Prevention of Children’s Burnout at School through the Tutoring and Guidance Process. A Structural Equation Model Analysis

: The school burnout of children, deﬁned as physical and mental exhaustion due to a lack of adjustment to the educational context, constitutes a serious problem in contemporary education. Thus, the determination of the elements that inﬂuence it and the possible strategies for avoiding it are key in the process of improving children’s well-being. A descriptive and cross-sectional study was conducted with a sample of 569 children aged 8 to 13 at eight primary education schools in Granada (Spain). With the aim of examining the association between the possible exhaustion of children and planned tutoring and guidance activities, a structural equation model (SEM) as the analytical technique was used. The results show good ﬁt indices for the model (comparative ﬁt index (CFI) = 0.955; normalised ﬁt index (NFI) = 0.956; incremental ﬁt index (IFI) = 0.946; root mean squared error approximation (RMSEA) = 0.089), which reﬂects the need for tutoring and guidance activities that are infused throughout the entire teaching process in order to preserve children’s well-being.


Children's Burnout
Recent classifications consider burnout as a work-related illness that is associated with stress [1]. It is characterised by a lack of motivation and mental or emotional exhaustion, dehumanisation and feelings of low personal achievement [2,3]. Concretely, [4] defines burnout as a lack of physical and mental energy that is associated with cynical attitudes and feelings of inadequacy in the context within which one develops. Specifically, [4] establishes three basic dimensions linked to burnout, among which are cynicism, dishonest attitudes associated with the situation experienced; exhaustion, a feeling of fatigue and lack of energy; and inadequacy, a feeling of a lack of correspondence between practices, the demanded task and personal motivations [5,6].
This state of exhaustion has been studied within a number of different contexts such as health, professional work or higher education [5,7]. The educational ambit has taken on great relevance in recent years, with a special emphasis being placed on teaching perspectives [8]. Nevertheless, school failure and dropout are elements that are also present at all levels of the educational system when taking the teaching perspective. This is because, on a large number of occasions, students must rise to academic demands despite perceiving themselves as lacking the competence to tackle them [9,10]. Indeed, both in the work and school settings, the reasons for this tend to be traditionally associated

School Burnout Prevention
In this sense, the solutions employed by governments, authorities and educational institutions to tackle exogenous challenges typically refer to palliative or corrective measures that target a problem that already exists. Such measures are delivered via social support programmes and scholarships or financial aid [17]. By contrast, issues that are endogenous in nature are addressed through academic guidance programmes and educational support [18,19].
Nevertheless, whilst all of these actions are necessary, they are insufficient if they are not complemented by other preventative actions that prevent or, at least, reduce the likelihood of these problems occurring [20,21]. For this reason, the present study focused on a search to identify solutions to the endogenous causes associated with the guidance and tutoring of children in formative processes. These acts transcend those that are merely academic and curricular, facilitating the integral training of school aged children and helping to prevent the emergence of children's burnout [3,22]. The study [23] showed that social competence is essential for finding an association between school engagement and students' well-being. In research on social capital and school exhaustion, it was shown that school social capital, as well as positive and supportive relationships between students and teachers, may produce less school exhaustion and positively influence academic performance [24].
Thus, the first idea is the recognition of burnout as a school problem that affects both pupils and the teaching process itself [25]. Specifically, a lack of well-being is established in the school context that provokes exhaustion in learning situations, the rejection of opportunities to study and feelings of incompetence or the lack of ability to adapt to these situations [4,26]. Furthermore, it is a complex problem that can only be approached from transversal standpoints, which tackle problems in a multi-dimensional way and offer multi-disciplinary solutions, such as student guidance and tutoring [27,28].
Secondly, a new way of guiding students based on global tutorial action must be considered. This tutorial action of teachers encompasses all actions that act synergistically with and in parallel to the development of the school curriculum, with the educational action being formed in concert with students [29,30]. Likewise, the relevance of the guidance and tutoring given to children before, during and after the educational process is revealed as one of the fundamental elements in this process.
It must bring diverse perspectives together in a multi-dimensional way whilst considering the complex personal, academic and social reality of the developing scholar [22,31].
The orientation actions performed through tutoring must transcend those that are considered from a more traditional perspective, which include a time-limited dedication to the resolution of minor co-habitation problems or the communication of examination results to families [3,27]. Likewise, a central dedication of teaching time and effort must be directed towards facilitating the personal development of students, alongside their academic learning and skills, for social and workplace integration [29,32]. Finally, it is indicated that this process of tutorial action must tackle children's well-being, school satisfaction and emotional well-being [27,33]. In this line, the development of the tutorial action and orientation in the educational context and from a holistic perspective should be oriented to the process of help and accompaniment in seven basic aspects: the learning process, dedication and involvement, co-existence, cooperation, socialisation, emotions and satisfaction.

Purpose of the Study
The present study raises as a research problem the existence of high levels of burnout and mental exhaustion in schoolchildren, which are due to inadequate teaching practices, a lack of methodological resources and a poor process for tutoring students. The aim of this research was to develop a structural equation model based on obtained empirical data in order to provide an explanation of the associations produced between the development of tutoring in the educational context and the different indicators of children's burnout. Based on these premises, the present study formed the following hypotheses: Hypothesis 1. The guidance of the learning processes, emotional management and coexistence will be the most relevant indicators in tutoring.

Hypothesis 2.
Those students who are tutored in the educational context will show less development of cynicism, exhaustion and inadequacy linked to burnout.

Materials and Methods
The study design was non-experimental and ex post facto. It was descriptive and transversal in nature, with measurements being taken at a single time-point and from a single group. The entire research process proposed, the procedures and the use of information collection instruments were submitted for the approval of the Research Commission of the Department of Research Methods and Diagnostics in Education (University of Granada, Granada, Spain).

Population, Sampling and Participants
The study sample was made up of Compulsory Primary Education students between the ages of 8 and 13, from eight public schools in different locations in the province of Granada (Spain) that provide Early Childhood and Primary Education and are similar in terms of student population. Non-random convenience sampling was performed in two stages. In the first stage, the nature of the research project was explained to various schools as a part of an educational programme run by the research team, with the purpose of requesting permission from the school councils with the representation of educational authorities, teachers and families. Likewise, in a second stage, parental permission was requested, to consider only those groups (classes) where there were no objections to collaborating in the study. Sample selection was to include primary education students from the fourth to sixth courses with the intention of only involving individuals who were mature enough to adequately utilise the data collection instruments.

Data Collection Instruments
Prior to commencing the study, in order to check suitability for the target students, the instruments were presented to a selected group of individuals who did not participate in the final study. This permitted confirming the understanding of the items and the response scale. The instruments used were:

1.
A self-registration sheet, which collected demographic information such as gender, age and school year, alongside tutoring characteristics such as the weekly dedication of hours to tutoring and the type of activities addressed, collected via broader instruments [22].

2.
A school burnout inventory [4], adapted into Spanish by [34]. This instrument is composed of 9 items that are rated along a Likert-type response scale with 5 response options. The responses run from completely disagree to completely agree. Its internal structure groups burnout according to three dimensions: exhaustion (items 1, 4, 7 and 9), cynicism (items 2, 5 and 6) and insufficiency (items 3 and 8). Its application in the present study produced adequate internal consistency, with a Cronbach's α of 0.746.

Procedure
The research process conformed to the ethical principles relating to this type of research and followed recommendations by [35,36]. The process followed the guidelines set by the ethics committee of the University of Granada for this type of research. The access to and selection of the sample, the use of the information collection instruments, and the analysis procedures were submitted for the approval of the Research Commission of the Department of Research and Diagnostic Methods in Education.
The access to the research scene began with a preliminary stage in which the study centres were provided with an explanation to the school councils of the purpose of the research, the procedures to be followed and a commitment to provide them with a summary report on the results.
Once permission to access the schools was granted, interviewers who were adequately trained in the use of the instruments due to being members of the research team explained the instruments to the students. At the same time, they were on hand to resolve doubts whilst proceeding with the administration of the data collection instruments. Finally, they checked the quality of the collected data before finalising the data collection processes.
Once the data had been analysed and a research report adapted to the audience had been developed, an explanatory report of the results and educational recommendations was sent to the participating schools. Figure 1 shows the path model constructed. In the first instance, the endogenous variables in the model were determined, with which error terms were associated. One of the primary variables was Overall Tutorial Action (AT-G), which is an endogenous variable, and it comprises a number of different indicators. Amongst these indicators, which are observable and exogenous variables, we found dedication to tutoring of one hour a week (AT-DED), working on emotions in tutoring (AT-EMO), working on socialisation (AT-SOC), working on cooperation with classmates (AT-COO), the approach to learning challenges (AT-APR), the consideration of co-existence challenges (AT-CON) and students' satisfaction levels with the tutors (AT-SAT). On the other hand, variables linked to burnout acted as exogenous variables within the model. Amongst these variables, we found burnout-associated exhaustion (B-AGO), burnout-associated cynicism (B-CIN) and burnout-associated inadequacy (B-INA). Educ. Sci. 2020, 10, x FOR PEER REVIEW 5 of 12 To complete the construction of the model, error terms were added for the endogenous variables. It is worth mentioning that the one-way links show the possible associations between the embedded variables. The estimation of the parameters was carried out using the maximum likelihood method. Furthermore, with the objective of verifying the compatibility of the model, a fit analysis of the empirical data was performed. The adequacy of the model fit was checked according to criteria stated by [37], considering:

Results
According to the calculations carried out, a good fit was observed for all the indices analysed. The chi-squared statistic revealed a significant p-value (Chi Square χ 2 = 178.345; degree of freedom df = 32; p < 0.001), although it must be taken into account that this statistic does not have an upper limit, and thus, it cannot be interpreted in a standardised way. This also needs to be considered in light of the problems created by its aforementioned sensitivity to sample size. In consideration of these flaws, other indices of standardised fit were employed that are less sensitive to sample size. The CFI produced a value of 0.955, this being excellent. The NFI also specified an excellent value of 0.956, whilst the IFI was 0.946, this being acceptable. The analysis of the RMSEA obtained a value of 0.089. Table 1 and Figure 2 show the values produced for the relationships between the variables of the structural equation model. Considering the first level of the model, the regression weights can be observed for each indicator of the variable describing the overall tutorial action. Statistically significant differences were obtained in all cases (p < 0.005). In the case of this variable, it is observed that the variable with the strongest association describes the dedication to tackling learning problems through tutoring (AT-APR; b = 0.923). This is followed by the variable describing satisfaction with To complete the construction of the model, error terms were added for the endogenous variables. It is worth mentioning that the one-way links show the possible associations between the embedded variables. The estimation of the parameters was carried out using the maximum likelihood method. Furthermore, with the objective of verifying the compatibility of the model, a fit analysis of the empirical data was performed. The adequacy of the model fit was checked according to criteria stated by [37], considering:

1.
Chi-squared values with non-significant p-values as an indication of good fit.

Results
According to the calculations carried out, a good fit was observed for all the indices analysed. The chi-squared statistic revealed a significant p-value (Chi Square χ 2 = 178.345; degree of freedom df = 32; p < 0.001), although it must be taken into account that this statistic does not have an upper limit, and thus, it cannot be interpreted in a standardised way. This also needs to be considered in light of the problems created by its aforementioned sensitivity to sample size. In consideration of these flaws, other indices of standardised fit were employed that are less sensitive to sample size. The CFI produced a value of 0.955, this being excellent. The NFI also specified an excellent value of 0.956, whilst the IFI was 0.946, this being acceptable. The analysis of the RMSEA obtained a value of 0.089. Table 1 and Figure 2 show the values produced for the relationships between the variables of the structural equation model. Considering the first level of the model, the regression weights can be observed for each indicator of the variable describing the overall tutorial action. Statistically significant differences were obtained in all cases (p < 0.005). In the case of this variable, it is observed that the variable with the strongest association describes the dedication to tackling learning problems through tutoring (AT-APR; b = 0.923). This is followed by the variable describing satisfaction with how tutoring is carried out (AT-SAT; b = 0.877) and tackling problems of co-existence in the classroom (AT-CON; b = 0.847). By contrast, the indicators with the weakest regression weights were dedication of one hour a week (AT-DED; b = 0.616), followed by cooperative working (AT-COO; b = 0.726) and emotional education (AT-EMO; b = 0.733).      Next, associations are shown between overall tutorial action and the different dimensions of burnout. A positive and direct association is shown between burnout-associated exhaustion and tutorial action (p < 0.005; b = 0.196), whilst a negative and indirect association is seen between burnout-associated cynicism and tutorial action (p < 0.05; b = −0.113). As a final point, no statistically significant differences were obtained between burnout-linked inadequacy and overall tutorial action (p = 0.382).
Finally, associations were determined between the different dimensions of school burnout, revealing positive and direct associations in all cases. The greatest regression weight was obtained for the association between cynicism and inadequacy (p < 0.005; b = 0.560), followed by the association between exhaustion and cynicism (p < 0.005; b = 0.497). The weakest association was that observed between exhaustion and inadequacy (p < 0.005; b = 0.465).

Discussion
The present study pursued the development of an explanatory model for the existing relationships between the development of tutorial action in primary education and its different indicators such as working on emotions, co-existence and learning strategies. This was examined in light of its relationship with the different dimensions of school burnout. In this way, the study sought to examine whether this holistic vision of the tutoring process and its integrated actions exerted a protective effect against the development of the aforementioned problem of mental health, which has emerged within the educational context. In this regard, some studies have taken a similar approach, such as those conducted by [3,10,37,38].
In reviewing the proposed model, it can be seen that the influence of the different indicators in relation to the proposed concept of tutorial action was determined at the first level. In this way, the indicator exerting the greatest influence on tutorial action is that which refers to tackling learning problems in tutorials and class groups. Logically, learning is a central question when developing teaching tasks and those relevant to tutoring [39]. Nevertheless, it serves to point out that the excessive centralisation of academic processes is a characteristic of more traditional educational models [40], and it is therefore necessary to include other indicators.
Children's satisfaction with respect to the way in which the tutor acts was the element that exerted the second strongest influence over the dimension constructed for tutorial action. This demonstrates the relevance of general subjective children's well-being and shows that they feel it is intertwined with the practices developed in the educational context [41,42]. Likewise, actions relating to this indicator enable an appropriate learning climate to be constructed and children's development to be based on a consideration of their emotions [43]. Furthermore, the factor with the third greatest influence on tutoring is tackling co-existence problems in the classroom through tutorial action and working on emotional education. This has the aim of finding solutions to problematic situations, as has been demonstrated by [26]. This indicator, together with that previously discussed, permits the development of what is considered to be a good school environment for children [22].
The indicator that exerted the least influence was dedication to specific tutoring sessions for one hour a week. This would also be predicted by a traditional model of tutorial attention [44]. Thus, it seems justified that a certain logical relationship still exists between the class or specific length of the tutoring process and the appreciation of the work of the tutor. However, in current models, all primary schools' teachers act as tutors, and, therefore, all class hours can be used to integrate tutoring activities into the teaching exercise. Given that the job of guiding and tutoring is of a transversal nature [29,32], it must be infused throughout all knowledge areas, classes and learning opportunities as a regulatory element of the learning process and children's welfare.
These premises are further endorsed by the negative relationship seen with burnout-linked cynicism. Thus, it is evident that when this dimension of tutorial action is enforced, attitudes of indifference or rejection towards the school and burnout in general are reduced, clearly demonstrating its preventative potential. Specifically, [3] demonstrated through multi-group analysis that young people who had not received tutorial attention possessed greater mental exhaustion and worse emotional regulation strategies. This might explain the findings in this respect found here. Likewise, [18,45] point to the relevance of exercising sufficient tutorial development that is ideographic in nature, in order to prevent the rejection and abandonment of education.
Finally, the tutorial action variable revealed a positive relationship with children's exhaustion resulting from burnout. Clearly, this relationship must be explained in some sense [46]. Children who present with burnout do not do so suddenly, but as a result of a process in which exhaustion is a typical characteristic [8]. Thus, this situation demands greater tutorial attention and actions when working on emotional control, learning, co-existence or aspects of socialisation [27]. To achieve this, this relationship should be understood as an alarm call to work profusely on these aspects in tutoring and prevent this aspect from forming elements that constitute children's burnout. In this way, teachers can be converted into effective and efficient tutors [45,47].

Conclusions
As the main conclusions, it was shown how tutoring in learning, satisfaction and coexistence are the three main indicators of tutoring. This reveals a partial fulfilment of Hypothesis 1, in which the emotional orientation was determined as more relevant. On the other hand, the existence of an inverse relationship between the development of cynical attitudes in the school-disapproval and a lack of honesty-and the development of tutoring in students was observed, showing how this helps in its prevention. It was also shown how an increase in exhaustion linked to burnout is associated with a greater development of tutoring and guidance actions, since their use helps in prevention. Nevertheless, the inadequacy in burnout was not related to tutoring, making Hypothesis 2 partially fulfilled.
The model presented in the present study shows that tutorial action is both a predictor of, and a protector from, children's burnout. A relationship was shown between the elements of tutorial action that have the strongest impact on the avoidance of children's burnout, amongst which attention to learning and co-existence problems most stood out. Nevertheless, an understanding of these factors should not only be considered in a corrective sense, in that they form the basis of action when the problem of burnout has already appeared, but should also be used to outline the elements upon which diagnostic care will be based. The entire student body must carefully observe and work on the elements discussed here, with a view to prevention to avoid this serious problem that can affect school-age children.

Limitations
It is essential to indicate the main limitations inherent in the present study. The first limitation is linked to the study sample, given that a representative sample of the population of Granada (Spain) was not employed. Nevertheless, it should also be pointed out that the number of participants was relatively high, and the selection of the sample implied a certain degree of randomisation, as it was conducted according to natural groups at schools, just as is recommended by the scientific literature. Furthermore, it should also be highlighted that the study design could constitute an intrinsic limitation, as it was not an experimental study. Causal relationships cannot be determined between the variables that formed the object of study. This being said, the analytical technique employed based on a structural equation model permits the links between variables to be effectively examined and compared to a theoretical model based on obtained empirical data. Finally, it is important to indicate that the scales employed could represent a third limitation. A very high Cronbach's alpha coefficient was not obtained despite the scales having been validated in populations with similar characteristics. For this reason, it would be interesting to conduct a review of these instruments with the aim of improving their adaptation to the school context.

Educational Implications
There are several implications that emerge from this study. Mainly, we highlight the importance that reflection on the integrated tutorial action in the teaching processes should acquire for teaching professionals. On the other hand, it is appropriate to emphasise tutorial action as a valuable tool for preventing situations of school burnout in Primary Education students, as well as helping to improve their learning tasks.
Thus, tutoring should focus on student learning as one of its main goals. This should not, however, be its only goal, nor should it solely draw on a merely academic content, as the integral development of individuals must transcend this. In this way, content based on aspects that refer to personal and social development could convert tutoring into a preventative model for children's burnout [48]. In this respect, tutoring and guidance play a fundamental role in the educational process [49]. This is because it provides the most appropriate setting and scenarios for the awakening of teachers and students to work together on aspects linked to emotions, satisfaction and a good classroom climate. At the same time, this enables an atmosphere appropriate for children's well-being to be generated, which, in both a direct and indirect way, can become a preventative element in relation to burnout and the emotional elements that construct it [3,9]. This will have a direct benefit for the academic performance of the students and contribute to their inclusion in the current information and knowledge society.

Prospective Actions
Further research in this field could include the adaptation and validation of the instrument in other educational stages where tutorial action is an essential factor (secondary, upper secondary and vocational training). In addition, experimental studies can be undertaken to verify what effects certain tutoring programmes have on the variables studied and thus obtain greater consistency in the relationships established in the present investigation. Finally, it is also intended to continue working on other relationships of variables that complete this perspective of tutoring as a key topic for enhancing learning processes. Some of the variables to examine for a relationship with tutorial action are academic performance, student motivation, self-concept, self-esteem, emotional intelligence and resilience. Funding: This research was carried out within the framework of the XXVI Program for supporting practical teaching, Vice-Rectorate for Teaching and Learning, University of Granada (Spain), grant number: 625/E/1.