Automated motivational support and academic self-regulation: evidence from a large-scale cross-sectional study

Abstract

This study examined motivational patterns among university students using the framework of Self-Determination Theory, focusing on the relationships between extrinsic motivation (EM), intrinsic motivation (IM), and amotivation (A). A quantitative cross-sectional design was applied to a sample of 723 undergraduate students from different academic fields at a public university in Peru. Data were collected using the Academic Situational Motivation Scale (EMSA) and analyzed using descriptive statistics, one-way ANOVA, Pearson correlations, multiple regression, and cluster analysis. The results revealed generally high levels of intrinsic and extrinsic motivation, accompanied by low levels of amotivation. ANOVA showed statistically significant differences in EM [F(4, 723) = 4.517, p = 0.001, η² = 0.024] and IM [F(4, 723) = 7.029, p < 0.001, η² = 0.037] across academic fields, while amotivation showed no significant differences across the fields. Correlation and regression analyses indicated that extrinsic motivation positively predicted intrinsic motivation, whereas amotivation negatively predicted it. Cluster analysis identified three motivational profiles: highly motivated, moderately motivated, and vulnerable groups. These findings highlight the heterogeneous nature of student motivation and suggest that universities may benefit from implementing scalable digital and pedagogical strategies aimed at strengthening adaptive motivational patterns and reducing motivational vulnerability in higher-education contexts.

1 Introduction

In higher education, students must face multiple challenges linked not only to aca-demic content but also to their personal, social, and emotional environment (Mohamed et al., 2025; Nannings et al., 2025). These challenges have intensified in recent years, driving the need to adopt more comprehensive and innovative approaches in training processes (Mhlongo et al., 2023; Zamiri and Esmaeili, 2024). Following the health crisis caused by the pandemic, various institutions have redefined their pedagogical strategies, accelerating the incorporation of educational technologies and emphasizing student well-being (Bustillos-Cotrado, 2023; Casas-Huamanta, 2022; Li, 2022). In this context, Goal 4 of the 2030 Agenda promotes inclusive, equitable, and quality education, focused on meaningful learning (United Nations, 2025). However, dropout rates and low academic performance persist, highlighting the need to address not only structural factors but also those related to motivation and self-regulation of learning (Al-Tameemi et al., 2023; Kocsis and Molnár, 2025).

In this regard, self-regulated learning implies that students take an active role in their educational process, managing their goals, strategies, and emotions in response to aca-demic demands (Brenner, 2022; Simón-Grábalos et al., 2025). This approach recognizes that learning does not depend exclusively on external instruction, but also on the individual's ability to monitor and adjust their own behavior (Radović et al., 2024). In this sense, three main phases have been identified: planning, execution, and self-evaluation, each mediated by metacognitive and motivational processes (Zimmerman, 2002). During these stages, students activate thoughts and emotions that influence academic decision-making as well as perseverance in the face of difficulties (Velásquez-Pérez et al., 2023). Strengthening these skills is key to consolidating autonomous and sustainable learning (Song et al., 2024). To this end, it is essential to understand the conditions that favor their development and the theories that explain how students regulate their behavior toward achieving personal and academic goals (Bandhu et al., 2024; Kryshko et al., 2022).

Aligned with these principles, contemporary educational environments have begun to integrate digital tools as a means to foster students' autonomous development (Mena-Guacas et al., 2025; Timotheou et al., 2023). Mobile technologies, due to their ubiquity and widespread acceptance among university students, represent a valuable opportunity to activate motivational processes in everyday learning (Mhlongo et al., 2023; Yu and Yang, 2024). Beyond their communicative function, these platforms allow for continuous formative connections, reinforce the sense of competence through reminders or feedback messages, and strengthen ties with the academic environment (Bazhenova et al., 2022; Omirzak et al., 2021). Thus, the strategic incorporation of technology can become a pedagogical resource that fosters self-determination and self-regulation (Chiu et al., 2024; Junaštíková, 2024; Starke and Ludviga, 2025). Particularly in Latin America, its potential has begun to be explored in educational experiences aimed at improving student engagement from a more human, accessible, and sustainable approach (Agasisti et al., 2023; Villa-Enciso et al., 2023; Viloria et al., 2020).

Among the digital tools widely used by university students, WhatsApp stands out for its high penetration and familiarity. Although traditionally associated with informal communication, in recent years it has generated growing interest as a medium to support educational processes (Fondevila-Gascón et al., 2019; Romero-Saritama et al., 2025; Suárez-Lantarón et al., 2022). In particular, the possibility of automating the sending of pedagogical messages offers a new approach to asynchronous support, aimed at reinforcing routines, promoting study habits, and activating cognitive processes related to self-regulation. This strategy becomes more valuable when grounded in solid theoretical frameworks, such as Deci and Ryan's self-determination theory (Deci and Ryan, 1985), which posits that quality motivation arises from meeting three basic psychological needs: autonomy, competence, and social relatedness. Complementarily, Nicholls' achievement goal theory (Nicholls, 1989) helps explain how personal goals and the perception of success influence academic effort. The integration of both approaches provides a robust framework for the design of student-centered educational interventions.

Considering these theoretical foundations and the possibilities offered by mobile technologies in educational settings, it is pertinent to explore strategies that promote academic self-regulation through accessible and contextualized mechanisms. In particular, the use of automated messaging systems, grounded in principles of autonomous motivation, emerges as a viable pedagogical alternative to support students in their educational process. From this perspective, the present study aims to evaluate the effect of an automated intervention based on self-determination theory, through the scheduled delivery of motivational messages via WhatsApp, on strengthening academic self-regulation in university students. This proposal seeks to contribute to the design of more personalized and sustainable educational environments, capable of fostering students' active engagement in their learning.

This study is guided by the following research questions: (1) What are students' perceptions in terms of extrinsic motivation (EM), amotivation (A), and intrinsic motivation (IM)?; (2) Are there significant differences in EM, A, and IM according to students' gender?; (3) Are there significant differences in motivational constructs according to students' field of study?; and (4) What relationships exist between EM, A, and IM? Addressing these questions will help identify key patterns and relationships that facilitate understanding of how these motivational factors interact in the university context. The results obtained will provide practical implications for the design of more effective and tailored educational interventions, while also contributing empirical evidence to guide future research in the field of motivation and academic self-regulation.

2 Related work

Bellhäuser et al. (2023) evaluated the effects of automatically generated adaptive feedback on daily self-regulated learning (SRL) in university students. The longitudinal experimental study included 194 participants randomly assigned to an experimental group with learning diary plus feedback (LDF, n = 98) and a control group without feedback (LD, n = 96). Participants completed electronic diaries for 36 days with morning and evening questionnaires on goals, planning, motivation, self-efficacy, adherence to schedules, procrastination, effort, and satisfaction. The LDF group received daily written feedback—confirmatory, informative, or transformative—automatically generated from their responses. Multilevel analyses showed that, compared to LD, the LDF group scored higher in goal setting (b = 0.19), planning (b = 0.24), self-efficacy (b = 0.24), satisfaction (b = 0.20), and schedule adherence (b = 0.40), with small to medium effect sizes, while intrinsic motivation, effort, and procrastination did not differ significantly. Exploratory analyses revealed that confirmatory and transformative feedback had more consistent positive effects on satisfaction, procrastination reduction, and time management, whereas informative feedback showed no relevant impact. No effects were found on trait-level SRL measures (pre- and post-test), leading to the conclusion that daily automatic feedback integrated into learning diaries enhances key components of SRL in daily processes

Chacón-Cuberos et al. (2025) developed a theoretical model using structural equation modeling to analyze relationships between self-concept, basic psychological needs, and learning strategies in 2,736 Spanish university students (66.2% women; 33.8% men). The García and Musitu (2001) Self-Concept Scale, the Sheldon and Hilpert (2012) Basic Psychological Needs Scale, and the MLSQ-SF were applied, with analyses in AMOS 22.0. Positive associations were found between self-concept and the needs for autonomy and competence, stronger among men; satisfaction of these needs was associated with greater use of learning strategies (p < .005), with no significant relationship with relatedness. Women scored higher in elaboration strategies (β = 0.784) and men in effort self-regulation (β = 0.828). The model showed acceptable fit (CFI = 0.91; RMSEA = 0.067), recommending educational environments that strengthen autonomy, competence, and self-concept to promote self-regulated learning.

The available evidence demonstrates that self-regulated learning, self-determined motivation, and student engagement can be strengthened through structured interventions that integrate pedagogical strategies, adaptive feedback, and support for basic psychological needs. The application of electronic diaries with automatic feedback, teacher autonomy-support programs, brief interventions based on motivational theories, and self-directed e-learning environments has shown improvements in planning, self-efficacy, motivational regulation, and perseverance, as well as in the use of effective learning strategies. These findings underscore the importance of replicating and adapting these proposals across diverse educational contexts, exploring their effectiveness at different levels and modalities, and leveraging digital tools to personalize feedback, monitor processes in real time, and broaden the reach of strategies that support self-regulated learning.

3 Materials and methods

3.1 Research design

This study adopted a quantitative cross-sectional research design to examine motivational patterns among university students and explore the relationships between intrinsic motivation (IM), extrinsic motivation (EM), and amotivation (A) within the framework of Self-Determination Theory. The study analyzed students' perceptions collected through a structured questionnaire and examined statistical associations among motivational constructs, as well as differences according to gender and field of study. Additionally, the predictive relationships between motivational variables were explored using regression analysis, and motivational profiles within the sample were identified through cluster analysis. This design allowed for the identification of general motivational trends and patterns within the university population while maintaining a descriptive and correlational analytical approach.

3.2 Context and participants

The study was conducted during the 2024-II academic semester at a public university in the San Martín region of northeastern Peru. The institution's mission is to provide high-quality professional training, promote applied research, and contribute to the sustainable development of the region and its surrounding areas of the Amazon rainforest. Its academic structure comprises faculties such as Education Sciences, Health Sciences, Engineering and Technology, Economic Sciences, and Environmental Engineering, offering a wide range of undergraduate programs. The student population is diverse, representing various social and academic sectors in the regional context.

A total of 723 students participated in the study, all of whom were enrolled in general education courses delivered transversally across different academic areas. Table 1 summarizes the sample's demographic information, including distribution by gender and field of study. Regarding gender, 53.0% were male (n = 383) and 47.0% were female (n = 340), indicating a relatively balanced proportion. In terms of field of study, 23.4% belonged to Economic Sciences, 22.4% to Education Sciences, 19.6% to Engineering and Technology, 19.4% to Health Sciences, and 15.2% to Environmental Engineering fields. This composition reflects a heterogeneous and representative sample of academic diversity within the institution.

3.3 Data collection

The study employed the Academic Situational Motivation Scale (EMSA) (Bruno et al., 2020), which was adapted to the university context of the present research. The instrument consisted of 27 items distributed across three constructs: EM (12 items), A (six items), and IM (nine items) (Table 2). Each item was rated on a 5-point Likert scale ranging from 1 (not at all true) to 5 (completely true). The adaptation process preserved the original structure of the EMSA, ensuring conceptual equivalence in terms of wording, content, and item order.

Table 3 presents the reliability analysis of the EM, A, and IM constructs using Cronbach's alphas. Cronbach's alpha is a measure of internal consistency that indicates the reliability of items within each construct. The EM construct, composed of 12 items, showed a Cronbach's alpha of 0.856. This value indicates a high level of reliability, as it exceeds the commonly accepted threshold of 0.7 for adequate internal consistency of the scale. Similarly, the A construct, with six items, obtained a Cronbach's alpha of 0.931, denoting excellent reliability and strong consistency among its items. The IM construct, comprising nine items, recorded a Cronbach's alpha of 0.927, reflecting high reliability. The Cronbach's alpha values for the three constructs indicate that the items within each construct are internally consistent and reliable for measuring their respective concepts. The high reliability supports the validity of these constructs in the present study.

3.4 Data analysis

The data analysis techniques applied in this study consisted of several complementary statistical procedures to assess participants' perceptions and examine the relationships among the constructs of EM, A, and IM, using SPSS software (version 31). First, descriptive statistics were employed to summarize and describe the participants' overall perceptions of each construct, including measures such as means and standard deviations, to provide an overview of central tendencies and data variability. Subsequently, a one-way analysis of variance (ANOVA) was applied to identify significant differences in the levels of EM, A, and IM according to sociodemographic variables of interest, such as academic area of study, allowing for the identification of statistically significant variations among the different groups.

Additionally, Pearson's correlation analysis was conducted to examine the strength and direction of the linear relationships among EM, A, and IM, enabling the identification of associative patterns between motivational constructs and verifying their consistency with the theoretical framework of motivation. After the correlation analysis, a multiple linear regression model was estimated to examine whether EM, A, gender, and field of study predicted IM. This regression model allowed us to identify the relative contribution of each variable to the variance in intrinsic motivation and to estimate the predictive influence of motivational constructs and demographic variables within the proposed analytical framework.

Finally, a cluster analysis using the K-means algorithm was conducted to identify motivational profiles among students based on the three constructs of EM, A, and IM. This procedure grouped students with similar motivational patterns into homogeneous clusters, enabling the identification of distinct motivational configurations in the sample. Together, these analytical procedures allowed the identification of descriptive trends, group differences, associative relationships, predictive effects, and motivational profiles within the student population, providing a comprehensive statistical basis for interpreting the motivational dynamics examined in this study.

3.5 Automated messaging system and intervention procedure

The proposed intervention model was implemented using a client–server architecture that integrated an administration panel, a central database, and an automated messaging bot (Figure 1). The control panel, developed in Django, enabled the management of messages, participants, and schedules, ensuring flexibility in content customization and control over the delivery sequence.

Figure 1

The MySQL database served as a central repository for storing information on messages, participant data, and scheduled times. The messaging bot, implemented with Venom Bot on Node.js, queried the database to select and send scheduled messages and interacted directly with the WhatsApp client.

The operational flow began with the configuration of messages and schedules from the control panel, followed by registration in the database. The bot periodically queried these records, selected the corresponding messages, and sent them to the participants via WhatsApp. All interactions were automatically logged, allowing precise monitoring of participation and feedback. This technological integration ensured a continuous, automated, and low-cost educational support process, thereby optimizing the implementation of the proposed model in this study.

In terms of intervention deployment, the automated messaging system was implemented over an eight-week period during the semester. All participating students received the intervention through WhatsApp, where motivational messages were delivered automatically according to the predefined schedules and interaction rules. Students received between two and five automated messages per week, depending on system interactions and content personalization, with the objective of reinforcing study routines, goal orientation, and academic persistence. Across the eight-week intervention period, participants received an estimated average of approximately 16–20 messages. The content of the messages was structured into three main categories aligned with the principles of self-regulated learning and Self-Determination Theory: academic reminders (40%), aimed at encouraging students to maintain consistent study routines and meet academic commitments; motivational prompts (35%), intended to reinforce persistence, confidence, and engagement with learning activities; and study strategy suggestions (25%), focused on promoting planning, goal setting, and effective learning habits. The messages were automatically distributed through a rule-based system to ensure consistent exposure across participants. System monitoring indicated an average message reading rate of approximately 91% throughout the intervention period, suggesting a high level of student exposure to the system-delivered motivational prompts. This procedure ensured consistent exposure to the intervention across all participants while allowing minor variations in message frequency owing to the personalization rules implemented in the system.

4 Results

4.1 Students' perceptions in terms of EM, A, and IM

To address the first research question, a quantitative approach was applied using a structured questionnaire that assessed three key constructs: EM, A, and IM. Table 4 presents the descriptive statistical analysis of the mean, standard deviation, and acceptance level for each construct, allowing for the identification of participants' overall assessment of their motivation toward class attendance and participation.

The descriptive analysis revealed that the IM construct recorded the highest mean (M = 4.00507, SD = 0.129118), classified as having a “High” level of acceptance, reflecting a strong orientation toward autonomous learning and intrinsic enjoyment of academic activities. This result indicates that students value personal progress and content comprehension as central elements of their educational experience, consistent with theoretical frameworks that associate autonomous motivation with greater persistence and higher-quality learning. Second, EM also reached a “High” level (M = 3.79979, SD = 0.180741), showing that external incentives, such as assumed responsibilities, expectations from others, and academic benefits, coexist with internal motivations and contribute to sustaining academic engagement.

In contrast, the A construct presented the lowest mean (M = 1.82596, SD = 0.539246), with a “Low” level of acceptance, indicating a minimal presence of negative perceptions associated with attending or participating in classes. This motivational profile, characterized by the predominance of IM, reinforcement of EM, and minimal incidence of A, suggests an academic environment that fosters active student engagement. The combination of high autonomous and controlled motivation with low levels of amotivation aligns with empirical evidence that emphasizes the importance of optimizing both types of motivation as a strategy to sustain performance and self-regulation over the long term.

4.2 Students' perception in terms of EM, A, and IM by gender

To address the second research question, a descriptive statistical analysis was conducted considering the participants' gender, evaluating the three key constructs: EM, A, and IM. Figure 2 presents the mean, standard deviation, and level of acceptance for each construct, differentiating between male and female students, allowing for the identification of possible variations in their motivational perceptions toward attending and participating in class.

Figure 2

In EM, both men (M = 3.79535, SD = 0.71892) and women (M = 3.80959, SD = 0.65864) reached a “High” level of acceptance, indicating that, regardless of gender, students value external factors that encourage class attendance, such as institutional commitment, the expectations of others, and academic benefits. In A, men scored higher (M = 1.98749, SD = 1.06092) than women (M = 1.68848, SD = 0.85724), with both groups remaining at a “Low” level, indicating a low incidence of negative perceptions toward academic participation, although with a slightly greater predisposition to demotivation in the male group.

In IM, women obtained a slightly higher mean (M = 4.05209, SD = 0.47724) than men (M = 3.96232, SD = 0.54899), with both groups classified as “High” acceptance. This pattern confirms that intrinsic motivation is a central component of the academic experience, particularly among women, who report greater enjoyment and personal satisfaction in relation to learning. Overall, the observed motivational profile was characterized by high levels of IM and EM and low levels of A. Although the differences were moderate, they suggest the relevance of designing gender-differentiated motivational interventions that strengthen IM in men and consolidate the strengths observed in women.

4.3 Comparison of EM, A, and IM by area of study

To examine whether differences exist in the levels of EM, A, and IM among students from different academic areas, a one-way analysis of variance (ANOVA) was conducted. Figure 3 presents the means, standard deviations, and sample sizes for each construct, as well as the corresponding F and p values, allowing the identification of statistically significant or non-significant variations depending on the field of study.

Figure 3

ANOVA revealed statistically significant differences in EM [F(4, 723) = 4.517, p = 0.001, η² = 0.024], indicating a small effect size. The highest mean was observed in Economic Sciences (M = 3.95611, SD = 0.62394) and the lowest in Engineering and Technology (M = 3.62735, SD = 0.61705), suggesting that although EM shows high values across all areas, students in Economic Sciences exhibit a slightly stronger influence of external factors associated with academic benefits and social expectations.

For amotivation (A), no statistically significant differences were identified [F(4, 723) = 1.349, p = 0.250, η² = 0.007], with low means in all groups (range: 1.66905–1.93075), indicating a low incidence of negative perceptions toward class attendance or participation across all academic fields.

Significant differences were detected for IM [F(4, 723) = 7.029, p < 0.001, η² = 0.037], indicating a small-to-moderate effect. The highest means corresponded to Economic Sciences (M = 4.12985, SD = 0.46007), followed by Health Sciences (M = 4.07341, SD = 0.45640) and Education Sciences (M = 4.00789, SD = 0.49452), while the lowest value was recorded in Engineering and Technology (M = 3.85955, SD = 0.51586).

Overall, the results indicate that although motivational levels are generally high across all academic areas, statistically significant variations exist depending on the discipline. These findings suggest that motivational orientations may be partially shaped by academic environments and disciplinary cultures, highlighting the importance of considering contextual factors when designing strategies to support students’ motivation.

4.4 Correlations between EM, A, and IM

To examine the relationships between the three constructs of interest—EM, IM, and A—Pearson correlation coefficients were calculated to identify the strength and direction of the associations between them. Figure 4 presents the resulting correlation matrix, which facilitates the visualization of the interrelationships among the motivational factors in the studied population.

Figure 4

The results indicate that the highest correlation was between EM and IM (r = 0.467), showing a moderate positive association. This pattern suggests that students with higher levels of EM linked to external rewards or recognition also tend to present higher levels of IM associated with personal interest and satisfaction derived from academic activity. The correlation between A and IM (r = −0.224) was negative and of low magnitude, indicating that an increase in A was associated with a decrease in IM, consistent with theoretical frameworks stating that the lack of purpose or meaning inhibits autonomous motivation. Meanwhile, the correlation between A and EM (r = −0.094) was also negative, although practically null, reflecting the minimal influence of A on EM in this context.

These findings show that EM and IM tend to reinforce each other, whereas A is inversely related to both, especially IM. This profile confirms the relevance of implementing educational strategies that simultaneously strengthen autonomous and controlled motivation while reducing factors that foster amotivation to optimize self-regulation and academic engagement.

4.5 Predictors of intrinsic motivation (regression analysis)

To further examine the relationships among motivational constructs, a multiple linear regression analysis was conducted to determine whether EM, A, gender, and field of study predicted IM. This analysis allowed us to identify the relative contribution of each variable in explaining the variance in intrinsic motivation among students. Table 5 presents the regression model coefficients and overall model statistics.

The regression model was statistically significant (F = 21.54, p < 0.001), explaining approximately 25.7% of the variance in intrinsic motivation (R² = 0.257). These results indicate that the set of predictors included in the model contributed significantly to understanding the differences in intrinsic motivation levels among participants.

As shown in Table 5, extrinsic motivation had a significant positive effect on intrinsic motivation (β = 0.337, p < 0.001), indicating that students who reported higher levels of extrinsic motivation also tended to exhibit higher intrinsic motivation. In contrast, amotivation showed a significant negative association with intrinsic motivation (β = −0.087, p = 0.009), suggesting that higher levels of demotivation were associated with lower levels of intrinsic engagement in academic activities.

Gender did not show a statistically significant effect on intrinsic motivation (β = −0.017, p = 0.771), indicating that motivational differences between male and female students were not relevant to the predictive model. Similarly, the field of study showed a marginal and non-significant effect (β = −0.038, p = 0.064), suggesting that disciplinary differences had a limited influence on intrinsic motivation when controlling for the motivational constructs included in the model.

Overall, these findings highlight the central role of motivational constructs —particularly extrinsic motivation and amotivation—in explaining intrinsic motivation among university students, while demographic variables such as gender and academic field showed comparatively weaker predictive contributions.

4.6 Motivational profiles identified through cluster analysis

To further examine the heterogeneity of students' motivational patterns, a cluster analysis using the K-means algorithm was conducted based on three motivational constructs: EM, A, and IM. This analysis aimed to identify groups of students with similar motivational configurations and reveal distinct motivational profiles within the sample. The optimal number of clusters (k = 3) was determined through iterative testing of different cluster solutions and by evaluating the interpretability and stability of the resulting profiles.

Cluster analysis identified three differentiated motivational profiles among the students. The centroid values for each construct across the clusters are presented in Table 6, and Figure 5 visually illustrates the motivational patterns characterizing each profile.

Figure 5

Cluster analysis identified three distinct motivational profiles among the participating students. The first cluster, labeled the “Highly motivated profile”, represented the largest group and was characterized by the highest levels of intrinsic motivation (IM = 4.32) and extrinsic motivation (EM = 4.22), combined with the lowest levels of amotivation (A = 1.35). This configuration suggests a highly adaptive motivational orientation in which students demonstrate strong engagement in academic activities and a clear internal disposition toward learning. In contrast, the second cluster corresponded to a “Moderately motivated profile”, presenting moderate levels of intrinsic motivation (IM = 3.65) and extrinsic motivation (EM = 3.22), together with relatively low levels of amotivation (A = 1.48). This group reflects students who maintain stable academic engagement, although with lower motivational intensity than the highly motivated profile.

The third cluster represented a “Vulnerable motivational profile”, characterized by comparatively higher levels of amotivation (A = 3.41) and moderate levels of intrinsic motivation (IM = 3.78) and extrinsic motivation (EM = 3.62). Students in this group may experience greater motivational uncertainty or disengagement, which could potentially affect their academic persistence and involvement over time. The motivational patterns of these clusters are visually illustrated in Figure 5, which highlights the contrasting relationships between intrinsic motivation, extrinsic motivation, and amotivation across the identified profiles.

To further illustrate the separation between motivational profiles, a three-dimensional visualization of the K-means clustering solution was generated using EM, A, and IM as axes (Figure 6). This graphical representation allows for a clearer visualization of how students are grouped according to their motivational characteristics and provides additional support for the existence of differentiated motivational patterns within the sample.

Figure 6

Overall, these results indicate that although the descriptive analyses suggested generally high levels of motivation among students, the cluster analysis revealed that the student population was not homogeneous. Instead, distinct motivational profiles coexist within the sample, reflecting different combinations of intrinsic motivation, extrinsic motivation, and amotivation among university students.

5 Discussion

The findings of this study indicate that university students generally exhibit high levels of IM and EM, accompanied by comparatively low levels of A. These results suggest a motivational orientation consistent with self-regulated academic engagement, in which students demonstrate both an internal interest in learning and responsiveness to external academic expectations. This pattern aligns with previous empirical evidence highlighting the importance of motivation and self-regulatory processes in higher education contexts. Prior research has shown that structured academic support and feedback mechanisms can strengthen key components of self-regulated learning, including goal setting, planning, and persistence in academic tasks (Bellhäuser et al., 2023).

The correlational analysis further revealed significant relationships among the motivational constructs examined in this study. Intrinsic motivation was positively associated with extrinsic motivation, whereas both constructs showed negative relationships with amotivation. These findings suggest that internally and externally oriented motivational processes may coexist in complementary ways in academic environments. Similar patterns have been documented in studies grounded in Self-Determination Theory, which proposes that controlled and autonomous forms of motivation can jointly support academic engagement when external incentives are aligned with students' personal goals and values (Deci and Ryan, 1985). In this regard, the observed negative association between amotivation and both motivational constructs reinforces the idea that stronger motivational orientations tend to coexist with lower levels of academic disengagement.

The regression analysis conducted in this study provides additional evidence of the predictive relationships among motivational variables. Extrinsic motivation was a significant positive predictor of intrinsic motivation, whereas amotivation was negatively associated with intrinsic motivation. These findings suggest that external motivational drivers may support internally regulated engagement when they reinforce meaningful goals and expectations. Conversely, higher levels of amotivation appear to weaken students' internal interest in academic activities. Although the cross-sectional design does not allow causal inferences, these predictive patterns highlight the interconnected nature of motivational constructs and their potential role in shaping students' engagement in learning.

The ANOVA results also revealed statistically significant differences in intrinsic and extrinsic motivation across academic fields, although the effect sizes were small. These differences suggest that disciplinary environments may influence students experiences and regulation of motivation. Academic cultures, pedagogical practices, and professional expectations associated with different fields of study may shape the relative importance of intrinsic and extrinsic motivation drivers. Nevertheless, despite these variations, the overall pattern across all fields was characterized by high levels of motivation and low amotivation, indicating that the general motivational orientation of the student population was relatively consistent.

Beyond examining the linear relationships among variables, cluster analysis provided additional insights into the heterogeneity of motivational patterns within the student population. Three distinct motivational profiles were identified: highly motivated, moderately motivated, and vulnerable. The highly motivated cluster was characterized by high levels of intrinsic and extrinsic motivation combined with low amotivation, suggesting a strong orientation toward self-regulated academic engagement. The moderately motivated cluster presented intermediate motivational levels and relatively low amotivation, while the vulnerable cluster exhibited higher levels of amotivation, indicating a subgroup of students who may experience greater motivational uncertainty or disengagement from their studies.

The three-dimensional visualization of the cluster structure further supports this interpretation by illustrating the spatial separation between motivational profiles across the dimensions of intrinsic motivation, extrinsic motivation, and amotivation. This graphical representation highlights that motivational orientations among students are not uniformly distributed but are organized into differentiated patterns of engagement. These findings reinforce the multidimensional perspective of Self-Determination Theory, which conceptualizes motivation as a dynamic and context-dependent phenomenon shaped by both individual psychological needs and environmental influences.

The gender-related differences observed in this study may also be interpreted in light of previous research examining motivational and self-regulatory processes among university students. Some studies have reported that female students tend to demonstrate stronger engagement with learning activities and greater use of elaboration and planning strategies, whereas male students may exhibit different patterns of effort regulation (Chacón-Cuberos et al., 2025). In the present study, the slightly higher levels of intrinsic motivation observed among female students and the marginally higher levels of amotivation among male students may reflect differences in academic engagement strategies rather than fundamental disparities in motivational capacity.

Taken together, the results of this study contribute to the growing body of research on motivational processes in higher education. Although this study did not evaluate the causal impact of the messaging intervention, the motivational patterns identified are consistent with theoretical frameworks emphasizing the role of structured academic support, autonomy-supportive environments, and goal-oriented regulation in fostering student engagement. These findings highlight the importance of understanding motivational diversity within university populations and provide a basis for future research exploring how digital learning supports and pedagogical strategies may contribute to strengthening adaptive motivational profiles among students.

6 Limitations of the study

Despite the contributions of this study, several limitations should be considered. First, the research was conducted within a single university context, which may limit the generalizability of the findings to other institutions or cultural environments. Although the sample included students from multiple academic fields, motivational dynamics may differ across universities with distinct pedagogical practices, technological infrastructures, and student populations. Future research should consider multi-institutional designs that allow for the comparison of motivational patterns across diverse higher education contexts.

Second, this study relied primarily on self-reported measures of motivation. While validated instruments were used to assess intrinsic motivation, extrinsic motivation, and amotivation, the absence of behavioral indicators, such as learning analytics, academic performance records, or system usage data, limits the possibility of triangulating students' perceptions with observable learning behaviors. Integrating objective indicators of engagement and performance would provide a more comprehensive understanding of how motivational perceptions relate to actual academic outcomes.

Finally, the cross-sectional design restricted the ability to examine the temporal evolution of motivational processes or to infer causal relationships among variables. Although regression and cluster analyses provide insights into predictive relationships and motivational profiles, these results should be interpreted as associative rather than causal. Future studies should employ longitudinal or experimental designs to examine the stability of motivational profiles over time and evaluate the potential effects of structured digital interventions on students' motivational development.

7 Conclusions

The present study revealed that IM and EM reached high levels in the analyzed population, while A remained comparatively low, indicating a general motivational orientation consistent with self-regulated academic engagement. Inferential analyses showed significant differences in motivational constructs according to gender and field of study, suggesting that contextual and disciplinary factors may influence how students internalize and regulate their academic goals. These findings are consistent with the principles of Self-Determination Theory, which proposes that motivation emerges from the interaction between individual psychological needs and contextual learning environments.

Correlational and regression analyses further indicated that motivational constructs operate in an interconnected manner. Extrinsic motivation was identified as a significant positive predictor of intrinsic motivation, whereas amotivation showed a negative predictive effect, suggesting that externally oriented motivations may coexist with internally driven engagement when they are aligned with meaningful academic goals. Additionally, cluster analysis revealed the presence of three distinct motivational profiles—highly motivated, moderately motivated, and vulnerable—demonstrating that student motivation is not homogeneous within the population. These profiles highlight the coexistence of different motivational configurations, ranging from strongly self-regulated engagement to patterns characterized by higher motivational vulnerability.

Future research should adopt longitudinal designs to examine the stability and evolution of motivational profiles across university students' academic trajectories. Experimental and quasi-experimental studies would also be valuable for evaluating whether structured interventions, such as autonomy-supportive feedback systems or digital motivational support, can influence the development of adaptive motivational patterns. Moreover, integrating behavioral indicators, such as learning analytics, academic performance measures, and engagement data, could provide a more comprehensive understanding of how motivational constructs translate into observable academic outcomes and persistence in higher education. From an institutional perspective, these findings suggest that universities may benefit from implementing scalable digital support strategies aimed at strengthening adaptive motivational patterns and reducing students' motivational vulnerability.

References

• 1

  AgasistiT.AntequeraG.DelpratoM. (2023). Technological resources, ICT use and schools efficiency in Latin America—insights from OECD PISA 2018. Int. J. Educ. Dev.99, 102757. 10.1016/j.ijedudev.2023.102757

  • CrossRef
  • Google Scholar

• 2

  Al-TameemiR. A. N.JohnsonC.GitayR.Abdel-SalamA.-S. G.Al HazaaK.BenSaidA.et al (2023). Determinants of poor academic performance among undergraduate students—a systematic literature review. Int. J. Educ. Res. Open4, 100232. 10.1016/j.ijedro.2023.100232

  • CrossRef
  • Google Scholar

• 3

  BandhuD.MohanM. M.NittalaN. A. P.JadhavP.BhadauriaA.SaxenaK. K. (2024). Theories of motivation: a comprehensive analysis of human behavior drivers. Acta Psychol. (Amst)244, 104177. 10.1016/j.actpsy.2024.104177

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  BazhenovaE.ShuzhebayevaA.KuntuganovaS.BazhenovaM.MuryginaS. (2022). The impact of Mobile learning on undergraduate Students’ cognitive learning outcomes: a meta-analytic review. Int. J. Eng. Pedagog.12, 42–53. 10.3991/ijep.v12i5.32821

  • CrossRef
  • Google Scholar

• 5

  BellhäuserH.DignathC.TheobaldM. (2023). Daily automated feedback enhances self-regulated learning: a longitudinal randomized field experiment. Front. Psychol.14:1125873. 10.3389/fpsyg.2023.1125873

  • CrossRef
  • Google Scholar

• 6

  BrennerC. A. (2022). Self-regulated learning, self-determination theory and teacher candidates’ development of competency-based teaching practices. Smart Learn. Environ.9, 3. 10.1186/s40561-021-00184-5

  • CrossRef
  • Google Scholar

• 7

  BrunoF. E.Fernández LiporaceM.StoverJ. B. (2020). Escala de motivación situacional académica para estudiantes universitarios: desarrollo y análisis psicométricos. Interdiscip. Rev. Psicol. Cienc. Af.37, 129–144. 10.16888/interd.2020.37.1.8

  • CrossRef
  • Google Scholar

• 8

  Bustillos-CotradoJ. A. (2023). COVID-19: educación remota para el desarrollo de competencias investigativas en estudiantes de una diplomatura. Desafios14, 60–61. 10.37711/desafios.2023.14.1.384

  • CrossRef
  • Google Scholar

• 9

  Casas-HuamantaE. R. (2022). Acceso a recursos tecnológicos y rendimiento académico en tiempos de pandemia y aislamiento social obligatorio. Rev. Cient. Sist. Inf.2, e296. 10.51252/rcsi.v2i1.296

  • CrossRef
  • Google Scholar

• 10

  Chacón-CuberosR.Serrano-GarcíaJ.Serrano-GarcíaI.Castro-SánchezM. (2025). Self-Concept modulates motivation and learning strategies in higher education: comparison according to sex. Educ. Sci.15, 873. 10.3390/educsci15070873

  • CrossRef
  • Google Scholar

• 11

  ChiuT. K. F.FalloonG.SongY.WongV. W. L.ZhaoL.IsmailovM. (2024). A self-determination theory approach to teacher digital competence development. Comput. Educ.214, 105017. 10.1016/j.compedu.2024.105017

  • CrossRef
  • Google Scholar

• 12

  DeciE. L.RyanR. M. (1985). Intrinsic Motivation and Self-Determination in Human Behavior. Boston, MA: Springer US. 10.1007/978-1-4899-2271-7

  • CrossRef
  • Google Scholar

• 13

  ElizondoK.ValenzuelaR.PestanaJ. V.CodinaN. (2023). Self-regulation and procrastination in college students: a tale of motivation, strategy, and perseverance. Psychol. Sch.61, 887–902. 10.1002/pits.23088

  • CrossRef
  • Google Scholar

• 14

  Fondevila-GascónJ. F.Marqués-PascualJ.Mir-BernalP.Polo-LópezM. (2019). Uses of WhatsApp in the Spanish university student. Pros. Cons. Rev. Lat. Comun. Soc.74, 308–324. 10.4185/RLCS-2019-1332

  • CrossRef
  • Google Scholar

• 15

  GarcíaF.y MusituG. (2001). Manual AF5 Autoconcepto. Madrid: TEA Ediciones.

  • Google Scholar

• 16

  JunaštíkováJ. (2024). Self-regulation of learning in the context of modern technology: a review of empirical studies. Interact. Technol. Smart Educ.21, 270–291. 10.1108/ITSE-02-2023-0030

  • CrossRef
  • Google Scholar

• 17

  KocsisÁMolnárG. (2025). Factors influencing academic performance and dropout rates in higher education. Oxford Rev. Educ.51, 414–432. 10.1080/03054985.2024.2316616

  • CrossRef
  • Google Scholar

• 18

  KryshkoO.FleischerJ.GrunschelC.LeutnerD. (2022). Self-efficacy for motivational regulation and satisfaction with academic studies in STEM undergraduates: the mediating role of study motivation. Learn. Individ. Differ.93, 102096. 10.1016/j.lindif.2021.102096

  • CrossRef
  • Google Scholar

• 19

  LiF. (2022). Impact of COVID-19 on the lives and mental health of children and adolescents. Front. Public Health10:925213. 10.3389/fpubh.2022.925213

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  Mena-GuacasA. F.López-CatalánL.Bernal-BravoC.Ballesteros-RegañaC. (2025). Educational transformation through emerging technologies: critical review of scientific impact on learning. Educ. Sci.15, 368. 10.3390/educsci15030368

  • CrossRef
  • Google Scholar

• 21

  MhlongoS.MbathaK.RamatsetseB.DlaminiR. (2023). Challenges, opportunities, and prospects of adopting and using smart digital technologies in learning environments: an iterative review. Heliyon9, e16348. 10.1016/j.heliyon.2023.e16348

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22

  MohamedM. G.Al-YafeaiT. M.AdamS.HossainM. M.RaviR. K.JaloF. M.et al (2025). The significance of emotional intelligence in academic stress, resilience, and safe transition from high school to university: an SEM analysis among northern Emirati university students. Glob. Trans.7, 109–117. 10.1016/j.glt.2025.02.003

  • CrossRef
  • Google Scholar

• 23

  NanningsM.van WeerdenburgM.van der ZandenP. J. A. C.HoogeveenL. (2025). Social–emotional and educational needs of higher education students with high abilities: a systematic review. Behav. Sci. (Basel)15, 819. 10.3390/bs15060819

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  NichollsJ. G. (1989). The Competitive Ethos and Democratic Education. Cambridge, MA: Harvard University Press.

  • Google Scholar

• 25

  OmirzakI.RalinA.KasatkinB.Vorona-SlivinskayaL.DubininaN. (2021). Students’ perception about the use of Mobile learning in solving engineering problems collaboratively. Int. J. Eng. Pedagog.11, 102–116. 10.3991/ijep.v11i6.24647

  • CrossRef
  • Google Scholar

• 26

  RadovićS.SeidelN.MenzeD.KasakowskijR. (2024). Investigating the effects of different levels of students’ regulation support on learning process and outcome: in search of the optimal level of support for self-regulated learning. Comput. Educ.215, 105041. 10.1016/j.compedu.2024.105041

  • CrossRef
  • Google Scholar

• 27

  Romero-SaritamaJ. M.Espinosa-TapiaS.Palacios-RodríguezA. (2025). Using WhatsApp in distance education: assessing the impact on academic interaction and influencing factors. Soc. Sci.14, 183. 10.3390/socsci14030183

  • CrossRef
  • Google Scholar

• 28

  SheldonK. M.HilpertJ. C. (2012). The balanced measure of psychological needs (BMPN) scale: an alternative domain general measure of need satisfaction. Motiv. Emot. 36, 439–451. 10.1007/s11031-012-9279-4

  • CrossRef
  • Google Scholar

• 29

  Simón-GrábalosD.FonsecaD.AláezM.Romero-YesaS.Fresneda-PortilloC. (2025). Systematic review of the literature on interventions to improve self-regulation of learning in first-year university students. Educ. Sci.15, 372. 10.3390/educsci15030372

  • CrossRef
  • Google Scholar

• 30

  SongP.CaiX.QinD.WangQ.LiuX.ZhongM.et al (2024). Analyzing psychological resilience in college students: a decision tree model. Heliyon10, e32583. 10.1016/j.heliyon.2024.e32583

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  StarkeS.LudvigaI. (2025). Unlocking digital potential—the impact of innovation and self-determined learning. Systems13, 396. 10.3390/systems13050396

  • CrossRef
  • Google Scholar

• 32

  Suárez-LantarónB.Deocano-RuízY.García-PeralesN.Castillo-RecheI. S. (2022). The educational use of WhatsApp. Sustainability14, 10510. 10.3390/su141710510

  • CrossRef
  • Google Scholar

• 33

  TimotheouS.MiliouO.DimitriadisY.SobrinoS. V.GiannoutsouN.CachiaR.et al (2023). Impacts of digital technologies on education and factors influencing schools’ digital capacity and transformation: a literature review. Educ. Inf. Technol.28, 6695–6726. 10.1007/s10639-022-11431-8

  • CrossRef
  • Google Scholar

• 34

  United Nations (2025). Goal 4: Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all. Available online at: https://www.un.org/sustainabledevelopment/education/(Accessed December 28, 2025).

  • Google Scholar

• 35

  Velásquez-PérezY.Rose-ParraC.Oquendo-GonzálezE. J.Cervera-ManjarrezN. (2023). Inteligencia emocional, motivación y desarrollo cognitivo en estudiantes. Cienciamatria9, 4–35. 10.35381/cm.v9i17.1120

  • CrossRef
  • Google Scholar

• 36

  Villa-EncisoE.García-MosqueraJ.Valencia-AriasA.Medina-ValderramaC. J. (2023). Exploring the role of Latin American universities in the implementation of transformative innovation policy. Sustainability15, 12854. 10.3390/su151712854

  • CrossRef
  • Google Scholar

• 37

  ViloriaA.Pineda LezamaO. B.Mercado-CaruzoN. (2020). Factors that describe the use of digital devices in Latin American universities. Procedia Comput. Sci.175, 127–134. 10.1016/j.procs.2020.07.021

  • CrossRef
  • Google Scholar

• 38

  WangC. K. J.ReeveJ.LiuW. C.KeeY. H.NgB.ChuaL. L.et al (2025a). An autonomy-supportive intervention program for STEM teachers to enhance engagement among students. Heliyon11, e42150. 10.1016/j.heliyon.2025.e42150

  • CrossRef
  • Google Scholar

• 39

  WangY.ZhaoY.ZhangJ.ManK.LuJ. (2025b). Fostering autonomous motivation in first-year college students: a self-determination theory-based online intervention. Learn. Motiv.90, 102118. 10.1016/j.lmot.2025.102118

  • CrossRef
  • Google Scholar

• 40

  YuX.YangD. (2024). The influence of Mobile technology on STEM education student learning outcomes. Int. J. Interact. Mob. Technol.18, 37–50. 10.3991/ijim.v18i20.50837

  • CrossRef
  • Google Scholar

• 41

  YangY.ChenJ.ZhuangX. (2025). Self-determination theory and the influence of social support, self-regulated learning, and flow experience on student learning engagement in self-directed e-learning. Front. Psychol.16:1545980. 10.3389/fpsyg.2025.1545980

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  ZamiriM.EsmaeiliA. (2024). Strategies, methods, and supports for developing skills within learning communities: a systematic review of the literature. Adm. Sci.14, 231. 10.3390/admsci14090231

  • CrossRef
  • Google Scholar

• 43

  ZimmermanB. J. (2002). Becoming a self-regulated learner: an overview. Theory. Pract.41, 64–70. 10.1207/s15430421tip4102_2

  • CrossRef
  • Google Scholar