Assessment of factors associated with academic engagement and student burnout in the Moroccan context .

: Introduction: Student burnout syndrome is a major concern in higher education, associated with high prevalence rates. Its consequences include dropping out of school and adverse effects on mental and physical health. Despite this, the importance of academic engagement, a positive attitude towards learning, is often overlooked in the Moroccan context. Methods: This is a cross-sectional study conducted over a period of one month in 2023, and relies on semi-structured interviews to collect data. The measurement instruments used in this study include a self-questionnaire and two well-established scales: the Copenhagenan Burnout Inventory (CBI) and the Utrecht Work Engagement Scale (UWES-9). Results: The study reveals complex links between socio-economic factors, academic failure and burnout, with p values < 0.05. No significant relashionship was revealed between engagement and burnout (p>0.05). Female students showed a higher level of burnout than male students (p < 0.05). Age appears to be a determining factor, with younger students showing higher levels of burnout. Housing had an influence on burnout (p < 0.05) and differences between faculties showed a significant effect on burnout (p < 0.05). The internal consistency of the burnout and engagement scales are satisfactory. Conclusion: This study highlights the need for a holistic approach to addressing burnout in students. Targeted interventions are essential to support students in diverse academic contexts, taking into account a variety of socio-economic factors.


Introduction :
Based on the literature review, the field of education has increasingly shown interest in exploring the various aspects of academic engagement and its intricate connection with student well-being.In fact, the quality of an educational system is assessed based on both academic success and student well-being 1 , as these two elements are intertwined.As highlighted by the Organisation for Economic Co-operation and Development, educational institutions aim to achieve both goals.However, it's worth noting that "well-being" doesn't feature in global university rankings, unlike academic performance and its impact on student well-being [2][3] .This discrepancy often results in an emphasis on success and productivity at the expense of student well-being and health.Awareness of student health has recently gained prominence, particularly due to health concerns arising from the European Bologna reform process and its effects on students [4][5] .Furthermore, students find themselves in a unique phase of their lives marked by various changes in their social environment and lifestyle, especially in developing countries 5 .This phase often includes moving away from their family homes, establishing new social connections, and transitioning into the professional world alongside their academic responsibilities.
Drawing from the work of Bakker AB, two contrasting factors in the professional sphere concerning well-being or its absence stand out: work engagement and burnout 5,6 .Although these factors were initially explored in a professional context, they have also found relevance in educational systems, including schools and universities [7][8] .This expansion has spurred scientific research into their impact on student health and well-being.
In Morocco, there is a notable gap in scientific studies examining the connections between students' academic engagement and academic burnout, as indicated by the existing literature.It's important to note that Morocco's specific socio-cultural and academic context provides an intriguing framework for understanding how various elements come together to shape students' academic experiences and emotional well-being.By meticulously assessing these factors, our research aims to offer valuable insights that can inform educational policies and practices.
Within this context, our study's ultimate objective is to advocate for a comprehensive approach to education that takes into account the intricate interplay between academic engagement and student burnout within the Moroccan context.To achieve this goal, we first set out to evaluate the psychometric properties of each measurement instrument.Subsequently, we aim to identify the factors associated with academic engagement and burnout within the Moroccan university setting.

Type and population of study
It is a cross-sectional study conducted over one month in 2023.The sample was solicited through semi-structured interviews.We excluded students with psychiatric or neurocognitive problems.The sample under observation comprised students from various faculties in the Rabat-Sale-Kenitra region.

Auto questionnaire
This form collects data to segment the sample.It includes socio-demographic variables such as gender, age and number of siblings.Socio-economic factors such as housing, means of transport, family social class and source of funding are also taken into account.Origin (urban, rural) is the only socio-cultural factor.Academic factors include baccalaureate grade, post-baccalaureate orientation, satisfaction with this orientation, course pace, exam preparation and preparation method.

Copenhagean Burn out Inventory (CBI)
The CBI includes 24 items assessing different types of burnout: Personal burnout (9 items), schoolrelated burnout (3 items), colleague-related burnout (6 items), and teacher-related burnout (6 items).For example, Personal burnout concerns physical and psychological fatigue accumulated over the day, while Colleague related burnout refers to fatigue due to working with colleagues 9 .

Utrecht Work Engagement Scale (UWES-9)
The UWES instrument measures work engagement along three dimensions: vigor (VI), dedication (DE), and absorption (AB).The UWES-9 consists of 3 items for each dimension, measuring a positive and fulfilling work-related state of mind 10 .A high score indicates a strong commitment to studies.These measurement instruments have been translated and re-translated according to international standards 11 .

Analyse de données:
The analysis begins by identifying questionnaire items with low variability and response rates.Descriptive statistics examine the distribution of sample characteristics and responses for each item.Items with a response rate below 85% are considered, while highly correlated items (coefficient > 0.75) are removed.The prevalence of burnout in the study population is estimated from the frequencies.For analytical tests,we used chi-square test, analysis of variance (ANOVA), Kruskal-Wallis test and Mann-Whitney test.The internal reliability of scales is assessed by Cronbach's alpha coefficient and inter-item correlations.Good internal reliability exceeds an alpha of 0.6.Factor analysis (principal component analysis) tests the structure of the scales.The KMO value and Bartlett's test evaluate sample representation.

Résults:
Sixty-five of the participants were aged between 22 and 25 (57.5%),43 between 17 and 21 (38.1%), and the remaining five between 24 and 28 (4.4%).There were twenty-seven male students (23.9%) and ninety-eight female students (76.1%).Most participants were from urban areas (69%), while the remainder were from rural areas (31%).The first group was from the Faculty of Science (68.1%), the second group was from the Faculty of Economics (15%), followed by groups from the Faculty of Medicine (14.2%) and the Faculty of Humanities (2.7%).The percentages of participants by university level (bachelor's, master's, doctorate) were 82.2%, 7.1%, and 9.8%, respectively.a) Psychometric properties of the CBI scale : The overall reliability of the burnout scale is entirely satisfactory, with an alpha of 0.87 (Table 2).This scale comprises four subscales (PB, SRB, CRB, and TRB).For the PB subscale, the internal consistency coefficient was expressed with an alpha of 0.776.For the SRB subscale, the coefficient was 0.795.The CRB and TRB subscales expressed alpha coefficients of 0.800 and 0.834, respectively.The results of the first exploratory factor analysis were not satisfactory.After eliminating item 10, the analysis proposed four factors (kaiser Mayer; KMO = 0.77 and Bartlett Test: χ2(276) =1147.46,p<.0001).a) Relationship between burn-out and certain environmental factors : For the gender factor, an analysis using the Student t-test highlighted a significant effect of gender on Burnout scores.Female students reflected higher scores (M=48.42;SD=12.5)than male students (M=42.0;SD=13.4).Similarly, the Student's t-test highlighted a gender effect on the PB dimension (p<0.05).Descriptive analyses showed that female suffered a higher PB (M=57.6;SD=14.9)than male (M=46.9;SD=18.9).On the other hand, no significant difference was revealed between women and men in the CRB (U=1279; p=0.

a) Psychometric properties of the UWES-9 scale:
A first analysis based on the three-factor structure was unsatisfactory, as items 3, 4, and 8 did not saturate their factors.Similarly, a second analysis proposed a two-factor structure, with item 7 saturated on the first factor.The deletion of item 7 enabled a twofactor structure to emerge, with the first factor (VI/ DE) and a second (AB).This structure showed good internal reliability alpha= 0.814.

b) Relationship between Burnout and engagement:
The table below shows that correlations between the dimensions of the CBI burnout scale range from 0.20 to 0.41.Correlations between CBI dimensions and the total scale score range from 0.68 to 0.73.For the Engagement scale, correlational analysis between the two dimensions (VI/DE and AB) was expressed by a Spearman coefficient of 0.419.The correlation between the total scores of the two scales and the total score of the UWES reflected strong correlations (0.89 and 0.75).Given the relationships between the two constructs (CBI and UWES), the analyses showed low negative correlations between -0.30 and 0.-21 between the VI/DE dimension and the four dimensions SRB, CRB, and TRB.On the other hand, for the Absorption dimension, the analysis revealed a weak correlation with the TRB dimension (rho=0.207)and an absence of a relationship with the PB, CRB, and SRB dimensions.The correlations between the total score of the Engagement scale and the four burnout subscales suggest an absence of a relationship between engagement and the three dimensions PB, CRB, and SRB.Interestingly, the correlational analysis suggests that the score on the Study Burnout (SRB) dimension was significantly related to the UWES total score and the new VI/DE dimension.As mentioned above, the present study can be segmented into two main axes.The first is to adapt the instruments used to measure burnout and engagement to the Moroccan university context.The second is investigating the relationships between environmental factors, burnout, and engagement in Moroccan students.
It is worth mentioning that it is the first study to assess the reliability of the CBI-S instrument among Moroccan students.Our results suggest that the psychometric properties of the CBI-S were satisfactory.Regarding internal consistency, the high values found in the CBI-S subscales align with data presented in previously published reports [12][13][14][15] .
These results confirm those proposed by Kristensen  et al. (2005) and corroborate those of Compos (2013) 12 .In light of this, the CBI can be used in the Moroccan context to assess student burnout levels.
After the elimination of item 10, the factor analysis results showed the same factor structure proposed by previous work [12][13][14][15][16] (e.g., Kristensen et al., 2005;  Compos (2013).Indeed, item 10 showed low negative loading.It is possible that a stereotype could have been created because all items had the same response direction.In other words, our participants did not account for the inversion of the response scale for item 10.As a good match, the same behavior was similarly revealed in the study by Yeh et al. (2007)  12 when evaluating the psychometric properties of the Chinese version of the CBI and in the study by Compos (2013) 16 when conducting an exploratory factor analysis on Brazilian and Portuguese samples of the CBI.It is also worth mentioning that only items 3 and 17 reflected factor loadings lower than those of the item set.In support of the four-factor structure, four factors emerged with acceptable correlations.
Reliability and internal consistency analyses of the UWES-S scale were satisfactory.In terms of factor structure, an initial analysis based on the three-factor structure 17 showed an unsatisfactory structure.Items 3, 4, and 8 did not saturate their factors 17 .In this vein, the study by Willmer et al. 18 indicates that the factor analysis revealed an overlap in item loadings between the three factors.In a second analysis, the results suggested two factors with saturation of item 7 on the first factor (VI).A third analysis, based on the deletion of item 7, revealed a factor structure with two distinct factors.The first factor groups the clean items (VI/DE) and the second (AB).This structure showed good internal reliability alpha= 0.814.As mentioned earlier, a recent review of the UWES factor structure showed inconclusive results, with some studies suggesting a better fit to a one-factor structure, others proposing an excellent fit to a threefactor structure, and others indicating a fit for a twofactor structure 19 .For example, one of the studies found poor fits for the one-factor and three-factor structures of the UWES-9 20 .
In good alignment, Kulikowski's 19 study proposed a two-factor structure, merging dedication and vigor into a single factor, and absorption constituted the second factor.In line with Kulikowski's results 19 , this was only possible after the elimination of item 7.
The results of the correlational analyses indicated an absence of a relationship between the total scores of two scales, CBI and UWES.On the other hand, some negative correlations were also highlighted between the engagement and burnout subscales.Indeed, the literature has widely discussed the relationship between engagement and burnout.A first conception of such a relationship was proposed by Maslach and Leiter (1997) 21 .According to these authors, burnout is characterized by exhaustion, depersonalization/ cynicism, and lack of a sense of personal accomplishment, whereas engagement refers to energy (as opposed to exhaustion), involvement (as opposed to depersonalization/cynicism), and a sense of autonomy, and effectiveness (as opposed to lack of a sense of personal accomplishment).As a result, these authors considered low burnout rates indicative of a relatively high level of commitment.On the other hand, other research has shown that the two concepts are separate 22 .This latter proposal is in line with the study by Schaufeli and Bakker 17 , which suggests two negatively associated independent dimensions.Thus, Schaufeli and Bakker argued that the notion of burnout and engagement being opposed is erroneous and that they are two distinct constructs that need to be measured separately.Similarly, Crawford, LePine, and Rick's 23 meta-analysis indicates that the two constructs are not antagonistic.Although our correlational analyses support good negative relationships between the "VI/DE" subscale and the burnout subscales, the correlation results between the UWES-S and CBI scales suggest a clear separation between the two constructs.The present study showed that 38.9% of students had symptoms of burnout.
The burnout prevalence appears to be much higher than that found in previous studies [24][25][26][27] and slightly higher than that proposed by a study conducted among medical students in Pakistan (30.6%) 24 .On the other hand, the prevalence of burnout seems to be lower than that found among students in several countries (between 45 and 70%) 28 .Indeed, the differences observed in burnout prevalence's across studies may be due to cultural and socio-economic differences in the study population.In good agreement with previous studies 26 , the present results suggest that burnout was significantly affected by gender, years of study, housing, means of transport and academic orientation.Because of the consequences of burnout syndrome on students' physical and psychological well-being, properly calibrated measurement and diagnostic devices for the target population becomes necessary.Our research contributes to examining the psychometric qualities of the CBI-S for a population of Moroccan university students and determining the prevalence of burnout in the university environment.
The results show that the CBI for students is a good instrument for assessing the occurrence of the syndrome and, consequently, a valuable tool for specialists in health assessment and psychology at the university.

Conclusion
This study investigated burnout and engagement in Moroccan students.The preliminary results of the psychometric analyses confirm the validity of the instruments used in this specific context.The findings on burnout prevalence and associated factors are valuable information for mental health and education professionals.This initial work in the Moroccan university context encourages the opening of a path for further investigations to better apprehend the complex relationships between these concepts and their determinants.

Conclusion:
This study investigated burnout and engagement in Moroccan students.The preliminary results of the psychometric analyses confirm the validity of the instruments used in this specific context.The findings on burnout prevalence and associated factors are valuable information for mental health and education professionals.This initial work in the Moroccan university context encourages the opening of a path for further investigations to better apprehend the complex relationships between these concepts and their determinants.

Table 1 . Factor analysis of the burnout scale (CBI)
For the gender factor, the results show that 40.7% of women suffer from moderate burnout, while 59.3% have low/No burnout.For men, 33.3% have moderate burnout, and 66.7% have low/No burnout.For the age factor, the results show that in the 17 to 21 group, KMO=0.77, Bartlett Test: χ2(276) =1147.46,p<.0001 a) Descriptive statistics and prevalence of burn-out : have low/no burnout.For students from the Faculty of Medicine, 25% suffer moderate burnout, and 75% have low/no burnout.Of the 77 students in the Faculty of Science, 42.9% have moderate burnout, and the remaining 57.1% have No/low burnout.Finally, for the Faculty of Human Sciences, 33.3% suffer moderate burnout, and 66.7% have low/No burnout.

Table 2 .
Prevalence of burnout

Table 4 . Pearson and Spearman correlations between UWES and CBI dimensions
We have respected ethical committee approval.