systemic nature of brain drain has critical implications for policy. Addressing economic push factors alone will not suffice. There is an urgent need for comprehensive strategies that also tackle governance inefficiencies and lack of institutional trust. Institutional quality is vital for reversing the trend of brain drain. Enhancing transparency and accountability processes inside public institutions is essential for fostering public confidence. Anti-corruption policies, consistent with international norms, must be implemented to restore confidence and trust in governance [73]. These include the promotion of meritocratic career systems, investment in research infrastructure, and diaspora engagement mechanisms that foster knowledge exchange and reverse migration. For countries like Turkey, such interventions could reframe brain drain as an opportunity for strategic reconfiguration of human capital flows rather than an inevitable loss.

While the limited sample size constrains the ability to generalize statistically or to control for variables such as age, occupational field, or socioeconomic status, this study primarily aims to demonstrate the methodological value of applying MCDM techniques to migration-related decision-making contexts. In support of the findings, the demographic and professional profile of Turkish immigrants further underscores the persistent nature of skilled migration. Data show that approximately 75% of Turkish professionals residing abroad hold either bachelor’s or master’s degrees, with a notable proportion also possessing doctoral qualifications. Their dispersion across countries such as Germany (25%), the UK (20%), the Netherlands (15%), and others—including the United States and several EU nations—demonstrates the global scope of Turkey’s human capital outflow. This underscores the significance of investing in educational and research infrastructure. Promoting international collaborations in research, facilitated by accessible funding mechanisms, will enhance Turkey’s status as a center for research and innovation [73].

Not only are these individuals highly educated, but they are also deeply experienced, with 70% having over five years of professional experience across critical sectors like automotive, IT, textiles, and academia. It will be essential to improve research and development incentives in these industries in order to mitigate the tendency of brain drain. The fact that only 10% express a desire to return to Turkey highlights the structural stickiness of emigration decisions, influenced by both positive host-country integration and negative home-country inertia. These patterns suggest that retaining or reattracting such talent will require not only economic incentives but also robust institutional reforms and strategic diaspora engagement to rebuild trust and opportunity structures in the home country.

Finally, this study underscores the methodological contributions of FMCDM approaches to migration research. The integration of fuzzy logic with decision-making models provides a valuable toolset for capturing expert insights in areas characterized by uncertainty and multidimensional complexity. The weighting outcomes across various domains offers a replicable framework that can be tailored to different country contexts.

6. Conclusions, Limitations and Further Research Directions

This study investigated the underlying motivations behind the emigration of highly skilled Turkish individuals using the IF-SWARA method within the FMCDM framework. The findings underscored that career-related (0.269) and personal motives (0.265) serve as the primary drivers of migration decisions. Within the career domain, workplace conditions (C2, 0.073) and employment opportunities (C1, 0.058) emerged as the most influential factors, followed by salary policies (C3, 0.056). On the personal motives dimension, living conditions (P3, 0.067) emerged as the most influential factor, outweighing both the desire to reside overseas (P2, 0.051) and cultural diversity (P4, 0.051). These results aligned with previous literature identifying economic insecurity and labor market mismatches as principal triggers of brain drain, confirming that the issue is deeply rooted in structural imbalances within national systems [4,15]. The proposed decision model is designed to assist policymakers in identifying the needs of skilled individuals contemplating immigration. Successful policy implementation necessitates a comprehensive understanding of the factors driving migration. Therefore, the insights derived from this research could prove to be invaluable at the national level. Merely identifying the causes of brain drain is insufficient; it is equally essential to develop viable solutions.

The application of FMCDM methods enabled a nuanced and systematic evaluation of a broad spectrum of drivers, capturing both the relative importance and expert consensus across sixteen decision-makers. Notably, the results indicated that personal or contextual factors, such as identity or disaster risk, while not irrelevant, were markedly less influential in the decision calculus. This suggests that policy efforts should prioritize reforms in professional development, wage structures, and meritocratic employment practices.

This study situates the Turkish experience within a global context, emphasizing similar trends as the migration of skilled personnel and professionals is critical for many developing nations, which are losing their well-trained citizens to wealthier countries [13]. These cases reinforce the idea that brain drain is not a symptom of underdevelopment alone, but a reflection of deeper mismatches between human capital and opportunity landscapes. Such a perspective demands more nuanced, evidence-based interventions beyond temporary economic incentives.

Complementing the analytical findings, the demographic evidence reveals a concerning trend: the majority of skilled Turkish migrants are not only well-qualified and sector diverse but also exhibit a long-term commitment to residing abroad. While the sample size limits the extent to which findings can be generalized to broader populations, the aim of this study is to highlight the priorities among individuals who have experienced migration. The relatively low intention to return—reported at just 10%—reflects deep-rooted dissatisfaction with domestic conditions and confidence in foreign systems. This challenges policymakers to reconsider simplistic retention policies and instead adopt a multidimensional approach that addresses career advancement, research infrastructure, institutional trust, and sociopolitical climate.

The current study is constrained by the number of participants; consequently, expanding the respondent pool could yield more comprehensive insights. A larger and more diversified sample of respondents would facilitate occupation-specific categorization, potentially uncovering more nuanced patterns and trends associated with the brain drain phenomenon. Nevertheless, the use of FMCDM allowed for a structured and transparent prioritization of complex and interrelated criteria.

As emphasized above, the most important limitation of the study is its limited sample size. While this does not pose a significant issue for MCDM methods, which are effective with limited samples, it is important to acknowledge that the results do not necessarily represent all highly skilled Turkish immigrants. The primary objective here is to investigate the underlying causes of the brain drain among highly educated individuals through the application of an MCDM methodology.

The issue of brain drain presents a considerable challenge, especially for developing nations, and it recommends various avenues for future research. Future investigations could gain from an analysis of a broader array of developing countries to better comprehend the factors contributing to brain drain. Furthermore, concentrating on a particular developed country as a host could provide more insight into the characteristics that attract skilled individuals.

In conclusion, brain drain represents a multidimensional policy challenge with implications for economic growth, social cohesion, and long-term national development. However, when understood through robust analytical frameworks like FMCDM and addressed through systemic, inclusive strategies, it can also be reimagined as a lever for transformation. Future research should further explore the interplay between virtual mobility, digital labor markets, and migration intentions, thereby offering new paradigms for managing global talent flows in an interconnected world.