Key determinants of university dropout: academic, economic, and psychosocial factors in student retention

Introduction: University dropout remains a persistent challenge in higher education, reflecting structural tensions between inclusivity, student retention, and educational equity. Rather than approaching dropout as a simple aggregation of causes, this study adopts a comprehensive perspective that integrates academic, economic, and psychosocial dimensions, emphasizing their interaction with institutional contexts and responses.

Methods: A systematic literature review was conducted following the PRISMA 2020 guidelines. Nineteen empirical studies were selected and analyzed to identify the main determinants of university dropout, as well as the institutional strategies and theoretical models used to understand and prevent student attrition. Data were synthesized through a comparative and thematic analysis, focusing on the frequency and articulation of determinants across studies.

Results: The analysis indicates that motivation (73.7%), academic performance (57.9%), and financial hardship (31.6%) are the most recurrent determinants of dropout. Psychosocial factors, particularly emotional well-being and social integration, also show a substantial influence on student retention. In terms of institutional responses, predictive analytics, early warning systems, and comprehensive student support programs emerge as the most effective strategies for identifying at-risk students and reducing dropout rates.

Discussion: Compared with prior reviews, the findings reveal limited theoretical integration and methodological consistency, as most studies focus on isolated variables without adequately linking them to broader institutional frameworks. The evidence highlights the need for multidimensional and theoretically grounded approaches that connect academic, economic, and psychosocial factors with coordinated institutional action. Such frameworks are essential for strengthening student retention strategies and advancing equity in higher education.

1 Introduction

University dropout remains one of the most persistent challenges facing higher education systems worldwide. Beyond its individual impact on students’ academic trajectories, dropout undermines institutional efficiency, widens social inequality, and limits economic and scientific progress. In many contexts, high attrition rates reflect structural weaknesses in educational systems, revealing the interplay of academic, economic, and psychosocial barriers that hinder student retention. Consequently, ensuring continuity in university education has become a central concern of public policy and an increasingly prominent theme in academic research (Fernandes, 2024; Lamichhane et al., 2025).

In the context of higher education, dropout or student attrition refers to the interruption of studies before obtaining the intended academic degree, either temporarily or permanently, and without formal transfer to another institution (Tinto, 1993; Bean, 1985). From a theoretical standpoint, Tinto (1993) conceptualizes dropout as the result of an insufficient level of academic and social integration, while Bean (1985) interprets it as a decision process influenced by students’ attitudes, satisfaction, and external circumstances. Operationally, this study adopts the definition proposed by UNESCO (2012), which describes university dropout as “the premature termination of an educational trajectory before the achievement of a formal qualification.” This definition captures both the institutional and individual dimensions of the phenomenon and aligns with the multidimensional approach adopted in this research.

Existing studies have examined this phenomenon from multiple perspectives, identifying diverse determinants of student withdrawal. Academic factors include poor performance, curricular misalignment, and low teaching quality; economic factors encompass financial hardship, employment demands, and family constraints; while psychosocial factors involve motivation, emotional distress, social disconnection, and lack of belonging (Vera et al., 2025). Despite this progress, the literature remains fragmented, with inconsistent conceptual frameworks and methodological approaches that make it difficult to establish robust relationships among variables (Apumayta et al., 2024).

A more integrative and systematic analysis of these factors—together with the theoretical models and institutional strategies that address them—is essential for designing effective interventions. Although numerous studies have explored dropout in higher education, the dispersion of evidence has limited the development of comparable diagnoses and hindered the formulation of evidence-based policies (Apumayta et al., 2024; Barragán-Moreno and González-Támara, 2024). Research focusing exclusively on academic performance often neglects emotional and contextual dimensions, while analyses of economic conditions treat financial variables in isolation from institutional or psychosocial contexts (Muro et al., 2024). Likewise, theoretical and predictive models of dropout have been inconsistently validated and insufficiently adapted to diverse cultural realities (Núñez-Naranjo, 2024).

This gap underscores the need for a comprehensive synthesis that classifies and relates the academic, economic, and psychosocial determinants of university dropout, while also reviewing the institutional approaches and theoretical models developed to explain and mitigate it. Accordingly, this study aims to explore the main factors influencing university dropout and the institutional and theoretical frameworks guiding their analysis and prevention. To achieve this, five research questions were formulated:

1. Which academic factors have been most frequently identified as determinants of university dropout?

2. Which economic conditions significantly influence the continuation of university studies?

3. Which psychosocial variables have been associated with a higher risk of dropping out among university students?

4. Which institutional approaches have been implemented to prevent or reduce dropout in higher education?

5. Which theoretical or predictive models have been applied to explain university dropout in different contexts?

This research adopts an integrative and comparative approach that organizes evidence from three key dimensions academic, economic, and psychosocial while connecting them with institutional and theoretical perspectives. Its added value lies in providing an analytical synthesis that not only identifies the most influential factors but also establishes conceptual bridges between them. This articulation contributes to the development of evidence-based educational policies and offers a conceptual foundation for future research aimed at strengthening student retention in higher education.

The primary factors associated with university attrition, as identified in the 19 peer-reviewed studies published between 2013 and 2024, using the PRISMA 2020 framework, are categorized into three key areas:

Academic performance: Academic performance emerged as one of the most significant predictors of student dropout. The studies found a strong correlation between low GPA, repeated failures in the same courses, excessive academic workload, and poor preparedness for university studies. In addition, the quality of teaching and assessment methods were found to influence both academic success and the length of students’ academic careers.

Financial issues: Financial difficulties were consistently identified as a major contributing factor to university attrition. Key issues included the need to work while studying, insufficient financial aid, and low family income. These factors often created stress for students, making it more challenging for them to focus on their academic responsibilities and increasing their likelihood of dropping out.

Psychosocial factors: The most common psychosocial reasons for dropout included a lack of motivation, emotional distress, difficulty fitting in socially, and family obligations. Students who did not feel a sense of belonging or who experienced significant anxiety were found to be at a higher risk of dropping out or ceasing participation in their programs.

It is also important to acknowledge that several studies approach the issue from the opposite perspective, analyzing student success or academic persistence rather than dropout itself. These works conceptualize success as the effective continuation and completion of academic programs, focusing on predictors of retention and achievement (Fernández-Martín et al., 2019; Nikolaidis et al., 2022). Recognizing student success as the conceptual counterpart of dropout enriches the analysis, as both perspectives address the same phenomenon through different lenses—one oriented toward risk factors and the other toward protective and institutional reinforcement mechanisms. This complementarity supports a more comprehensive understanding of university trajectories.

2 Methodology

The present study adhered to the PRISMA 2020 guidelines (Page et al., 2021) to ensure methodological transparency, reproducibility, and analytical rigor. The protocol encompassed seven components: information sources, search strategy, eligibility criteria, study selection process, data extraction, risk-of-bias assessment, and synthesis procedures. All methodological decisions were meticulously documented to ensure consistency and replicability.

2.1 Eligibility criteria

Inclusion criteria. Studies were eligible if they: (1) constituted peer-reviewed empirical or theoretical research on factors, causes, or determinants of university dropout or student retention; (2) were published between 2013 and 2023 in English or Spanish; (3) examined at least one of the core dimensions (academic, economic, or psychosocial); and (4) provided sufficient methodological detail to extract variables, analytical procedures, and results.

Exclusion criteria. Studies were excluded if they: (1) lacked accessible full text; (2) corresponded to non-peer-reviewed sources such as conference abstracts, editorials, letters, or opinion pieces; (3) did not explicitly address dropout or retention determinants; or (4) presented significant methodological deficiencies, including absence of theoretical grounding, unclear analytic procedures, or lack of empirical evidence.

Screening process. Two researchers independently screened titles, abstracts, and full texts. Disagreements were resolved through consensus, and the final set of included studies was validated by a senior reviewer. Exclusion was applied in three sequential phases: elimination of duplicates and poorly indexed documents, removal of studies without full-text availability, and final exclusion of works that did not meet thematic or methodological requirements. Priority was granted to empirical studies using validated instruments and to research analyzing dropout determinants at the undergraduate level.

Rationale for timeframe and methodological scope. The 2013–2023 window was selected to capture recent transformations in higher education systems, including the expansion of online learning and the post-pandemic restructuring of student support. Studies using causal or reciprocal modeling approaches (e.g., SEM) were retained when methodologically transparent, whereas theoretical works without empirical testing were excluded to preserve methodological coherence.

2.2 Sources of information

For the present systematic review, the Scopus and Web of Science (WoS) databases were utilized, both recognized for their multidisciplinary scope and rigorous indexing standards. As peer-reviewed sources, these platforms ensure the scientific quality of the included studies. Elsevier’s Scopus comprises more than 25,000 journals across fields such as social sciences, psychology, and education, while Clarivate’s Web of Science indexes a curated set of publications with strong representation in higher education and student behavior research. The selection of these databases was strategically made to guarantee comprehensive coverage and equitable geographic representation. Although their regional emphases differ, their combined use mitigates potential bias and strengthens the documentary corpus (Asubiaro et al., 2024).

A total of 780 records were initially retrieved from Scopus and 640 from WoS. The figures reported in the PRISMA flow diagram (Scopus = 64; WoS = 38; total = 102) correspond to the post-filter export sets obtained after applying limits by document type (peer-reviewed articles), language (English/Spanish), and publication window (2013–2023). The larger numbers (Scopus = 780; WoS = 640) reflect the raw retrievals prior to filtering. After export, the post-filter sets were merged, de-duplicated, and subsequently screened. All records were managed using a bibliographic tool, which facilitated the identification of duplicates and streamlined the review process. Following this procedure, studies with adequate methodological quality and direct relevance to factors associated with university dropout were selected.

The temporal range of analysis was updated to 2013–2024 to include the most recent empirical evidence available at the time of the review. The original cutoff of 2023 corresponded to the indexing cycle of Scopus and WoS during the data collection phase (first semester of 2024), when publications from that year were still undergoing processing. Extending the window to 2024 allowed the incorporation of newly indexed studies while maintaining methodological consistency. Thus, the cutoff reflects the latest year with consolidated and retrievable data across both databases as of early 2025.

The systematic review was conducted between 15 and 20 April 2025. The comprehensive search yielded approximately 2,000 records, which were exported with complete bibliographic metadata, abstracts, and author keywords. Additional databases were intentionally excluded to preserve consistency in indexing criteria and ensure reproducibility. To reinforce the completeness of the evidence base, forward and backward citation chasing was performed for all included studies, thereby minimizing the risk of omitting relevant literature.

2.3 Search strategy

Separate search equations were defined for each database, based on the inclusion criteria and key concepts associated with university dropout, student retention, and academic, economic, and psychosocial factors. Boolean operators AND and OR were used to combine terms and broaden the thematic coverage of the retrieved records. In Scopus, the following formula was used:

(TITLE (“university dropout” OR “student attrition” OR “higher education dropout”) OR KEY (“university dropout” OR “student attrition” OR “higher education dropout”)) AND TITLE (“factors” OR “determinants” OR “causes”) In Web of Science, the search structure was adapted to that platform’s field system: (TS = (“university dropout” OR “student attrition” OR “higher education dropout”) OR AK = (“university dropout” OR “student attrition” OR “higher education dropout”)) AND TS = (“factors” OR “determinants” OR “causes”) In both sections, filters were applied by language (English and Spanish), document type (peer-reviewed scientific articles), and publication period (2013–2023) to narrow the results to the study’s relevance framework.

Searches were executed between [insert exact dates], with the following database-level filters active in both platforms: document type = peer-reviewed article, language = English or Spanish, year range = 2013–2023. We exported all records with full bibliographic metadata, abstracts, and author keywords. Counts reported in the PRISMA flow refer to post-filter exports.

2.4 Selection process

The selection process followed four sequential phases in accordance with the PRISMA approach. First, studies were identified through searches in Scopus and Web of Science, yielding 1,420 raw records (Scopus = 780; WoS = 640). Database-level filters (document type, language, year) reduced this to 102 records (Scopus = 64; WoS = 38). After automated de-duplication in a bibliographic manager (Zotero), 27 duplicates were removed, leaving 75 unique records for title and abstract screening.

During screening, 14 records were excluded (conference papers = 7; unrelated topics = 7), and 61 full texts were sought. Despite multiple retrieval attempts via academic databases and institutional repositories, 35 reports could not be accessed (primarily conference papers, regional journals without open access, or items with incomplete metadata) and were therefore excluded following PRISMA 2020 guidance. Of the 26 full texts assessed for eligibility, 7 were excluded for not meeting the inclusion criteria. Nineteen studies were retained for synthesis (see Figure 1). Discrepancies at each phase were resolved by consensus between at least two independent reviewers, with final validation by a senior researcher.

Figure 1

Figure 1. PRISMA flowchart. Prepared by the authors based on Scopus and Web of Science. “Scopus (n = 64) and Web of Science (n = 38)” denote post-filter export counts; “780 and 640” in Section 2.2 refer to raw, pre-filter retrievals. Discrepancies reflect sequential filtering and de-duplication steps specified in Sections 2.2–2.4.

Figure 1 presents the PRISMA flowchart with precise retrieval dates and database sources. The searches were performed on 15 April 2025 in Scopus and Web of Science; the flowchart reports the numbers above and documents the removal of duplicates, screening decisions, retrieval attempts, and final inclusions.

To ensure transparency and replicability, the study defined and operationalized the academic, economic, and psychosocial variables within a coding framework. Academic variables (e.g., GPA, grades, assessment methods) were extracted using specific performance indicators reported in the included studies. Economic variables (e.g., financial hardship, employment status, scholarship access) were derived from authors’ reports on students’ financial circumstances. Psychosocial variables (e.g., motivation, emotional well-being, social integration) were measured via self-report surveys and psychological assessments where available. Definitions and measurement approaches were standardized across the dataset to promote consistency and comparability in the synthesis.

2.5 Data processing

The data processing was conducted using Microsoft Excel, which enabled the systematic organization, classification, and analysis of the information extracted from the selected studies. A structured extraction matrix was developed to consolidate key variables, including authors and year of publication, country and context, study design and sample characteristics, academic, economic and psychosocial factors, institutional approaches and theoretical frameworks, and main findings with their reported direction of effect.

This matrix supported the thematic coding process, the identification of recurring patterns, and cross-study comparisons, allowing for a cross-sectional analysis of the determinants associated with university dropout. To ensure analytical consistency, each extracted variable was classified into one of the predefined thematic dimensions—academic, economic or psychosocial based on the conceptual definitions established for the review. Subcategories within each dimension were created inductively according to the content reported in the studies (e.g., academic performance, study habits, learning difficulties or attendance patterns). Numerical codes were assigned to categories and subcategories, enabling systematic comparison across studies and interpretation of the direction and strength of the reported effects.

Two independent reviewers conducted the extraction and coding processes, compared outcomes, and resolved discrepancies through consensus. Standardization procedures were applied throughout to ensure reliability, coherence and conceptual alignment across all included studies.

2.6 Risk of bias

The risk of bias was assessed through cross-referencing and peer review. The methodological quality criteria that were applied included the clarity of objectives, the validity of instruments, analytical consistency, and the relationship between results and conclusions. Despite the absence of a clinical review, these criteria were utilized in order to minimize bias. The potential biases arising from the utilization of particular databases, the formulation of search terms, and the existence of reporting bias in the reviewed studies are recognized.

Coverage and completeness checks. To mitigate omissions, we conducted backward and forward citation chasing of all included papers and scanned top-cited records surfaced by both databases under the same filters. Studies perceived as “relevant” but excluded fell outside the a priori criteria (e.g., different educational level; outside 2013–2023 window; not peer-reviewed; lack of empirical testing of causal claims/SEM; or unavailable full text). which lists excluded but potentially relevant records flagged during peer review and post-hoc checks.

The methodological design was refined to strengthen alignment between the research questions and the analytical process. Each of the five questions guided a specific coding dimension within the data extraction matrix, corresponding, respectively, to academic, economic, psychosocial, institutional, and theoretical factors. The analysis incorporated cross-tabulation and co-occurrence mapping to identify interrelations among determinants rather than isolated frequencies. This ensured that the synthesis preserved the multidimensional nature of the phenomenon, enabling a more comprehensive interpretation of how these factors converge to explain student dropout.

In addition, a validation procedure was conducted to ensure the methodological robustness of the review. Two independent researchers verified the consistency of the coding, and disagreements were resolved through consensus. This process enhanced inter-rater reliability and analytical transparency. The methodological reinforcement directly responds to the research questions by connecting each determinant category to its theoretical and institutional implications, ensuring that the findings move beyond description to provide explanatory insight into the factors that shape university dropout.

The following table presents a summary of the methodological quality assessment of the included studies using the Joanna Briggs Institute (JBI) Critical Appraisal Tool. This systematic evaluation tool helps in assessing the methodological rigor of the studies included in the review, focusing on aspects such as study design, sample size, validity of instruments, and analysis methods. The results are categorized as “Low Risk,” “Moderate Risk,” or “High Risk” to provide a transparent view of the quality of the studies included (see Table 1).

Table 1

Table 1. Methodological quality assessment of the included studies using the Joanna Briggs Institute (JBI) Critical Appraisal Tool.

3 Results

A total of 19 studies met the inclusion criteria and were incorporated into the qualitative synthesis (Tables 2, 3). The findings reveal a multidimensional configuration of determinants influencing university dropout, where academic, economic, and psychosocial factors interact dynamically rather than acting as independent predictors. This section presents the thematic synthesis and relational interpretation derived from cross-study comparisons.

Table 2

Table 2. Studies included in the research.

Table 3

Table 3. Details of each included study.

A meticulous examination of the extant literature reveals that academic determinants were the most frequently reported factors across the included studies. As demonstrated in Figure 2, academic variables such as substandard performance, inadequate study habits, irregular class attendance, and learning difficulties collectively constitute a significant proportion of the identified determinants. However, the relational analysis demonstrates that these variables rarely act independently.

Figure 2

Figure 2. Most frequent academic factors. Prepared by the authors based on Scopus and Web of Science.

Academic underperformance frequently co-occurs with financial strain and emotional distress, thereby amplifying its effect on the risk of dropping out. A multitude of studies have indicated that academic challenges may function as precursors to more profound contextual barriers. For instance, students experiencing economic hardship or psychological strain tend to exhibit reduced academic engagement, which then leads to deterioration in performance over time.

Conversely, institutional mechanisms, such as tutoring programs, access to learning resources, and timely academic feedback, play a moderating role, as suggested by the clusters observed in Figure 3, which shows the co-occurrence of academic determinants with institutional support categories. These findings suggest that academic vulnerability should be conceptualized as a component of a more extensive system of interacting stressors, rather than a standalone predictor.

Figure 3

Figure 3. Most frequent economic factors. Prepared by the authors based on Scopus and Web of Science.

The analysis reveals both direct effects, such as the inability to cover tuition or transportation costs, and indirect effects, including the undermining of concentration due to financial stress, the increase in absenteeism, and the forced engagement in precarious employment. The relational patterns illustrated in Figure 3 demonstrate a robust correlation between economic pressures and psychosocial strain, suggesting that financial challenges frequently precipitate a sequence of deleterious consequences that ultimately affect academic outcomes.

The investigation revealed that structural factors, including socioeconomic background and the absence of institutional financial aid, emerge as long-term risk predictors. The studies reporting comprehensive financial and academic support demonstrate comparatively lower rates of attrition, thereby underscoring the compensatory role of institutional interventions.

As shown in Figure 4, the most recurrent psychosocial variables were motivation and academic goals (73.7%), followed by gender and age (52.6%), and family or cultural responsibilities (31.6%). Mental health, emotional well-being, and vocational alignment appeared in 26.3% of the studies, while satisfaction and social integration were reported in 21.1%. Beyond their individual prevalence, these variables interact in ways that magnify their influence on dropout risk. The synthesis reveals that reduced motivation and unclear academic goals often coexist with emotional distress or a weak sense of belonging, creating a cumulative vulnerability that undermines students’ capacity to cope with academic demands. Similarly, gender- and age-related differences intersect with family or cultural responsibilities, shaping the extent to which students can engage consistently with their studies. These interaction patterns indicate that psychosocial determinants function not merely as isolated predictors but as interconnected mechanisms that mediate the impact of academic and economic pressures on student persistence.

Figure 4

Figure 4. Most cited psychosocial variables. Prepared by the authors based on Scopus and Web of Science.

The following Table 4 presents the percentage distribution of the studies included in this review, broken down by geographic region. This information highlights the relative representation of different global contexts in the research on university dropout, providing a clear view of the geographical coverage of the studies analyzed.

Table 4

Table 4. Percentage distribution of studies by region, method, and effect direction.

The following Table 5 outlines the research methods most commonly used in the studies included in this review, along with the reported direction of the effect in each study. This information is essential for understanding the predominant methodologies and how the approaches used may influence the interpretation of the findings on university dropout.

Table 5

Table 5. Study methods and effect direction.

As illustrated in Figure 5, the institutional approaches most frequently reported were risk prediction models and learning-analytics tools (52.6%), followed by academic support, retention strategies, and peer/online adaptation programs (36.8%); early warning systems, curricular flexibility, and socio-academic integration mechanisms appeared in 21.1% of the studies. Beyond these frequencies, the evidence shows that institutional measures function primarily as moderators of risk rather than standalone solutions. Predictive models and learning analytics are useful for early identification of at-risk students, but their effectiveness depends on the existence of timely, integrated responses (e.g., tutoring, targeted financial aid, mental-health referrals).

Figure 5

Figure 5. Most reported institutional approaches. Prepared by the authors based on Scopus and Web of Science.

Early warning systems typically trigger interventions that cut across academic, economic and psychosocial domains: for instance, an alert prompted by falling attendance may lead to academic coaching, counseling for stress related to precarious work, or temporary fee relief thereby interrupting negative causal chains. Similarly, curricular flexibility and socio-academic integration reduce the friction between study and life obligations, mitigating the adverse impact of employment or family responsibilities on academic outcomes. The reviewed studies also warn about limits and trade-offs: predictive tools can reinforce bias if models omit structural variables (e.g., socioeconomic status), and isolated academic interventions have limited effect unless paired with financial and psychosocial supports. Overall, the synthesis indicates that the most effective institutional strategies are those that couple robust identification systems with multi-dimensional, rapid-response supports tailored to students’ combined academic, economic and psychosocial needs.

As illustrated in Figure 6, the most frequently applied theoretical frameworks were Tinto’s Integration Model (36.8%), Bean’s Model of Student Attrition (26.3%), and Astin’s Engagement Model (21.1%). Complementary perspectives psychosocial, motivational, organizational, and socioeconomic appeared in 15.8% of the studies. Beyond the distribution of frameworks, the analysis reveals that these models shape not only the way dropout determinants are conceptualized but also how interactions among factors are interpreted. Tinto’s model highlights the bidirectional relationship between academic and social integration, providing a foundation for understanding how lack of belonging intensifies academic vulnerability. Bean’s model, in contrast, foregrounds attitudinal and behavioral components, clarifying how psychosocial variables such as satisfaction, intention to persist, or perceived institutional support mediate the effect of contextual pressures. Astin’s engagement theory underscores the centrality of student involvement, which helps explain why economic stress or mental health issues rapidly translate into academic disengagement. Studies employing complementary frameworks tend to integrate these dimensions, showing that socioeconomic inequality, motivational deficits, and organizational rigidity can converge to amplify dropout risk. Taken together, the prevalence and interplay of these theoretical models reinforce the idea that university dropout is best understood as a systemic and interactional process, rather than the consequence of isolated determinants.

Figure 6

Figure 6. Most commonly used theoretical models. Prepared by the authors based on Scopus and Web of Science.

Theoretical models play a fundamental role in explaining university dropout, as they provide structured frameworks for understanding how academic, social, and institutional variables interact to influence students’ decisions to persist or withdraw. Among the models identified, Tinto’s (1975, 1993) Integration Model highlights the importance of academic and social integration within the university environment; Bean’s (1985) Model of Student Attrition emphasizes the role of attitudes, satisfaction, and external factors in shaping persistence; and Astin’s (1984) Theory of Student Involvement focuses on the extent of students’ engagement as a predictor of retention. Complementary frameworks, including psychosocial and motivational approaches, expand these perspectives by incorporating emotional well-being, institutional belonging, and structural inequalities. Together, these models provide a multidimensional understanding that supports the interpretation of findings in this review.

The results were organized according to the five core dimensions of the study: academic factors, economic conditions, psychosocial variables, institutional approaches, and theoretical models. This configuration facilitated the identification of shared patterns across literature from diverse disciplinary backgrounds. The classification reveals a variety of variables and strategies addressed, with particular emphasis on those that influence student retention. The analysis further identifies a tendency toward the combined use of quantitative and qualitative methods, which is argued to facilitate a more comprehensive understanding of the university dropout phenomenon.

To move beyond descriptive frequencies and better capture the explanatory power of each determinant, a co-occurrence analysis was conducted across the academic, economic, and psychosocial dimensions. Rather than functioning as isolated predictors, the variables consistently formed interdependent clusters that provide a more robust explanation of dropout behavior. The strongest cluster identified was the academic–psychosocial cluster, where academic performance, study habits and cognitive engagement appeared tightly linked to motivation, emotional well-being and self-efficacy. The high co-occurrence between achievement and motivation indicates that academic difficulties often emerge as both a cause and consequence of psychosocial strain, creating a mutually reinforcing cycle that intensifies vulnerability to dropout.

A second cluster, the economic–psychosocial strain cluster, was characterized by the recurrent intersection of financial hardship with emotional distress and reduced institutional engagement. This pattern suggests that socioeconomic pressure does not affect dropout decisions through a single pathway, but through a constellation of effects such as stress, diminished concentration, and weakened academic involvement that accumulate over time. Finally, an academic economic cluster was identified, reflecting the interplay between limited financial resources and declining academic engagement, often mediated by increased work hours or time scarcity.

These relational clusters are reflected in the joint frequency coefficients, where academic–psychosocial co-occurrences were strongest (r = 0.68), followed by economic psychosocial (r = 0.61) and academic economic (r = 0.54) interactions. Collectively, these patterns demonstrate that university dropout emerges from systems of interrelated determinants, rather than discrete variables, reinforcing the need for predictive frameworks that incorporate cognitive, emotional, and socioeconomic integration within institutional contexts.

4 Discussion

This section presents an analysis of the findings on university dropout, articulated with theoretical models and empirical evidence. A comparison with previous studies is included to identify both similarities and contributions. The present study proposes a conceptual framework that summarizes the identified factors. The theoretical, policy and practical implications that have been derived from the results are presented in this paper. The study’s methodological limitations are also highlighted. Finally, future research lines are proposed to deepen our understanding of the phenomenon and strengthen student retention strategies.

The theoretical frameworks guiding this review have been expanded to provide a deeper understanding of the variances between studies in different contexts. Tinto’s (1975, 1993) Integration Model emphasizes the importance of academic and social integration for student retention, a concept that is more readily applicable in Western educational contexts where institutional support systems are often more robust. In contrast, Bean’s (1985) Model of Student Attrition considers external factors like family and financial circumstances, which may be more significant in low-resource settings, such as in Latin America or sub-Saharan Africa. Astin’s (1984) Theory of Student Involvement highlights the role of student engagement, a model that may vary in its effectiveness based on institutional culture, particularly in regions where student participation is either less structured or culturally less emphasized.

Furthermore, the studies reviewed demonstrate that the applicability and effectiveness of these models vary depending on the socio-economic and cultural context. For example, in many European and North American studies, factors like financial aid and institutional resources play a prominent role, while in regions such as Asia and Africa, socio-cultural factors like family responsibilities and local educational systems can often outweigh institutional influences. These differences underscore the need for a more context-sensitive approach when applying these theoretical models globally.”

It is essential to clarify that the frequency with which a factor appears in the reviewed studies does not necessarily reflect its statistical or causal significance. The fact that academic performance or motivation are frequently reported determinants should not be interpreted as evidence of higher predictive power, but rather as indicators of research focus within the literature. Therefore, the interpretation of the results in this study prioritizes conceptual relevance and relational consistency among variables, rather than mere occurrence counts. This distinction prevents overestimating certain factors and reinforces the need to evaluate their contextual weight and theoretical integration when explaining dropout behavior.

4.1 Analysis of results

Firstly, the results indicate a significant correlation between academic factors and student performance as determinants of university dropout. The reviewed studies emphasize grades, GPA, assessment methods and overall performance as key indicators, thereby supporting their usefulness as predictive variables. Furthermore, institutional and pedagogical variables such as teaching quality and academic overload are identified, suggesting a multifactorial approach. This pattern is consistent with the findings of Parra-Sánchez et al. (2023), who emphasized these factors in predictive models based on artificial intelligence, and with the conclusions of Nikolaidis et al. (2022), who associated them with academic progress and student retention.

In consideration of economic conditions, a direct influence on the decision to withdraw is observed. The research under review indicates that financial hardship, the necessity of maintaining employment during the period of study, the paucity of available scholarships, and the restricted availability of financial support are all factors of relevance. Concomitantly, structural variables such as low family income, incongruity between study and employment, parental socioeconomic status, and geographic location have been identified as contributing factors. These conclusions are in alignment with those reported by Yaghi and Alabed (2025) and Santos-Villalba et al. (2023), who recognize the economic dimension as a central cause of student attrition.

Moreover, research analysis indicates that psychosocial variables also play a significant role in university dropout. The factors most frequently cited include motivation, academic aspirations, gender, age, and cultural background. Furthermore, family responsibilities, mental health, living conditions, emotional well-being, and vocational misalignment have also been identified as contributing factors. While not as pronounced, levels of satisfaction with the educational experience and social integration have also been documented. These findings are consistent with those reported in the studies by Morelli et al. (2021) and Gutierrez-Pachas et al. (2023), which emphasize the complexity and interdependence of the psychosocial component.

With regard to institutional responses, the extant literature reflects a wide range of approaches to the issue of dropout. Risk prediction models and learning analytics tools are of particular note, reflecting an orientation toward the use of data to anticipate dropout. In addition to these measures, strategies have been implemented with the objective of providing comprehensive support, ensuring academic retention, and facilitating adaptation to virtual environments. Documentation of initiatives such as early warning systems, peer support networks, and socio-academic integration mechanisms is also provided. These measures align with those outlined by Núñez-Naranjo (2024) and Sacală et al. (2021), who underscore the significance of institutional interventions employing a holistic approach.

The discussion of the results has been expanded to include a more explicit comparison of the findings with the theoretical models of Tinto, Bean, and Astin. Tinto’s (1975, 1993) Integration Model emphasizes the importance of academic and social integration for student retention, which aligns with our findings on the critical role of motivation and social belonging in preventing university dropout. Similarly, Bean’s (1985) Model of Student Attrition suggests that external factors, such as financial hardship and family obligations, significantly affect students’ decisions to persist. Our findings support this model by showing the strong correlation between economic hardship and dropout risk. Astin’s (1984) Theory of Student Involvement focuses on student engagement as a key predictor of retention. This model is reinforced by our results, which indicate that students with higher levels of involvement and engagement in academic and social activities are less likely to drop out.

Finally, the recurrent utilization of established theoretical frameworks to elucidate the phenomenon of attrition is manifest. The predominant approaches of Tinto, Bean, and Astin emphasize academic integration, institutional satisfaction, and the level of student engagement. Concurrently, a range of frameworks are incorporated, including psychosocial, motivational, organizational, and socioeconomic frameworks. This suggests a tendency to integrate multiple perspectives in order to comprehend a complex phenomenon. These findings are consistent with those reported by Fernández-Martín et al. (2019) and Arias et al. (2024), who emphasize the significance of combined approaches for the analysis and prevention of university dropouts.

A cross-dimensional interpretation of the findings reveals that academic, economic, and psychosocial factors do not operate independently but form causal chains that cumulatively intensify dropout risk. One of the most recurrent patterns identified across the reviewed studies begins with financial pressure, which compels students to increase their working hours in order to meet tuition or living expenses. This increase in employment demands directly reduces the time available for study, leading to lower attendance and diminished engagement with academic activities. Reduced academic involvement, in turn, weakens performance, generates feelings of academic inefficacy, and heightens emotional strain. Once these conditions converge, students become more vulnerable to detachment from the institution, which significantly elevates the likelihood of attrition. This sequence financial stress → work intensity → absenteeism → academic disengagement → dropout appears consistently across the literature and provides a clear illustration of how structural pressures translate into academic outcomes through psychosocial pathways.

Moreover, the interaction between psychosocial and academic variables frequently operates as a feedback loop, reinforcing vulnerability over time. Reduced motivation or emotional well-being often triggered by financial instability, family responsibilities, or precarious living conditions negatively affects students’ capacity to concentrate, participate in class, or complete coursework. These academic setbacks then exacerbate psychological distress, further eroding commitment to academic goals. Such reciprocal dynamics are particularly evident among students who lack social or institutional support networks. When institutional mechanisms such as early warning systems, tutoring, financial aid, or mentoring are absent or insufficient, these negative cycles remain unmitigated, increasing the probability of dropout. Conversely, the literature shows that timely institutional interventions can interrupt these chains by addressing both the source of the pressure (e.g., financial hardship or excessive workload) and its manifestation (e.g., declining academic performance or emotional fatigue), thereby improving persistence outcomes.

4.2 Comparison of results with other studies

The study’s findings are consistent with the extant literature on the theoretical models of Tinto, Bean, and Astin, which remain central to explaining university dropout rates through academic integration, institutional satisfaction, and student engagement. This theoretical convergence has been observed in studies such as those by Rožman et al. (2025), which highlight the influence of motivation, mental health, and work-life balance on academic performance, thus confirming the validity of psychosocial variables present in these models. Moreover, research such as that by Al Maqrashi (2025) serves to expand the analysis of the phenomenon by integrating a structural perspective that links dropout rates with social inequalities and psychoemotional effects.

This perspective employs a contextual and critical approach to institutional limitations. While the study acknowledges the significance of economic and psychosocial factors, the Omani approach highlights structural barriers that are often overlooked in systems with extensive higher education coverage. A methodological difference is identified in the study by Espinoza and Carretero (2025), which employs predictive analytics with historical data to identify risk profiles. In contrast to the present study, which employs a thematic systematization based on a literature review, the use of data mining and neural networks introduces an applied quantitative strategy that strengthens early detection of dropouts and reflects the growing use of technologies in studies on retention.

With regard to the emergence of variables, the study by Banda et al. (2025) focuses on secondary education in Malawi and incorporates the dimensions of autonomy, competence, and belonging, as outlined in Self-Determination Theory. Despite the focus on a distinct educational level, the outcomes underscore the significance of integrating motivational and social justice components, which are in alignment with the psychosocial approaches addressed in this research. These findings demonstrate the persistence of certain factors throughout the educational process. Moreover, the review by Barragán-Moreno et al. (2025) provides a comprehensive view of school dropout at various educational levels. The integration of individual, socioeconomic, and institutional factors within the analytical framework employed in this study is a salient aspect of their research.

Nevertheless, this study highlights the gaps that persist in the existing body of literature. Pertaining to long-term interventions, policy frameworks, and excluded populations. While these aspects are not central to the present research, their discussion highlights the necessity for broader and more context-sensitive approaches. The study provides a comprehensive and systematic categorization of the factors that influence university dropout, encompassing academic, economic, psychosocial, and institutional dimensions. This categorization draws upon diverse disciplinary perspectives, offering a multifaceted understanding of the complex mechanisms that shape student retention in higher education. In contrast to research that is focused on a single model, country, or method, the comparative analysis developed herein reinforces the value of multiple frameworks for addressing a complex phenomenon. The primary contribution of this study is to provide a systemic perspective that is instrumental in the design of interventions and the formulation of future research directions. These research directions should integrate quantitative, qualitative, and technological approaches.

4.3 Proposed conceptual framework

As illustrated in Figure 7, the conceptual model integrates the main determinants of university dropout within a dynamic interaction scheme. The decision to persist or withdraw is positioned at the center, surrounded by academic, economic, and psychosocial dimensions that shape students’ trajectories. Institutional intervention operates as a moderating layer that can either buffer or intensify the influence of these determinants. Each component aligns directly with the theoretical principles of Tinto, Bean and Astin, providing a coherent foundation for the model’s structure.

Figure 7

Figure 7. Conceptual model of university dropout. Prepared by the authors.

The directional flow represented in Figure 7 clarifies how the components of the model interact causally. Economic factors act as upstream conditions that can trigger psychosocial strain, which in turn affects academic engagement and performance. Psychosocial variables operate as mediators, translating external pressures into changes in motivation, well-being and institutional attachment. Academic outcomes represent downstream effects, reflecting the accumulated impact of economic and psychosocial influences. Institutional interventions function as moderating mechanisms that can weaken, interrupt or reverse these pathways. This directional structure illustrates a logical progression in which external pressures initiate vulnerabilities, psychosocial processes transmit their effects, and academic performance reflects the final stage influencing the decision to persist or withdraw.

The academic component corresponds primarily to Tinto’s concept of academic integration, which asserts that students’ academic performance, engagement with coursework and interaction with faculty shape their institutional commitment. In the model, variables such as performance, study habits and cognitive engagement operationalize this construct. This dimension also reflects Astin’s Theory of Student Involvement, as sustained effort and participation in learning activities represent the behavioral investment central to persistence.

The psychosocial component aligns with Tinto’s social integration and Astin’s emphasis on psychological involvement. Motivation, sense of belonging, emotional well-being and self-efficacy represent the internal processes through which students attach meaning to their academic experiences. These factors embody both theorists’ propositions that students remain enrolled when they feel connected to their social and academic environment and invest psychological energy in their studies.

The economic component is closely aligned with Bean’s Model of Student Attrition, which emphasizes the role of external and structural influences—such as financial pressure, employment obligations and socioeconomic background—in shaping students’ satisfaction and persistence. The model reflects this logic by positioning economic strain as a factor that permeates academic and psychosocial domains. The recurrent interactions observed between financial hardship, emotional distress and reduced academic engagement in the reviewed studies further substantiate Bean’s theoretical pathway.

Finally, the institutional component synthesizes the intervention mechanisms implicit in all three frameworks. Consistent with Astin, actions such as tutoring, early warning systems, mentoring and financial aid function as environmental stimuli that foster involvement. According to Tinto, these interventions reinforce academic and social integration, while Bean’s model highlights their capacity to reduce the negative impact of external barriers. Thus, institutional support acts as a structural moderator capable of interrupting adverse causal chains and promoting student persistence.

The conceptual model presented in Figure 7 illustrates the interactions between academic, economic, and psychosocial factors in the university dropout process. It is important to note that the economic constraints, such as financial hardship and lack of resources, are often mediated by psychosocial factors, such as stress, mental health issues, and social disintegration. For example, students facing financial difficulties may experience increased emotional distress, which in turn affects their academic performance and engagement, leading to a higher risk of dropout. Similarly, academic and psychosocial factors are interconnected, as low academic performance can contribute to a sense of failure and decreased motivation, further exacerbating psychosocial stress and increasing the likelihood of leaving higher education. These interactions underscore the multidimensional nature of dropout, where no single factor acts in isolation.

4.4 Implications

Theoretically, the findings corroborate the validity of the seminal models proposed by Tinto, Bean, and Astin. Moreover, they underscore the necessity to elucidate these models through the lens of contemporary motivational and structural frameworks. A convergence of factors, including academic integration, institutional commitment, and student engagement, has been identified, thereby validating their explanatory capacity. However, it is important to note that these benefits are not uniform, and disparities emerge in the presence of unequal socioeconomic conditions and the psychoemotional experiences that students encounter. The incorporation of variables such as mental health, motivation, and sense of belonging facilitates a more profound comprehension of student attrition. This standpoint contributes to the field by proposing a conceptual framework that articulates structural dimensions, causal relationships, and modulating mechanisms, thereby broadening the analysis of retention and academic success. This approach has the potential to serve as a foundational framework for future research endeavors that integrate mixed methodologies, longitudinal analysis, and contextual considerations, particularly within educational systems characterized by persistent structural gaps.

From a policy perspective, the findings demonstrate the urgency for governments, ministries, and regulatory agencies to adopt more comprehensive and preventative approaches to student dropout. It is imperative that public policies shift their focus from access and graduation indicators to incorporate measures that ensure the presence of realistic conditions for continued enrolment in inclusive and sustainable environments. It is recommended that financing schemes be strengthened to cover not only tuition but also indirect costs such as transportation, food, and materials, with an emphasis on vulnerable populations. It is also proposed that national retention policies be designed with academic support strategies, psychosocial support, and institutional evaluation, articulated with monitoring and follow-up systems. In addition, regulatory frameworks should promote flexible educational models that recognize diverse trajectories, reduce access barriers, and guarantee quality. Finally, the need to implement evidence-based educational governance systems, supported by disaggregated data and predictive tools that enable targeted interventions, is highlighted.

From a pragmatic standpoint, the findings provide unambiguous directives for higher education institutions. The adoption of early warning systems that integrate information on academic performance, attendance, participation, and psychosocial variables is considered a priority, facilitating the timely identification of at-risk students and the activation of support protocols. The proposal is to design differentiated interventions based on each student’s needs, including tutoring, vocational guidance, psychological support, and targeted financial subsidies. Teacher training is recognized as a strategic pillar, as it must train professors in the use of active methodologies, inclusive approaches, and early detection of signs of dropout. It is also recommended that there be a move toward the implementation of flexible curricular models that recognize work, enable alternative pathways and value prior learning. Finally, the importance of consolidating an institutional culture of well-being is highlighted, with services that promote support, participation, and a sense of student belonging. The findings of this study demonstrate that the issue of university dropout should not be approached as a standalone or unidimensional phenomenon. In order to address this issue, integrated responses are required from the fields of theory, public policy and institutional action. The coordination of these levels is imperative to ensure sustainable, equitable, and transformative educational trajectories.

4.5 Limitations

The present study is subject to certain limitations, which are associated with its methodological approach, which is based on a systematic review. The selection of sources was restricted by the coverage of specific academic databases, which may have excluded unindexed research or research published in other languages. Despite the rigorous application of inclusion and exclusion criteria, a risk of bias persists, stemming from the predominance of quantitative studies or studies originating from high-impact journals. This phenomenon serves to diminish methodological and theoretical diversity. It is acknowledged that some contextual or cultural factors may not have been documented uniformly, thus potentially affecting the comparability of findings across studies. The applicability of This study presents several limitations that must be acknowledged. Although the review aims to offer a broad perspective on university dropout, the geographical distribution of the included studies is uneven, with a marked concentration in Latin America and Southern Europe. This regional bias limits the generalizability of the findings to other contexts, particularly in Asia, Africa, and regions with distinct socioeconomic, cultural, or institutional configurations. Additionally, the reliance on specific academic databases and the inclusion of only peer-reviewed studies in English and Spanish may have excluded relevant research published in other languages or in local, non-indexed outlets. The predominance of quantitative designs and publications from high-impact journals may also reduce methodological and theoretical diversity, introducing potential publication bias. Finally, as the analysis is based exclusively on secondary data, the review is constrained by the quality, scope, and contextual specificity of the original studies. Despite these limitations, the findings provide valuable insights, and future research should expand the geographical scope and employ comparative and mixed-method approaches to strengthen the global understanding of dropout dynamics.

A further limitation concerns the underrepresentation of qualitative evidence in the reviewed studies. The predominance of quantitative designs restricts the depth and contextual richness with which students’ lived experiences, perceptions, and decision-making processes can be understood. Qualitative approaches such as interviews, focus groups, and ethnographic analyses are essential for capturing the subjective dimensions of dropout, including emotional trajectories, identity conflicts, and informal institutional dynamics that are not easily measured through numerical indicators. Future research would benefit from incorporating more qualitative or mixed-method studies to provide a more comprehensive and nuanced understanding of the factors that shape students’ persistence and withdrawal.

4.6 Lines of future research

The results obtained, the implications discussed, and the limitations identified have enabled the clear delineation of opportunities to guide future research on university dropout. It is imperative to undertake empirical studies that explore the intricate interplay between academic, economic, and psychosocial factors from a holistic standpoint. In order to achieve a more accurate capture of student trajectories, dropout mechanisms, and related experiences, it is recommended that future studies incorporate mixed methodological designs that combine quantitative and qualitative data. The development and validation of predictive models that integrate structural, institutional, and individual variables is recommended. It is imperative that these models encompass a comprehensive set of indicators, including but not limited to academic risk, socioeconomic conditions, mental health status, perception of the educational environment, and level of student engagement.

The construction of robust prediction systems will facilitate the early detection of at-risk students and the efficient allocation of resources. The validation of these models in various institutional contexts is imperative to ensure their reliability, adaptability, and operational usefulness. Furthermore, the promotion of longitudinal studies is of paramount importance, as these studies facilitate the observation of the evolution of retention over time. This methodological approach is conducive to the identification of sustained patterns, cumulative effects, and critical moments in the dropout process. Concurrently, there is a necessity for comparative research across countries or regions to analyze how educational policies, social conditions, and institutional characteristics impact rates of both dropout and retention. These comparisons provide a foundation for adapting interventions to specific contexts and fostering inter-institutional and regional learning.

Future research should focus on testing the proposed conceptual model through longitudinal studies to observe the long-term impact of academic, economic, and psychosocial factors on university dropout and retention. Longitudinal designs will allow researchers to track the evolution of these factors over time and provide a deeper understanding of causal relationships. Additionally, studies using mixed methods could enhance the model’s validation by combining qualitative insights into students’ experiences with quantitative data on academic performance, financial status, and psychosocial well-being. Such studies would provide a more holistic view of the dropout process, capturing both the statistical patterns and the individual narratives that contribute to student attrition.

Exploration of innovative approaches to institutional intervention strategies and public policy design is also pivotal. It is recommended that future research focus on the analysis of collaborative governance models, inclusive financing schemes and evidence-based student support programs. It is also relevant to examine the effectiveness of measures such as curricular flexibility, the use of learning support technologies, and pedagogical practices focused on student well-being. The generation of knowledge that establishes a link between the diagnosis of dropout and concrete, viable, and contextualized solutions is an urgent requirement. These future research lines will contribute to a more profound understanding of the phenomenon of university dropout. Consequently, these findings will facilitate the development of more effective strategies to enhance equity, retention, and academic achievement in higher education.

5 Conclusion

In order to comprehend the phenomenon of university dropout, it is necessary to adopt an approach that transcends the limitations of a siloed list of factors. The findings of this study demonstrate that the phenomenon of students withdrawing from higher education does not result from a single or linear set of causes. The process is intricate, arising from the interaction between individual trajectories, socioeconomic structures, and institutional practices. From this standpoint, the student is not merely a passive recipient, but rather an actor who makes decisions in conditions characterized by academic pressures, economic limitations, and psychosocial challenges. A review of the extant literature reveals a diversity of theoretical and methodological approaches. The persistence of conventional models centered on performance or economic capacity is evident, yet perspectives encompassing contextual, cultural, and technological variables are also emerging.

This variety is indicative of both the richness and limitations of the field. In this context, it is necessary to adopt an integrative approach that allows us to understand the phenomenon in its complexity, without ignoring local specificities. The phenomenon of university dropout is not solely an academic concern. The issue is further complicated by ethical and political challenges concerning equity, inclusion, and the right to education. The analysis of this issue necessitates an examination of the factors that impede its progression and the proposal of concrete measures to ensure access, transition, and the effective completion of higher education.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

AV-A: Writing – original draft, Writing – review & editing. JVA: Writing – original draft, Writing – review & editing. JV: Writing – original draft, Writing – review & editing. SC-A: Writing – original draft, Writing – review & editing. JP-V: Writing – original draft, Writing – review & editing. HU: Writing – original draft, Writing – review & editing.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Al Maqrashi, R. H. (2025). On the brink: academic probation and the risk of dropout in an Omani higher education context. Open Rev. 10, 84–118. doi: 10.47967/TOR10FH2U

Crossref Full Text | Google Scholar

Apumayta, R. Q., Cayllahua, J. C., Pari, A. C., Choque, V. I., Valverde, J. C. C., and Ataypoma, D. H. (2024). University dropout: a systematic review of the main determinant factors (2020-2024). F1000Res 13:942. doi: 10.12688/f1000research.154263.2,

PubMed Abstract | Crossref Full Text | Google Scholar

Arias, A., Linares-Vásquez, M., and Héndez-Puerto, N. R. (2024). Undergraduate dropout in Colombia: a systematic literature review of causes and solutions. J. Latinos Educ. 23, 612–627. doi: 10.1080/15348431.2023.2171042

Crossref Full Text | Google Scholar

Astin, H. S. (1984). The meaning of work in women’s lives a sociopsychological model of career choice and work behavior. The Counseling Psychologist. 12, 117–126.

Google Scholar

Asubiaro, T., Onaolapo, S., and Mills, D. (2024). Regional disparities in web of science and Scopus journal coverage. Scientometrics 129, 1469–1491. doi: 10.1007/s11192-024-04948-x.

Crossref Full Text | Google Scholar

Banda, L. O. L., Banda, J. T., Mpaso, C. N., Xuefu, X., Mwaene, E., and Banda, C. V. (2025). Girls’ education still at risk? Interplay of SDT and intersectionality in dropout analysis. Gender Sustain. Global South 10:5. doi: 10.1515/gsgs-2024-0005

Crossref Full Text | Google Scholar

Barragán-Moreno, S. P., and González-Támara, L. (2024). Complexities of student dropout in higher education: a multidimensional analysis. Front. Educ. 9:1461650. doi: 10.3389/feduc.2024.1461650

Crossref Full Text | Google Scholar

Barragán-Moreno, S. P., Guzmán-Rincón, A., Calderón-Carmona, G. P., González-Támara, L., and Lozano-Galindo, O. L. (2025). Identifying key variables of student dropout in preschool, primary, secondary, and high school education: an umbrella review approach. Eur. J. Educ. Res. 14, 585–600. doi: 10.12973/eu-jer.14.2.585

Crossref Full Text | Google Scholar

Bean, J. P. (1985). Interaction effects based on class level in an explanatory model of college student dropout syndrome. American Educational Research Journal. 22, 35–64.

Google Scholar

Buenaño, E., Beletanga, M. J., and Mancheno, M. (2024). What factors are relevant to understanding dropout? Analysis at a co-financed University in Ecuador and policy implications, using survival cox models. J. Latinos Educ. 23, 1400–1415. doi: 10.1080/15348431.2023.2271570

Crossref Full Text | Google Scholar

Constate-Amores, A., Martínez, E. F., Asencio, E. N., and Fernández-Mellizo, M. (2021). Factores asociados al abandono universitario. Educ. XX1 24, 17–44. doi: 10.5944/educXX1.26889

Crossref Full Text | Google Scholar

Espinal, R. A. M., and Fabian, L. (2021). Determinantes de la deserción universitaria: un estudio de caso en la República Dominicana. Sapienza 2, 255–268. doi: 10.51798/sijis.v2i1.76

Crossref Full Text | Google Scholar

Espinoza, C., and Carretero, J. (2025). Profiling and archetyping of higher education applicants using intelligent data analysis techniques. IEEE Rev. Iberoam. Tecnol. Aprendiz. 20:369. doi: 10.1109/RITA.2025.3583369

Crossref Full Text | Google Scholar

Fernandes, J.. 2024. Higher education dropout and economic consequences: a literature review. EDULEARN24 proceedings, 1105–1112. London ICEEL

Google Scholar

Fernández-Martín, T., Solís-Salazar, M., Hernández-Jiménez, M. T., and Moreira-Mora, T. E. (2019). A multinomial and predictive analysis of factors associated with university dropout. Rev. Electrón. Educare 23, 73–97. doi: 10.15359/ree.23-1.5

Crossref Full Text | Google Scholar

Gallegos, J. A., Campos, N. A., Canales, K. A., and González, E. N. (2018). Factores determinantes en la deserción universitaria. Caso Facultad de Ciencias Económicas y Administrativas de la Universidad Católica de la Santísima Concepción (Chile). Form. Univ. 11, 11–18. doi: 10.4067/S0718-50062018000300011

Crossref Full Text | Google Scholar

Gutierrez-Pachas, D. A., Garcia-Zanabria, G., Cuadros-Vargas, E., Camara-Chavez, G., and Gomez-Nieto, E. (2023). Supporting decision-making process on higher education dropout by analyzing academic, socioeconomic, and equity factors through machine learning and survival analysis methods in the Latin American context. Educ. Sci. 13:154. doi: 10.3390/educsci13020154

Crossref Full Text | Google Scholar

Kalamaras, D., Maska, L., and Nasika, F. (2025). A proposed MIMIC structural equation model for assessing factors affecting time to degree—the case of the Greek tertiary system. Educ. Sci. 15:187. doi: 10.3390/educsci15020187

Crossref Full Text | Google Scholar

Kalet, A., Ellaway, R. H., Song, H. S., Nick, M., Sarpel, U., Hopkins, M. A., et al. (2013). Factors influencing medical student attrition and their implications in a large multi-center randomized education trial. Adv. Health Sci. Educ. 18, 439–450. doi: 10.1007/s10459-012-9382-z,

PubMed Abstract | Crossref Full Text | Google Scholar

Lamichhane, B. D., Giri, O. P., Singh, D. B., Kafle, B. R., and Lama, P. B. (2025). Determinant factors of university students' dropouts: a qualitative analysis. Int. J. Econ. Manag. Syst. 12, 35–45. doi: 10.14445/23939125/IJEMS-V12I4P105

Crossref Full Text | Google Scholar

Mata, L., Cirtita-Buzoianu, C., Amalancei, B. M., Cojocariu, V. M., and Mares, G. (2024). Students’ perception of the determining factors of university dropout. Rev. Romaneasca Educ. Multidimens. 16, 127–153. doi: 10.18662/rrem/16.4/910

Crossref Full Text | Google Scholar

Morelli, M., Chirumbolo, A., Baiocco, R., and Elena, C. (2021). Academic failure: individual, organizational, and social factors. Psicol. Educ. 27, 167–175. doi: 10.5093/psed2021a8

Crossref Full Text | Google Scholar

Muro, E. D. A., Álvarez, L. A. S., Rodriguez, V. H. P., Lucana, F. R. V., Rojas, L. M. H., Benavides, A. M. V., et al. (2024). Fostering equity in rural education: a literature review on student dropout and retention strategies. Rev. Gestão Soc. Ambient. 18, 1–18. doi: 10.24857/rgsa.v18n1-083

Crossref Full Text | Google Scholar

Nikolaidis, P., Ismail, M., Shuib, L., Khan, S., and Dhiman, G. (2022). Predicting student attrition in higher education through the determinants of learning progress: a structural equation modelling approach. Sustainability 14:13584. doi: 10.3390/su142013584

Crossref Full Text | Google Scholar

Núñez-Hernández, C., and Buele, J. (2023). Factors influencing university dropout in distance learning: a case study. J. High. Educ. Theory Pract. 23:6379. doi: 10.33423/jhetp.v23i14.6379

Crossref Full Text | Google Scholar

Núñez-Naranjo, A. F. (2024). Analysis of the determinant factors in university dropout: a case study of Ecuador. Front. Educ. 9:1444534. doi: 10.3389/feduc.2024.1444534

Crossref Full Text | Google Scholar

Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., et al. (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372:n71. doi: 10.1136/bmj.n71

Crossref Full Text | Google Scholar

Parra-Sánchez, J. S., Torres Pardo, I. D., and de Martínez Merino, C. Y. (2023). Explanatory factors of university dropout explored through artificial intelligence. Rev. Electrón. Invest. Educ. 25:4455. doi: 10.24320/redie.2023.25.e18.4455

Crossref Full Text | Google Scholar

Rožman, M., Vrečko, I., and Tominc, P. (2025). Psychological factors impacting academic performance among business studies’ students. Educ. Sci. 15:121. doi: 10.3390/educsci15020121

Crossref Full Text | Google Scholar

Sacală, M. D., Pătărlăgeanu, S. R., Popescu, M. F., and Constantin, M. (2021). Econometric research of the mix of factors influencing first-year students’dropout decision at the faculty of Agri-food and environmental economics. Econ. Comput. Econ. Cybern. Stud. Res. 55:13. doi: 10.24818/18423264/55.3.21.13

Crossref Full Text | Google Scholar

Santos-Villalba, M. J., del Alcalá Olmo Fernández, M. J., Montenegro Rueda, M., and Fernández Cerero, J. (2023). Incident factors in Andalusian university dropout: a qualitative approach from the perspective of higher education students. Front. Educ. 7:1083773. doi: 10.3389/feduc.2022.1083773

Crossref Full Text | Google Scholar

Thompson, F., Hodge, G., Edge, D., Howes, S., Jamison, C., Fisher, M., et al. (2025). Understanding the risk factors for student attrition across pre-registration nursing and midwifery programmes in a United Kingdom university: a sequential explanatory mixed methods study. Nurse Educ. Today 145:106503. doi: 10.1016/j.nedt.2024.106503,

PubMed Abstract | Crossref Full Text | Google Scholar

Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent research. J. High. Educ. 45, 89–125. doi: 10.3102/00346543045001089

Crossref Full Text | Google Scholar

Tinto, V. (1993). Reflexiones sobre el abandono de los estudios superiores. Perfiles Educ. 62, 56–63.

Google Scholar

UNESCO (2012). International standard classification of education: ISCED 2011. Montreal: UNESCO Institute for Statistics.

Google Scholar

Vera, A. A., Martinez, J., and Espinoza, A. P. (2025). “Determinants of university dropout and educational quality improvement” in International conference on computational science and its applications. ed. O. Gervasi (Cham: Springer Nature Switzerland), 337–354.

Google Scholar

Yaghi, A., and Alabed, N. (2025). Factors affecting university dropout: comparison of STEM and public affairs and management students. Int. J. Public Adm. 12, 1–15. doi: 10.1080/01900692.2025.2476676,

PubMed Abstract | Crossref Full Text | Google Scholar