University tutoring is an academic support process aimed at guiding and assisting students during one or more stages of their university education. It is carried out by a tutor whose purpose is to facilitate adaptation to the university environment, optimize academic performance, and promote the overall development of students at academic risk (González et al., 2024; Guffante et al., 2022). This support is based on the interaction between the tutor and the student, establishing a relationship rooted in trust, effective communication, and mutual commitment to achieving formative and personal goals (Elgegren, 2022; Pérez et al., 2024; Reséndiz and Zepeda, 2021).

Success in higher education has evolved; it is no longer limited to the acquisition of knowledge but extends to the development of transversal competencies and student well-being. In this context, university tutoring has become an essential pedagogical strategy to guide students in their academic and personal journey, especially those in adverse conditions or at academic risk (Esteban and Caro, 2023; Garay et al., 2024; Vita et al., 2021).

In the Latin American context and the Peruvian university system, the implementation of tutoring programs faces several challenges, ranging from students’ socioeconomic conditions to the management and resources available within universities. Often, these programs lack systematic evaluation of their impact and an analysis of the perceptions of key stakeholders: students, tutors, and administrators. This disconnect between design and practice limits the effectiveness of existing initiatives (García and Vilca, 2021; Navarro, 2025; Wakelin, 2021; Vita et al., 2021).

To address this gap and propose concrete guidelines for designing a contextualized and adaptable tutoring plan for the Peruvian reality, the present study analyzes the perception and impact of university tutoring. This work is based on the premise that a successful design must stem from a clear understanding of the needs and expectations of the university community (Alarcón et al., 2024; Chaverra et al., 2023; Martínez et al., 2022). The relevance of this study is related to higher academic performance for at-risk students in Peru and serve as a tool against university dropout by strengthening student support systems. Its objectives are to explore the correlations and perceptions of students regarding the tutoring service (Escalante et al., 2023).

University tutoring has become a key strategy in the holistic educational process of students, especially in the context of the massification of access to higher education and the diversity of student trajectories. This role has gained central importance, extending beyond mere academic support (Esteban and Caro, 2023; Cruzata et al., 2020; Pantoja et al., 2022). The tutor’s role has evolved from its traditional conception into a formative agent committed to the student’s development, acting as a critical companion, a facilitator of reflective processes, and a guide through the student’s academic journey (Cruzata et al., 2020).

Tutoring actions should be linked to the development of personal, academic, and professional competencies (Guffante et al., 2022; Moreno et al., 2023; Walker, 2020). It contributes to the student’s sense of belonging and encourages a critical attitude toward their education. It is emphasized that a non-directive educational relationship is necessary, where active listening and empathetic understanding are foundational to the tutoring relationship (Alarcón et al., 2024; Valdez et al., 2024; Yucra, 2022).

In the Latin American context, tutoring is recognized as an effective mechanism to reduce dropout rates, improve retention, and facilitate university adaptation (Benites, 2020; Escalante et al., 2023; Esquivel and Basilio, 2024). Tutoring enhances students’ responsibility and autonomy, which in turn boosts academic achievement and leads to significant improvements in soft skills such as leadership, communication, and decision-making. It plays an essential role in academic, institutional, and personal integration, with three complementary modalities: individual, group, and peer tutoring (Alcarraz and Sanchez, 2021; Benites, 2020; Bernardo et al., 2021).

The value of student perception depends on how students perceive and experience the quality of support, extending beyond the content provided. Emotional components should be explicitly included in tutoring programs, recognizing the importance of emotional sensitivity in addressing the psychosocial distress associated with the university transition (Angulo, 2021; Arámburo, 2023; Valdez et al., 2024). Successful programs share attention to individual needs, continuous teacher training, and an optimal relationship between the number of tutors and students (Alegre et al., 2024; Pari et al., 2024). Tutoring plans should consider the structural conditions of universities and the socio-emotional characteristics of students, integrating approaches to well-being and diversity (Elgegren, 2022; Vita et al., 2021).

The 21st-century university faces a dilemma between humanistic education and utilitarian orientation, with tutoring emerging as an alternative to personalize learning and facilitate adaptation to institutional and labor environments (Espinales et al., 2022; Gonzáles and Martínez, 2020; Martínez et al., 2020; Yana et al., 2024). Tutoring plays a role in minimizing constant failure rates and dropout, strengthening students’ capacities through individual or group guidance (Escalante et al., 2023). It must address students in an integrated manner, with clear objectives and continuous evaluation, as its effectiveness depends on the commitment and specialized training of the faculty (Medianero, 2017; Pantoja et al., 2022; Reséndiz and Zepeda, 2021).

Moreover, it is an effective mechanism for developing generic competencies such as critical thinking and self-regulation (Arámburo, 2023; Navarrete and Tomé, 2022; Yana et al., 2024). Student perception of tutoring is linked to the strengthening of interpersonal skills and the consolidation of transversal competencies. From an institutional perspective, it is necessary to document the importance of defining the scope of action, providing continuous training to tutors, and having monitoring systems to feed back the program based on evidence (Angulo, 2021; Arakaki et al., 2019; Guffante et al., 2022).

This study aims to provide large-scale empirical evidence on the perception and impact of university tutoring in a Latin American context, an understudied setting. Specifically, it seeks to: (1) quantify the relationship between tutoring, student satisfaction, and academic performance using quantitative analytical methods; (2) examine the moderating role of socioeconomic status and employment conditions through regression and structural equation modeling; and (3) identify critical gaps—particularly between students’ valuation of tutoring as a principle and their perception of its pedagogical execution—in order to inform the design of more effective and context-sensitive tutoring programs.

Rather than asking whether tutoring works, this study advances the literature by examining how tutoring is perceived, under which conditions it is more strongly associated with academic outcomes, and why students may simultaneously value tutoring while critically evaluating its implementation. This distinction is largely absent from prior research in Latin American higher education contexts.

This study follows an observational, cross-sectional design. Tutoring participation was not randomly assigned; instead, it was institutionally mandated as part of the university’s academic support system. In accordance with policies established by the Academic Vice-Rectorate, tutoring is a compulsory component of undergraduate education. Faculty members are formally assigned tutoring responsibilities—including specific student groups, academic sections, or semesters—as part of their teaching workload, and these tutoring hours are financially remunerated.

Tutors are allocated to student groups by the Faculty or School of Psychology, which oversees the tutoring system and assigns students based on institutional criteria rather than student choice. Students identified as being at academic risk due to low academic performance or repeated course failure are systematically included in the tutoring program. Although assignment to tutoring is institutional rather than voluntary, levels of student engagement and interaction with tutors may vary, which precludes causal inference. Consequently, all comparisons between students who did and did not actively receive tutoring should be interpreted with caution and understood as associative rather than causal.

The main variables were defined as follows. Tutor perception and satisfaction with tutoring were treated as independent and mediating variables within the SEM framework. Perceived academic impact and grade point average were modeled as outcome variables. Socioeconomic status and employment condition were specified as moderators, while gender, age, scholarship status, and academic field were included as control variables.

The study adopted a quantitative approach with a descriptive level. To deepen the understanding of the relationships between the studied variables and provide a more comprehensive view of the impact of university tutoring on academic performance, the analysis was complemented with a Structural Equation Modeling (SEM) approach (Hair et al., 2022; Kline, 2023).

The analysis was conducted using the Maximum Likelihood (ML) method, suitable for large samples and multivariate normal distributions (Byrne and Cahyono, 2022). Before estimating the structural model, assumptions of multicollinearity, normality, absence of extreme outliers, and sample adequacy were checked using the KMO index (> 0.80) and Bartlett’s Test of Sphericity (p < 0.001), confirming the suitability of the data for prior confirmatory factor analysis (Schumacker and Whittaker, 2022).

The model was estimated based on two main components. First, the measurement model was used to examine the convergent and discriminant validity of the dimensions associated with university tutoring. Standardized factor loadings above 0.60 and composite reliability coefficients (CR > 0.70) demonstrated adequate internal consistency. Additionally, the Average Variance Extracted (AVE) exceeded the recommended threshold of 0.50, confirming satisfactory convergent validity.

Second, the structural model allowed the evaluation of the hypothesized relationships between latent and observed variables, estimating standardized regression coefficients (β), statistical significance (p < 0.05), and the proportion of explained variance (R²) for each dependent variable.

The global model fit evaluation was conducted through various indices, according to international consensus criteria. A relative chi-square (χ²/df) value lower than 3.00 was considered indicative of an acceptable fit. Furthermore, the Comparative Fit Index (CFI) and Tucker–Lewis Index (TLI) with values equal to or higher than the recommended thresholds reflected an adequate model fit. Finally, values of the Root Mean Square Error of Approximation (RMSEA) and Standardized Root Mean Square Residual (SRMR) equal to or lower than 0.08 indicated a satisfactory match between the theoretical model and empirical data (Byrne and Cahyono, 2022; Hair et al., 2022; Kline, 2023).

Although Bartlett’s Test of Sphericity yielded a relatively modest chi-square value (χ² = 295.47, p < 0.001), this result should be interpreted in light of the reduced number of observed variables included in the factor analysis. In this context, the significance of the test confirms the presence of sufficient inter-item correlations to justify factor extraction, in combination with a high KMO index (0.88).

The findings of this study highlight an understudied dimension of university tutoring: the distinction between perceived institutional value and experienced pedagogical quality. While prior research has predominantly focused on overall satisfaction indicators or academic outcomes associated with tutoring programs, fewer studies have examined how direct exposure to tutoring reshapes student evaluations, often revealing discrepancies between normative expectations and actual pedagogical practice. This distinction is particularly relevant in institutional contexts where tutoring is formally established as a central academic support mechanism.

By documenting this divergence, the present study contributes to a more nuanced understanding of why tutoring programs may be simultaneously valued and criticized within the same university setting. The results suggest that satisfaction with the idea or principle of tutoring should not be conflated with satisfaction regarding its pedagogical delivery. Evaluations provided by students who actively participated in tutoring appear to reflect a more informed and critical assessment of tutor practices, whereas non-users may rely on indirect perceptions or institutional narratives. This finding underscores the importance of incorporating the perspectives of actual service users when assessing tutoring quality and effectiveness.

Overall, this study provides large-scale empirical evidence on the relationships between students’ perceptions of tutoring, satisfaction, and academic performance in a Peruvian university context. Using regression analysis and structural equation modeling, the results reveal a complex landscape in which positive associations with academic outcomes coexist with important limitations in program implementation. Given the cross-sectional and observational design of the study, all reported relationships must be interpreted as statistical associations rather than causal effects. Although the SEM results support theoretically coherent predictive pathways, the absence of random assignment or longitudinal data precludes definitive causal inference.

This research makes three primary contributions to the literature on university tutoring in Latin American contexts. First, it provides empirical evidence from a large and institutionally representative sample (N = 5,930) in a region that remains underrepresented in higher education research. Second, it demonstrates that associations between tutoring-related variables and academic outcomes are moderated by socioeconomic status and employment conditions, highlighting the relevance of contextual and structural factors. Third, it identifies a critical dissonance between students’ valuation of tutoring as an institutional support mechanism and their more critical evaluations of its pedagogical execution, particularly regarding tutors’ instructional competencies.

The regression analyses indicate a strong association between positive perceptions of tutoring and academic performance. However, this association is not uniform across academic contexts. Significant differences between faculties, with more favorable perceptions in Biomedical Sciences than in Engineering and Social Sciences, suggest that disciplinary cultures, curricular structures, and academic workloads may substantially shape how tutoring is experienced and evaluated. These findings align with previous research indicating that standardized, one-size-fits-all tutoring models may limit effectiveness in diverse academic environments and that discipline-sensitive approaches may enhance tutoring outcomes (Alcocer et al., 2022; Capa et al., 2020; Elgegren, 2022; Sánchez and Mares, 2025).

Students with stronger academic skills or higher intrinsic motivation may both perform better academically and evaluate institutional programs more positively. While the models control for several sociodemographic factors, residual confounding cannot be ruled out.

A central paradox emerging from the results is that students who received tutoring reported higher levels of satisfaction with their academic experience while simultaneously expressing more critical evaluations of tutoring quality. This pattern suggests that direct engagement with tutoring does not necessarily lead to more favorable perceptions but rather facilitates more realistic and discerning evaluations. Students appear to recognize the potential value of tutoring while also identifying shortcomings in its pedagogical implementation, particularly in instructional guidance. Similar discrepancies between affective support and pedagogical effectiveness have been reported in prior studies on university tutoring and mentoring relationships (Guffante et al., 2022; Moreno et al., 2023; Walker, 2020).

Further analysis reveals disparities in how different dimensions of tutoring are evaluated. Interpersonal attributes such as empathy, respectful treatment, and availability received relatively high ratings, suggesting that tutors generally demonstrate adequate relational skills. This finding is consistent with conceptualizations of tutoring as a fundamentally relational practice that fosters trust and institutional belonging (Ceniz et al., 2024; Navarrete and Tomé, 2022; Topping, 2017). In contrast, pedagogical competencies and the perceived importance attributed to the tutoring role were evaluated less favorably, pointing to potential gaps in pedagogical training and role clarity. These deficiencies may constrain the educational effectiveness of tutoring interventions and help explain why tutoring-related variables, although statistically significant, account for a limited proportion of variance in academic performance.

The persistence of associations between tutoring variables and academic performance after controlling for sociodemographic factors reinforces the relevance of tutoring while simultaneously underscoring its contextual dependence. Paid employment and scholarship status emerged as significant correlates of academic outcomes, indicating that tutoring operates within a broader ecosystem of economic, temporal, and institutional constraints. These findings are consistent with prior research emphasizing that academic support initiatives interact with students’ socioeconomic realities rather than functioning as isolated interventions (Esteban and Caro, 2023; Cruzata et al., 2020; Pantoja et al., 2022).

Structural equation modeling results provide further theoretical coherence to the proposed framework. Tutor perception was positively associated with satisfaction with tutoring, explaining a substantial proportion of its variance. Satisfaction, in turn, was significantly associated with perceived academic impact. In this sense, satisfaction can be understood as a key intervening construct linking perceptions of the tutor–student interaction with students’ perceived academic benefits. This interpretation is consistent with prior work highlighting the role of tutoring experiences in shaping student confidence, self-regulation, and perceived academic usefulness (Álvarez, 2020; Castro et al., 2022). These findings support conceptual models that frame tutoring not merely as an informational service but as a relational and motivational process that may strengthen persistence and engagement in higher education (Hair et al., 2022; Kim and Lundberg, 2015; Klug and Peralta, 2019; Solis and Cañarte, 2025).

The inclusion of socioeconomic status and employment condition as moderators allowed for a more differentiated understanding of tutoring effectiveness. Partial moderation by socioeconomic status suggests that the association between satisfaction and perceived academic impact is stronger among students from medium- and high-socioeconomic backgrounds, potentially due to greater access to complementary learning resources that facilitate the application of tutor guidance (Elgegren, 2022; Esquivel and Basilio, 2024). Employment status showed significant moderation, with weaker associations observed among working students. Time constraints, academic stress, and reduced availability for extracurricular support activities may limit working students’ capacity to fully benefit from tutoring, as suggested by previous studies on student workload and academic engagement (Coila et al., 2023; Garay et al., 2024).

Taken together, these findings indicate that while university tutoring is positively associated with key academic and experiential outcomes, its effectiveness depends critically on pedagogical quality, contextual adaptation, and institutional support. Improving tutor training, clarifying pedagogical role expectations, and designing flexible tutoring modalities that account for students’ socioeconomic and employment conditions may enhance the contribution of tutoring programs to academic success and student well-being.

An important limitation is the potential influence of unobserved confounding variables, such as prior motivation, academic self-efficacy, or institutional engagement, which may simultaneously affect academic performance and positive perceptions of tutoring. As a result, the estimated associations should be interpreted as upper-bound estimates rather than causal effects.

A conceptual limitation of this study concerns the use of the term tutoring. Although the activities analyzed may resemble what is referred to as mentoring in other educational systems, in the Peruvian higher education context mentoring is neither formally defined nor institutionally regulated. Consequently, the study adopts the term tutoring as used in local policy and practice, with its operational definition explicitly specified in the Methods section. Caution is therefore warranted when generalizing these findings to contexts where tutoring and mentoring represent distinct and formally differentiated models.

Future research should employ longitudinal designs to better understand the temporal dynamics of tutoring effects. Experimental studies with random assignment to tutoring conditions would help establish causal relationships. Mixed-methods approaches could provide deeper insights into the qualitative experiences of both tutors and students. Additionally, research comparing different tutoring models across diverse institutional contexts would help identify best practices adaptable to various educational settings.

In light of the findings, university tutoring emerges as a dual-nature intervention in the Peruvian context studied: it demonstrates potential associations with improved academic performance and enhanced student trajectories, but its current implementation appears constrained by pedagogical competency gaps. The study reveals that students recognize the value of tutoring while remaining cognizant of its deficiencies. A positive perception of the tutor is associated with higher satisfaction with tutoring, and this satisfaction is linked to greater perceived impact on academic performance—relationships that are moderated by students’ socioeconomic and employment conditions.

The evidence suggests that merely having a tutoring program is insufficient; how it is designed and for whom it is adapted are crucial considerations. The marked differences between faculties are not mere variations but potential indicators of an overly standardized model within a diverse academic ecosystem. Twenty-first century tutoring cannot be merely an administrative protocol; it should be a reflective practice, contextualized to disciplinary needs, and supported by tutors who are not only companions but active facilitators of the learning process.

The future of tutoring lies not in its mere existence but in its capacity to evolve and adapt to student needs: it should transition from well-intentioned support to an intelligent strategy, integrated into the curriculum, and backed by substantial tutor training. Institutional commitment to ongoing tutor development, contextualized program design, and systematic evaluation will be essential to realize the full potential of university tutoring as a tool for educational enhancement and student support.