In the academic year 2020/2021, 1,793,210 students were enrolled in Italian universities (
Ministero Dell’istruzione 2020), even though Italy is one of the European countries with the lowest percentage of young students who continue their studies. Despite the different academic paths and contexts, many students shared the same experience of higher education. The university student population has often been considered a privileged target group due to their age and living conditions (
Lederer and Oswalt 2017;
Lederer et al. 2019;
LeViness et al. 2019;
Lipson et al. 2018;
Schwartz and Kay 2009); however, in recent years, more attention has been paid to the physical and mental well-being (or lack thereof) of this population, both due to an increased focus on these issues and the broader topic of student services, and due to changes in the population itself. Thanks in part to policies promoting the right to education, students from different socioeconomic backgrounds are participating in academic life: however, given the high cost of academic study and information barriers (
Barone et al. 2017;
Contini and Scagni 2013), the difference in economic resources means that participation in academic life is not equal for all. At the same time, the phenomenon of globalisation and recent changes in the labor market, together with the widespread precariousness of work, reinforce the sense of insecurity even among those who are still studying, a perception that negatively affects their well-being (
Musumeci and Ghislieri 2020). Students face a weak labour market that often does not adequately employ them, even when they earn their degrees on time, at prestigious universities, and with very high grades. In other words, students perceive the labour market as extremely demanding and full of inequality and injustice (
Capone et al. 2020).
Apart from the systematic surveys on the well-being of university students, these aspects were highlighted in Italy by an episode that shook up public opinion, institutions and academics. In the summer of 2021, three female students from the Scuola Normale di Pisa, one of the most prestigious universities in Italy, gave a powerful and forceful graduation speech about the great opportunities offered by the education system, while being highly critical of the current neoliberal system, which exaggerates competition, emphasises differences, and thus exacerbates inequalities, with negative effects on students’ psychological health.
This and other powerful evidence led several researchers to question the experiences of university students, especially during the pandemic COVID-19 (
Schröpfer et al. 2021). The pandemic context and subsequent containment measures exacerbated an already delicate situation by introducing new elements into academic life: the reduction, if not complete negation, of social interactions in presence; the unprecedented massive reliance on distance learning that accompanied low digital skills among faculty and students; the need to use information and communication technologies (ICT) in every aspect of life, sometimes in inappropriate spaces and with less appropriate devices (
Amerio et al. 2020); the inability to access higher education resources, most of which were closed or had limited access anyway (
Lederer et al. 2021).
These profound changes have led many universities to establish and expand psychological support services for students and make them remotely accessible to meet the increasing demand for help in the wake of a drastic deterioration in student well-being. These requests have come from different sources, such as teaching staff, student services desks, and other entities that capture the student body’s needs.
Given this complex scenario, our study has two objectives. It aims to explore how Italian students perceived academic life during the pandemic and which variables contributed to emotional exhaustion and engagement by examining the relationship between specific variables, relevant during the COVID-19 pandemic, in a large sample of Italian college students during the second lockdown. The study is part of a larger data collection project aimed at expanding knowledge on this topic and identifying strategies and interventions at universities to improve quality of life (
Ghislieri et al. 2022b). Unlike other studies (e.g.,
Lipson et al. 2018;
Son et al. 2020), this work does not inquire about the presence of mental disorders in the population of college students but focuses on dimensions that contribute to or worsen quality of life in academia. Our study takes into account the framework of the Job Demands and Resources model (JD-R model;
Bakker and Demerouti 2017), which has been used in other studies to understand the dynamics of higher education students’ well-being (
Naylor 2022). Specifically, we considered emotional exhaustion (a dimension of burnout) and engagement, both adapted to the academic context. Engagement refers to a positive state of mind, characterised by vigour (energy and mental resilience), dedication (being strongly involved with a sense of significance, enthusiasm, and challenge) and absorption (being happily focused on activities) (
Schaufeli and Taris 2014). Indeed, engagement has been analysed from various perspectives, being considered a resource for subjective well-being but also as a consequence of subjective well-being or as a related variable (
Datu and King 2018). With respect to the possible determinants of exhaustion and engagement, we first considered a study demands (study load), alongside two dimensions particularly relevant during the pandemic period (
Giusti et al. 2021;
Salimi et al. 2021), namely reduced academic performance (RAP) and internet addiction (IA); we also considered academic self-efficacy (ASE) as a personal resource.
1.1. Higher Education Students during the Pandemic
In an international study conducted in 62 countries,
Aristovnik et al. (
2020) highlighted college students’ experience of apprehension and boredom during the first lockdown, along with a general appreciation for the role of universities and hospitals (relative to other institutions). In addition to the increasing need for support, several studies have highlighted a number of important risk factors: overuse of technology and lack of adequate digital skills for effective remote learning, leading to potentially poorer academic performance (
Giusti et al. 2021;
Salimi et al. 2021); the perception of a heavier workload, which is seen as a major source of frustration (
Tasso et al. 2021); concerns about one’s performance, continuity of study, and economic situation, with associated changes in sleep patterns (
Gavurova et al. 2022). It should be noted that some of the dimensions associated with decreased academic achievements (e.g., delayed graduation;
Aucejo et al. 2020) have been considered both an outcome and a cause of stress.
In the Italian context, unequal access to ICTs and poor Internet connectivity, combined with the inaccessibility of academic spaces, were also major problems for many students. Even when universities kept dormitories or some spaces open on campus, students in need remained mostly isolated and had limited access to services. In addition, the quality and impact of online teaching and learning, which we will not address in this paper, has been the subject of several debates (
Balestra et al. 2021;
Ghigi and Piras,
2021;
Goglio 2022).
Psychological dimensions and personal resources were analysed as important protective factors in coping with the pandemic and containment measures. In addition to self-efficacy, discussed below, other studies have highlighted the role of the sense of responsibility and belonging (
Procentese et al. 2020); the latter lies between personal and organisational resources, as it is fostered by organisational support and care measures such as counselling services. Even before the pandemic, the demand for additional resources to support mental health and well-being interventions on campus was higher than the available resources (
LeViness et al. 2019;
Lederer et al. 2021), a situation that was exacerbated during the pandemic, further threatening students’ future well-being and continuity of study. Finally, some studies have pointed to gender differences (
Sánchez-Teruel et al. 2021), geographic background, and socioeconomic status. Regarding gender,
Ding et al. (
2020) found that female students had greater risk perception related to COVID-19.
Alsaady et al. (
2020) also observed higher levels of exam anxiety among female students, and
Prowse et al. (
2021) found a more pronounced negative effect on academic performance, social isolation, stress, and mental health among female students compared to their male peers. Regarding performance (measured by the number of credits earned),
Bratti and Lippo (
2022), using administrative data from a public college in northern Italy, showed that the gender difference did not change, while women improved their performance compared to men in some fields of study (social sciences and humanities). These data, as well as some considerations from the extensive work of
Aristovnik et al. (
2020), suggest that special attention should be paid to gender differences (
Sánchez-Teruel et al. 2021).
1.2. Emotional Exhaustion and Engagement in Higher Education
The JD-R model is a widely known model of occupational well-being that assumes that each organisational context has specific job demands and resources, and therefore well-being depends on the balance of these two aspects (
Bakker and Demerouti 2017). Due to its flexibility and adaptability to different professional and educational contexts, the JD-R model allows for parallel reflection between the work and academic contexts. While the model is generally used in conventional work contexts, there have been recent studies that have applied the JD-R model to university students, such as the work of
Naylor (
2022), who conducted a study during the pandemic and used this model in conjunction with the self-determination theory (
Ryan and Deci 2000) to analyse psychological distress, highlighting the importance of motivational processes and analytic, data-driven measures that address various aspects of being a student.
Regarding the components of the model, job demands have been defined as psychological, social, or organisational aspects that require skills and efforts with associated physiological and psychological costs. Resources (related to the job or personal;
Schaufeli and Taris 2014), on the other hand, refer to aspects that perform one of the following three functions: reducing job demands and their costs; supporting the achievement of work goals, and promoting personal growth and learning (
Bakker and Demerouti 2017).
The relationship between demands, resources, and outcomes is complex; for example, it usually involves the moderating effect of resources on the relationship between demands and negative outcomes (
Ibrahim et al. 2021;
Kim and Wang 2018;
Tadić et al. 2015;
Xanthopoulou et al. 2007). In this study, however, we focus only on the direct relationships and pursue a more general goal of identifying determinants of emotional exhaustion and engagement and possible interventions to promote well-being. Specifically, we focus on the two well-known direct processes of the JD-R model, namely health impairment and motivation. The first process states that demands, especially when perceived as excessive or poorly designed, can deplete workers’ energy, leading to strain and other negative consequences such as emotional exhaustion (and more generally burnout); moreover, this process is negatively related to resources. Conversely, the other process states that resources have a motivational potential for workers, leading to engagement and thus positive outcomes, with a negative relationship to demands.
Concerning demands, in this study we focused on study load, a demand that refers to the amount and intensity of academic activities, which, as the name suggests, mainly revolve around studying and preparing for exams. Although study load is not fully synonymous with the workload dimension, the two variables are known to have some conceptual similarities and have similar effects on burnout and engagement; for example, one of the dimensions of student burnout is defined as exhaustion caused by study demands (
Schaufeli et al. 2002), which is not dissimilar to the definition of work-related burnout. Two recent studies have indicated that students perceived an increased study load during the pandemic, which is considered a major source of frustration and is positively correlated with stress and poorer mental and physical health (
Tasso et al. 2021;
Yang et al. 2021). Consistent with status as a requirement in the JD-R model and in line with the literature, our hypotheses are that study load is positively related to emotional exhaustion and negatively related to engagement.
In the most recent review of the JD-R model,
Bakker et al. (
2023) consider the various antecedents of burnout, taking into account the sources already reviewed by
Lee and Ashforth (
1996); these authors included stressful events in their meta-analysis on the determinants of burnout. Reduced academic performance, already mentioned among the relevant aspects during the pandemic, can be considered a distressing event. Multiple aspects can be identified as determinants of reduced academic performance during the pandemic, from social isolation to the radical and sudden changes in class attendance and studying. Increased study demands may also have played a role through their potential to reduce existing resources; in this study, given its cross-sectional nature, we simply observed the direct relationship between reduced academic performance and emotional exhaustion and engagement. Thus, we hypothesise that reduced academic performance is positively related to emotional exhaustion and negatively to engagement.
Alongside these variables, considering the specific pandemic experience of increased ICT use and an intensified presence of internet addiction issues, we included the latter in our study. Together with contextual factors, personal variables were considered possible determinants of emotional exhaustion and engagement; for example, some studies highlight the effect of neuroticism and internet addiction (
Lubbadeh 2020;
Pohl et al. 2021;
Toth et al. 2021). Internet addiction is a form of behavioural addiction that has attracted increasing interest since its conceptualisation in the late 1990s, both in terms of its prevalence and epidemiology and its association with multiple personal and social psychological dimensions (
Fumero et al. 2018;
Pan et al. 2020;
Young 1998a). This is particularly true for young adults, who have reported high levels of problematic Internet use in recent years (
Lozano-Blasco et al. 2022). While there are different models of Internet addiction, it can be broadly defined as compulsive and excessive Internet use with serious and negative effects on personal lives (
Poon 2018). Specifically, this work draws on
Young’s (
1998a,
1998b) model, which defines Internet addiction as a psychological dependence characterised by loss of control and problems with time management, interference with daily tasks and social relationships, excessive preoccupation and inability to moderate Internet use, mood changes, and hiding behaviours.
The association between Internet addiction and stress, anxiety, and depression has been documented in the student population both before (
Younes et al. 2016) and after (
Gavurova et al. 2022) the COVID-19 pandemic.
Gavurova et al. (
2022) found that the emergency context highlighted the importance of ICTs in students’ lives, not only for academic activities but also for leisure and maintaining meaningful social relationships. However, in addition to the positive aspects of ICTs use, we must also consider the sudden increase in technology-mediated academic activities, which not only already act as an independent demand, but also increase the risk of excessive Internet use, which is positively associated with academic burnout (
Zhu et al. 2022). Moreover, excessive Internet use (especially for distraction and entertainment) might have been employed as a maladaptive coping strategy (
Widyanto and McMurran 2004) for mood regulation or simply to avoid an overwhelming situation characterised by increased demands, dwindling resources, uncertainty, stress, and boredom. In other words, the Internet may have been the perfect tool to escape academic obligations, but at the same time the source of overstudying. Therefore, we hypothesise that Internet addiction has a positive relationship with emotional exhaustion and a negative relationship with engagement.
Self-efficacy is widely viewed in the literature as a positive factor in academic achievement, including during the pandemic (
Capone et al. 2020;
Alemany-Arrebola et al. 2020). This widely studied dimension represents a personal resource that comes into play in the interplay of demands and resources (
Salanova et al. 2002). Bandura defined self-efficacy as the “beliefs in one’s capabilities to organise and execute the courses of action required producing given attainments” (
Bandura 1977, p. 3). Bandura emphasised the role of self-efficacy in the pursuit of meaningful goals and in adaptive self-regulation, which positively affect overall health and well-being; based on one’s achievement, feedback received, and modeling, self-efficacy beliefs influence one’s ability to perform well and to develop positive and proactive attitudes toward academic challenges. Regarding gender differences, results are sometimes contradictory; while some studies found no significant differences (
Rivera Heredia et al. 2016), others reported higher levels of self-efficacy in men (
Molino et al. 2018;
Olmedo et al. 2018). Thus, we hypothesise that academic self-efficacy has a negative relationship with emotional exhaustion and a positive relationship with engagement.
H1. Study load (a) reduced academic performance (b) and internet addiction (c) have a positive relationship with emotional exhaustion.
H2. Academic self-efficacy has a negative relationship with emotional exhaustion.
H3. Study load (a) reduced academic performance (b) and internet addiction (c) have a negative relationship with engagement.
H4. Academic self-efficacy has a positive relationship with engagement.