*Article* **Students' Emotions and Engagement in the Emerging Hybrid Learning Environment during the COVID-19 Pandemic**

**Elizabeth Acosta-Gonzaga 1,\* and Elena Fabiola Ruiz-Ledesma <sup>2</sup>**


**Abstract:** Due to the COVID-19 pandemic, classes in schools acquired a hybrid learning model. Students took their classes both in person and, at other times, remotely. However, students are currently facing situations that they are not familiar with after a period of two years of confinement due to the COVID-19 pandemic. This emerging model of learning, to which the students had to adapt, not only impacted on their emotions during learning but also influenced their perceptions of their abilities and skills in being able to perform adequately in a situation of uncertainty, which also influenced the degree of academic engagement that they had. This study applied the structural equation modeling technique, using PLS-SEM software, to a sample of 194 students. The results show that their self-efficacy to act in a situation of vulnerability was affected, which is why their negative emotions increased and their positive emotions decreased. This in turn influenced the degree of engagement and effort they invested in developing a school activity.

**Keywords:** self-efficacy; emotions; academic engagement; hybrid learning; higher education; COVID-19 pandemic

### **1. Introduction**

During the pandemic, classes in schools were taught using a hybrid learning model, which involved students receiving classes that fluctuated between face-to-face and online periods. This model, which is called a hybrid [1], consists in dividing the total number of students into two blocks. The first block of students attended school to take their classes in person, while the second worked remotely. This hybrid model is an emerging adaptation that had to be implemented given the sanitary conditions [1].

This model is also called emergency remote education, according to Hodges et al. [2], as it was an emergency response to a situation in which the main measure was confinement; therefore, its effectiveness cannot be compared to other educational processes that were designed from the very beginning to be provided online [3]. The main difference is that e-learning or online education was created so that students could build their knowledge based on their individual learning experiences, supported by electronic media [4].

The hybrid learning model with which IPN higher education institutions have worked in Mexico City is an emerging adaptation in which the students of a group coexist simultaneously in different spaces [5]. For one week, half of the group studied in the physical classroom, and the other half was connected to classes in a virtual room; in the following week, the situation was reversed.

During this unexpected situation at the time of the COVID-19 pandemic, the students were affected academically, financially, physically, and psychologically [6].

Faced with the situations presented by this emerging hybrid environment, the students were caught up in a carousel of emotions, both positive and negative [7,8]. The negative emotions included anxiety [8,9] about the possibility of catching COVID-19 while traveling

**Citation:** Acosta-Gonzaga, E.; Ruiz-Ledesma, E.F. Students' Emotions and Engagement in the Emerging Hybrid Learning Environment during the COVID-19 Pandemic. *Sustainability* **2022**, *14*, 10236. https://doi.org/10.3390/ su141610236

Academic Editors: Zubair Ahmad and Noora J. Al-Thani

Received: 28 June 2022 Accepted: 4 August 2022 Published: 17 August 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

between home and school or at the school itself [10] and anger due to not being able to hear or take the entire class during the weeks in which they were learning remotely because of the failures of the school's internet connection [9,10]. Another negative emotion that the students may have experienced was boredom during online classes due to the aforementioned technical problems related to the lack of a fast internet connection [11].

Fortunately, positive emotions also arose, such as the joy of obtaining a satisfactory grade or the hope of reaching a planned level of success and the pleasure of seeing their classmates in physical form again [12]. These emotions play a fundamental role in student motivation, learning, and performance [13–16] and, crucially, in the degree of engagement with their learning [17].

This emerging model of learning, to which the students had to adapt, not only impacted their emotions during learning but also influenced their perceptions of their abilities and skills in being able to perform adequately in a situation of uncertainty; we are all born with the need to feel effective in our environment [18].

This raises the following questions: to what extent did the hybrid model of learning affect the students' efficacy in performing their schoolwork and what emotions prevailed? To what extent did self-efficacy and emotions affect school engagement?

Therefore, this study explored how the students' degree of self-efficacy when performing in a situation of uncertainty may have influenced their emotions and their degree of engagement with their studies. This research is important as it provides empirical evidence that analyses the effects of the COVID-19 pandemic on the essential constructs for school success, such as academic feelings, self-efficacy, and academic engagement.

#### *1.1. Academic Emotions, Engagement, and Self-Efficacy*

Several theories have been proposed to study the interrelationships between motivation, engagement, and self-efficacy, and two of them have been given particular attention. The first is the control–value theory by Pekrun [19], which provides an approach to exploring emotions in achievement environments such as the academic context. This theory implies that students' emotions can be affected by fostering their perceptions of competence and control. It is beneficial for schools because shaping educational environments properly can contribute to changing emotions in such a way that students know how to use them to their advantage.

The second theory is the expectancy–value theory of motivation by Eccles [20] and Pintrich [21], which analyzes the role of motivation in the performance of goals or tasks. The authors propose two main components as the foundation of this theory. The first plays a crucial role in determining school performance and refers to students' beliefs about their ability to perform their schoolwork and whether they have control and are responsible for their performance. This component is called expectative and includes self-efficacy and beliefs about the control of their learning. The other component is called value and includes extrinsic and intrinsic motivation towards an objective and the value the student gives to a task.

The emotional state of students is directly related to the involvement that they have with their learning activities. Emotions are feelings towards a situation, thought, or person, which are distinguished by being real or unreal [22]. Among the positive emotions that students can experience are fun, enthusiasm, enjoyment, satisfaction, and vitality. Negative emotions can include boredom, frustration, anxiety, depression, or anger. Emotions during the learning process play a fundamental role in students' motivation, learning, and school performance [13–15] and in the level of engagement that they show with their education [17]. In this study, positive emotions included enjoyment, enthusiasm, amusement, pride, and interest. Negative emotions included boredom, disinterest, frustration, sadness, and anxiety.

In defining school engagement, first Skinner and Belmont [23], then Fredricks et al. [24], and later Reschly and Christenson [25] found that academic engagement is a meta-construct that includes emotional, behavioral, and cognitive engagement. Regarding emotional

engagement, the affective, positive, or negative responses that students manifest in the development of academic tasks are evident [24]. The opposite behavior to engagement is the lack of interest, disaffection, or detachment that students may have towards academic work. When a student presents this kind of behavior, they are capable of manifesting negative emotions such as frustration, boredom, anxiety, depression, or anger, as well as passivity, distraction, or mental disconnection [18].

Regarding behavioral engagement, González et al. [26] pointed out that this refers to the time, attention, and effort that students invest in their academic process, that is, in the resolution of assigned tasks and in their learning environment in general. On the other hand, Reschly and Christenson [25] and Reschly [27] commented that behavioral engagement is the level to which a student participates in academic, social, or extracurricular tasks involving attention, concentration, and persistence [18].

Cognitive engagement is the mental effort that students put into their learning process [26]. Student engagement is high when using sophisticated cognitive strategies, such as self-regulation or metacognition [28]. Some metacognitive actions involve learning to plan, monitor, and regulate learning, according to what was pointed out by Ramirez-Arellano et al. [29]. Planning allows students to have knowledge about the way in which the problem will be addressed, while monitoring helps students to understand the problem and monitor its evolution and finally regulate or control their learning [30].

Martin et al. [31] commented that one of the primary factors in understanding the level to which students have to get involved and take responsibility for their educational work is motivation. There are several authors who have analyzed this relationship and verified the effects of motivation on academic engagement [32]. In the case of Wang and Eccles [33], they confirmed that motivation affects student engagement, which they divided into emotional, behavioral, and cognitive engagement.

In the expectancy–value theory, [20,21,34] have proposed that within the motivation component, self-efficacy is an essential element in determining school performance and refers to the students' beliefs about their ability to accomplish school tasks. In this regard, Sun et al. [35] found that there is also a positive relationship between self-efficacy as a component of motivation and academic performance.

Self-efficacy is defined as the confidence that a person has towards their ability to achieve an objective in a specific situation, and it is also considered fundamental for a student to achieve school goals and to be able to face the different situations that arise during their life [36].

In the school environment, the effects demonstrated by a student's self-efficacy in meeting goals are crucial as the perception that the student has about their capacity and ability to solve the situations that arise is decisive [37]. Likewise, the control that students have over emotional reactions to success or failure is a strong predictor of effort and persistence in student activity [18].

It has been found that self-esteem, self-efficacy, and expectations are some of the elements that could be involved in student engagement, which in turn would have a considerable effect on the quality and level of engagement [38]. On the other hand, Skinner et al. [18] pointed out that every person innately has the need to feel effective within the environment in which they work, and the more that students, within their school environment, feel this sense of dominance, the better the quality of their involvement is.

Even today, inconsistent results are shown regarding the effects of self-efficacy and academic engagement in hybrid and face-to-face environments [39].

This section provides the theoretical foundation on which this research proposes a hypothetical model (see Figure 1) to study the relationships between self-efficacy, emotions, and academic engagement. The theoretical bases for each relationship are presented in the following section.

**Figure 1.** Hypothetical model of causal relationships.

#### *1.2. Related Work*

This section presents previous research concerning the effects of self-efficacy and emotions on students' engagement with their schoolwork.

#### Self-Efficacy and Its Relationship with Emotions and Academic Engagement

Within the academic field, it has been found that emotions during learning are positively related to motivation, learning strategies, and academic performance [40]. Other researchers have found that positive or negative emotions directly impact cognitive and learning strategies differently. For example, if students feel amused or curious about what they are learning (positive emotions), then they will use learning strategies such as self-regulation and critical thinking [14,40].

Similarly, it was found that fun (a positive emotion) is related to motivational beliefs and self-regulation strategies, which in turn were shown to have effects on the level of current and future engagement that the students demonstrated [17]. Another study also suggested that the effects of fun are essential for the adoption of online assessments for mathematics instruction [41].

Thus, positive academic emotions impel students to feel more engaged in schoolwork [42]. This was also observed in the study by Carmona-Halty et al. [43], who surveyed high school students in order to observe the effects of academic engagement and its role as a mediator between students' emotions and performance. Their findings showed that academic engagement mediates the link between positive emotions and academic performance. In addition, the authors added that increasing students' positive emotions is a challenge for educators and parents.

In contrast, negative emotions minimize the use of self-regulation strategies and promote external guidance/help [14,44,45].

Similarly, the research of Tze et al. [45] showed that when students feel bored it impacts on their motivation and learning strategies, with the greatest effect on their performance. Likewise, the study by Marchand and Gutierrez [15] demonstrated evidence of the strong negative effect of anxiety on the use of learning strategies.

Therefore, negative academic emotions could negatively on students' achievements, making it more difficult for students to be engaged in their learning activities [46]. For example, in a longitudinal study on adolescents, Salmera et al. [47] analyzed the effects of self-efficacy as a component of personal resources and school burnout (as feelings of inadequacy) on academic engagement. Their results showed that self-efficacy had a positive relationship with academic engagement and a negative association with burnout. Additionally, one year after the beginning of this study, it was observed that academic engagement was negatively associated with school burnout.

King et al. [48], in a cross-sectional and longitudinal study, explored the effects of positive and negative emotions on academic engagement and its counterpart, academic disengagement. Their results proved that students with high levels of positive emotions were more engaged in their learning. Conversely, students with high levels of negative emotions reported higher disaffection. It was also observed that students who experienced positive emotions were more willing to strive for higher levels of engagement in their school.

Furthermore, in the context of the COVID-19 pandemic, through a longitudinal study Zhang et al. [49] explored the role of adaptability, academic emotions, and school engagement. Their results showed that academic engagement was positively correlated with positive emotions (such as enjoyment) and negatively correlated with negative emotions (such as anxiety, fear, and boredom).

Regarding motivation, it has also been found that there is a positive relationship between motivation and fun and a negative relationship between motivation and boredom [50]. Negative emotions decrease aspects related to intrinsic motivation. However, they increase extrinsic motivation, helping students in a certain sense, as this encourages them to avoid failure [14].

Another researcher, Pellas [51], who analyzed the effects that self-efficacy had on academic engagement in online education courses at a university, found that students who presented high levels of self-efficacy also reached high levels in the use of cognitive strategies and self-regulation. Likewise, it was found that self-efficacy was a significant predictor of the emotional engagement of those involved. As previously mentioned, it can be confirmed that self-efficacy in a given domain is essential to experiencing positive emotions, such as the enjoyment of that domain [17].

Likewise, other studies have found similar results, which explain that self-efficacy impacts on academic engagement [39] and learning performance [52]. Salmela-Aro and Upadyaya [47], when conducting a longitudinal study with adolescent students, also found that there is a positive linear relationship between self-efficacy and academic engagement; they point out that the level of self-efficacy that is attributed to a student influences the level of engagement and effort invested in developing an activity.

This was shown more clearly in the study by Heo et al. [39], who analyzed the relationship between self-efficacy and school engagement in students with and without symptoms of depression. The results showed a direct relationship between self-efficacy and school engagement in non-depressed students.

The research carried out by Borrachero et al. [53] analyzed the emotions of graduate teachers towards science teaching, finding that when teachers believed in their own ability to learn certain content, their positive emotions increased. Moreover, when teachers did not consider themselves capable of learning such content, their negative emotions increased.

#### **2. Materials and Methods**

#### *2.1. Data Collection*

A sample of 194 students from a higher education school was obtained, with ages ranging from 18 to 24. The students were pursuing a bachelor's degree in computer science engineering. The surveys were anonymous and answered by Google Forms (Google, Mountain View, CA, USA). All the students were invited to participate in answering the questionnaire; they were informed of its objective and that the data collected would be used exclusively for research purposes. The students' consent was requested to participate in the study, and all agreed to participate [54].

The questionnaire was distributed to the students on a regular day of classes; those in the physical classrooms answered on their mobile phones, and those taking classes online answered from home. It was applied at the end of March 2022, three months after starting their semester in the hybrid model, so that they could express their feelings about this model before returning to the face-to-face model; due to the sanitary conditions, they were to return to face-to-face classes in May 2022.

The sample size was five cases per variable observed in the low limit and ten cases per variable observed in the upper limit [55].

#### *2.2. Instruments*

The positive and negative emotions of the students were assessed using the Student Engagement and Disaffection in School (SED) instrument [18]. Positive emotions (PE) included enjoyment, enthusiasm, amusement, pride, and interest. Negative emotions (NE) included boredom, disinterest, frustration, sadness, and anxiety. The answers were evaluated using a 5-point Likert scale, where 1 meant totally disagree and 5 meant totally agree.

As the expectancy–value theory of motivation was included [20,21,34], we adopted the Motivated Strategies for Learning Questionnaire (MSLQ) instrument by Pintrich et al. [34,56]. This instrument includes self-efficacy (SE) as the motivational factor. Eight items were used and were evaluated with a 7-point Likert scale, where 1 was totally disagree and 7 was totally agree. Examples of self-efficacy for learning included "I'm certain I can understand the most difficult material presented in the readings for this course" and "I'm confident I can do an excellent job on the assignments and tests in this course".

Both instruments, the SED and the MSLQ, have been validated previously in the educational context [29,57,58].

To assess learning engagement (LE), the Schreiner and Louis [59] Engaged Learning Index instrument was implemented, using ten items. This instrument also evaluated the responses using a 5-point Likert scale, where 1 meant totally disagree and 5 meant totally agree. This instrument has been validated in previous research, such as in the study by [39], for example. To use this instrument, translation into Spanish was carried out and it was reviewed by a group of academics and a professional translator. When all the professionals had given their approval, the instrument was applied at the school where this study took place. This instrument included items such as "I can usually find ways of applying what I'm learning in class to something else in my life" and "I feel energized by the ideas that I am learning in most of my classes".

Likewise, as the instruments were measured with different Likert scales, all the factors were standardized [60].

#### *2.3. Analysis*

The proposed model was analyzed using the structural equation modeling (SEM) statistical technique, using the partial least squares method (PLS\_SEM) and using SmartPLS version 3 software (SmartPLS GmbH, Oststeinbek Germany). This technique is particularly useful for small sample sizes and makes better predictions when compared to the covariance-based technique CB-SEM [61]. In addition, the PLS-SEM technique does not require statistical assumptions such as data normality and sample size; so, it is considered a non-parametric SEM technique. The PLS-SEM technique has been used for both confirmatory and predictive research [62].

Utilizing a PLS-SEM technique is a two-tier process. The first tier involves calculating the measurement model or outer model. This model can include reflective or formative indicator variables. The difference between the two measurement approaches lies in the causal priority between the constructs and their indicator variables. In this study, reflective measures were used as the indicator variables of each construct are competitive with each other and represent manifestations of the construct [63].

The second tier consists in estimating the relationship among the constructs in the structural model or inner model. The links between the constructs are hypothesized according to theoretical reasoning [64].

As the measurement model evaluates the quality of the relationships between indicator variables and their underlying constructs, there were some criteria to validate. The construct

validity was verified using Cronbach's alpha (A); for many researchers, values greater than 0.70 are considered acceptable [65].

The composite reliability (CR) and the average of variance extracted (AVE) evaluated the convergent validity; values greater than 0.7 are considered acceptable for CR and 0.5 for AVE [66].

The discriminant validity is verified when the values of the diagonal that represent the square root of the AVE values of each construct are greater than the values below them, which represent the values of the correlations between the constructs [67].

The indicator reliability shows how much of the indicator's variation can be explained by the construct. Hulland [68] suggested that empirical studies can include loadings smaller than 0.4 in a PLS-SEM model or, as Hair et al. [66] mentioned, larger than 0.5.

The value of Cronbach's alpha, CR, and AVE, as well as the Fornell–Larcker discriminant validity criterion (shown later), are indicators of the goodness of fit of the proposed model [69].

The structural model was estimated by assessing collinearity issues, the path coefficient significance level, the level of determination coefficient R2, the effect size f2, and the predictive relevance Q2.

If two or more constructs (as independent variables) of the structural model are highly related, there is a multicollinearity issue, and it is observed when the variance inflation factor (VIF) coefficient is higher than 5 [66]. The goodness of the path coefficients was tested by the bootstrapping technique and t-statistics [70].

The R2 reflects the level of the construct's explained variance and can take values between 0 and 1. For social science research, the following thresholds are applied: weak (0.25), moderate (0.50), and substantial (0.75) [71].

The effect size f2 shows the change in R2 when a certain (exogenous) construct is omitted from the model. According to Cohen [72], the effect sizes of 0.02, 0.15, and 0.35 are small, medium, and large, respectively.

The Q<sup>2</sup> value assesses the predictive accuracy of the structural model. Stone [73] and Geisser [74] suggest using the Stone–Geisser test. The predictive relevance should be positive and with values higher than zero [75].

#### **3. Results**

#### *3.1. Descriptive Statistics, Reliability, and Validity*

The proposed model was analyzed using the structural equation technique, and the validity of the constructs was verified. The measurement model evaluated the construct reliability, convergent validity, discriminant validity, and indicator reliability. In order to achieve validity, five observed variables were removed from the learning engagement (LE) construct, and four observed variables from the negative emotions (NE) construct. No observed variables were removed from the self-efficacy (SE) and positive emotion (PE) constructs.

The observed variables and their factor loadings, including Cronbach's alpha, composite reliability (CR), and the average variance extracted (AVE) of the constructs involved in the proposed model, are shown in Table 1.

The reliability of each construct (construct reliability) was evaluated using Cronbach's alpha (A). Values greater than 0.70 were obtained; so, these values were considered acceptable. The convergent validity was also evaluated using the values of the composite reliability (CR) and the average variance extracted (AVE). Values greater than 0.7 for CR and 0.5 for AVE were observed; so, it was considered that construct validity was reached. The factor loadings show values higher than 0.4 [68], reaching indicator reliability.

Table 2 shows the discriminant validity of the measurement model, in which it is observed that the AVE values (on the diagonal) are greater than the values of the square root of the correlations between the constructs (the rest of the values). Therefore, discriminant validity was reached.


**Table 1.** Cronbach's alpha, composite reliability, and average variance extracted of the proposed model.

**Table 2.** Discriminant validity.


#### *3.2. Causal Model or Structural Model*

Figure 2 shows the results of the proposed model. These results demonstrate that the relationships between the constructs are statistically significant.

The results show that there is a direct and significant relationship between self-efficacy and academic engagement. Relationships between self-efficacy and positive emotions and self-efficacy and negative emotions were also observed. It was noticed that both emotions influence academic engagement. It was observed that the significance levels of the path coefficients were all significant (\*\*\* *p* < 0.001).

Hair [66] suggests that the VIF values should be lower than 5. The results show that the inner VIF values were NE (1.516), PE (1.399), and SE (1.443). Therefore, collinearity among the constructs was not an issue.

The determination coefficient R<sup>2</sup> of the construct learning engagement was 0.58, which is considered medium; for the negative emotions, it was 0.256, and for the positive emotions, it was 0.195, both of which are considered small [71]. This means that PE, NE, and SE were able to explain more than half the variance in the learning engagement construct.

**Figure 2.** Resulting causal model \*\*\* *p* < 0.001.

The effect sizes f2 of the structural model show the values of NE (0.407) and PE (0.202), which, according to Cohen [72], are large and medium, respectively. It was also observed that the SE construct had no effect on learning engagement as its value (0.015) was less than 0.02.

To evaluate the predictive accuracy of the structural model, the Q2 value of the endogenous constructs was used. The results show values of LE (0.289), NE (0.127), and PE (0.099), all of which are positive and higher than zero [75].

#### **4. Discussion**

During the health conditions imposed by the COVID-19 pandemic, in which an emerging hybrid teaching model was established, the manner in which self-efficacy influenced positive and negative emotions was analyzed, along with how these, in turn, have impacted the academic engagement of higher education students. Self-efficacy should be understood as the perception or judgement of students about their ability to complete school tasks.

Various studies have found the significant role that emotions play in the learning process [14,50,76,77]. The results show that self-efficacy has an impact on positive and negative emotions. As can be seen in Figure 2, there is a causal relationship between self-efficacy and negative emotions, which means that when students do not consider themselves capable of learning certain academic content, their negative emotions increase; that is, the low values of self-efficacy mean that students do not feel they have the skills and abilities to pass their subjects [53], which is why they feel anxious, stressed, distracted, and bored. This was also shown in the investigations by Sinatra and Taasoobshirazi [78] and Ramirez-Arellano et al. [29].

Additionally, it was found that there is a negative causal relationship between selfefficacy and the emotions of enjoyment, enthusiasm, and fun (positive). That is, when students do not believe in their own ability to learn the content of the subject, their positive emotions decrease, which could imply that students who feel less able to complete a school task have a decrease in their positive emotions and, therefore, that they might feel uninterested in the class [29]. As mentioned, it can be confirmed that self-efficacy in a given domain is essential to be able to experience positive emotions, such as the enjoyment of that domain [17].

What leads to the reflection on how the emerging hybrid teaching model has influenced the self-efficacy of the students during the uncertain conditions of the COVID-19 pandemic is the fact that they felt stressed, worried, and anxious about getting COVID-19 on their way from home to school and at the school itself. In the week during which students had to attend their online classes, feelings of anger or frustration were evident as a consequence of not being able to clearly hear their class due to failures caused by the school internet

from which the teacher transmitted. Likewise, the students felt bored in their online classes due to the same technical problems caused by an inadequate internet signal. Therefore, it can be assumed that the described problems interfere with the skills that the students must have to be able to attend and complete all school activities, consequently increasing their negative emotions.

The results also show the effects of negative and positive emotions on academic engagement. Research has been conducted that confirms that emotions predict academic engagement [51] and school performance [14]. Emotions are an essential aspect of the academic career of students, and they can vary over time [79]. During the health emergency, the students' emotions may have fluctuated due to the stress and anxiety caused by the environment of uncertainty about possible infections.

It has been proven that negative emotions are related to poor grades [80]; the fact that students experience emotions of stress, anxiety, worry, or boredom causes them to feel uncommitted to their learning, affecting their behavioral and cognitive engagement [81]. This was also observed in another study that showed that stress in students affected their degree of academic engagement [29]. Therefore, it is necessary to foster positive self-efficacy beliefs in students, which should be promoted to reduce negative emotions.

Conversely, if students feel enthusiastic and interested, these feelings cause a higher degree of engagement [53]. Therefore, it is vital to introduce motivating factors that promote enjoyment, enthusiasm, and fun in such a way that they manage to encourage a high degree of behavioral and cognitive engagement in students.

Other studies have found that self-efficacy influences the degree of academic engagement [39]. This research has also found a direct causal relationship between self-efficacy and academic engagement. When students show low levels of self-efficacy, this could imply that they are not very engaged with their homework and school activities. Conversely, high levels of self-efficacy have been shown to influence the degree of commitment of the students to their studies. Therefore, students who have a positive perception of the fact that they are capable of successfully carrying out their school activities will be more committed.

This was also observed in the research by Salmela-Aro and Upadyaya [47], who found a positive linear relationship between self-efficacy and academic engagement in adolescent students, pointing out that the level of self-efficacy that is attributed to a student influences their degree of engagement and effort invested in developing an activity. This implies that practices and academic activities that stimulate positive self-efficacy beliefs in students should be encouraged; this in turn will increase their motivation, performance, and skills when performing their school activities [58,82]. This encourages universities to apply support and tutoring models, especially for students in their first year at university, to develop skills such as autonomy, digital competence, and self-regulation [83].

#### **5. Conclusions, Limitations, and Future Research**

#### *5.1. Conclusions*

During the health emergency caused by the COVID-19 pandemic, universities had to implement an emerging hybrid teaching model, combining face-to-face classes and online classes. Within this new situation, students experienced feelings of uncertainty due to the fact that they were facing situations of possible contagion or technical problems due to the poor internet signal from the place where teachers transmitted classes. These problems affected the abilities and skills of the students in performing adequately in an uncertain situation. When students do not consider themselves capable of learning certain content, their negative emotions increase, and consequently, they feel anxious, stressed, distracted, and bored. Conversely, more motivated students feel more skillful in completing a school task; they feel interested in the class, enthusiastic, and even have a sense of fun. Emotions such as stress, anxiety, worry, and boredom cause students to feel less engaged in their learning. In response to this problem, the abilities and skills of students should be stimulated by providing the appropriate technical conditions and

continuing to promote preventive measures to avoid contagion, all with the aim of reducing their negative emotions in order to help build positive self-efficacy beliefs in the students.

It can be concluded that the conditions of uncertainty in the hybrid learning model, which was implemented due to the health emergency, have affected the level of self-efficacy that is attributed to a student. This, in turn, has influenced the degree of engagement and the effort that students invest in developing an activity.

#### *5.2. Limitations of the Research*

The study had the limitation that it was only carried out in an IPN engineering school because the other IPN schools continued the semester online due to a large number of infections in the city. Therefore, a small sample size was obtained.

#### *5.3. Future Research*

As a continuation of this study, future research could focus on factors related to the work of tutoring the students who showed little engagement with their schoolwork by monitoring the tutoring for a semester and measuring the impact at the end of it.

**Author Contributions:** Conceptualization, E.A.-G. and E.F.R.-L.; methodology, E.A.-G. and E.F.R.- L.; formal analysis, E.A.-G.; investigation, E.A.-G. and E.F.R.-L.; resources, E.A.-G. and E.F.R.-L.; writing—original draft preparation, E.A.-G.; writing—review and editing, E.A.-G. and E.F.R.-L.; supervision, E.A.-G. and E.F.R.-L.; funding acquisition, E.A.-G. and E.F.R.-L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Instituto Politécnico Nacional, grant number SIP20220747. The APC was funded by the Instituto Politécnico Nacional.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors would like to thank the Instituto Politécnico Nacional for their support of this research.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

