COVID-19 anxiety and uncertainty of classes: Devastating effect on students’ academic behavior and performance

Review of the literature

A search was done with the keywords of ‘School and College closure’, ‘Academic Behavior’, ‘Covid Anxiety’ and ‘Academic Achievement’ in numerous permutations to review the previous literature. The researcher found 65 related articles and later reviewed them for conceptual development.

COVID-19 and the education sector

According to Li (2020) and Tinsley (2020) underline the widespread recognition that COVID-19 has not only affected public health and the economy but has also cast a long shadow over education. These initial studies likely serve as foundational pieces that acknowledged the magnitude of the crisis and initiated the discourse surrounding the pandemic’s impact on the education sector.

Furthermore, the mentions of Suwidiyanti and Anshori (2021) and Tan (2021) emphasize that the closure of schools and colleges during the pandemic caught the education sector off-guard, particularly with regards to the transition to digital classes. This unpreparedness likely resulted in a host of challenges, including unequal access to online learning resources and difficulties in adapting to new teaching methods. The current study can build upon these observations by delving deeper into the specific consequences of unpreparedness, particularly on students’ academic behavior and performance.

The reference to “Back to Basic” by Sarah et al. (2021) hints at the cyclic nature of reopening and closure of educational institutions as waves of COVID-19 continue to surge. This highlights the need for a more sustained and resilient approach to education during the pandemic. The present study could critically analyze the implications of such cyclical disruptions on students’ learning trajectories and mental well-being. Moreover, the repetition of Li (2020) in the context of the second wave of COVID-19 in India underscores the persistence of the challenges faced by the education sector. Despite the experience gained from the first wave, there seems to have been a gap in addressing the emotional and psychological needs of students. This observation could be a catalyst for the current study to explore the long-term impact of these challenges on students’ academic behavior and performance.

In summary, the existing literature cited here provides an essential foundation for understanding the initial shockwaves of COVID-19 on the education sector. It raises crucial questions about the unpreparedness of educational institutions, the challenges of transitioning to digital learning, and the cyclical disruptions caused by subsequent waves of the pandemic. The current study can contribute by critically examining the long-term consequences of these challenges on students’ academic behavior and performance, thereby justifying its focus on this critical aspect of the education sector’s response to COVID-19.

COVID-19 anxiety among students

While the WHO report suggests that individuals under 25 are less prone to severe COVID-19 infection, it is evident from the research (Faize and Husain, 2020) that even young students have been mentally disturbed. This discrepancy between the lower health risk and heightened anxiety among students highlights the psychological toll of the pandemic, which should be a central concern for educators and policymakers. The literature underscores the various sources of COVID-19-related anxiety among students. These include exposure to excessive information about the virus, the proliferation of false information, uncertainty surrounding lockdown rules, financial difficulties, and food scarcity (Sarah et al., 2021). Exam postponements and the shift to online examinations and viva add to academic stress (Faize and Husain, 2020; Günther, 2021). These insights can serve as a foundation for understanding the multifaceted nature of students’ anxiety and its potential consequences.

Several studies, such as Balhara et al. (2020), have identified a positive relationship between COVID-19 anxiety and academic stress. This association is significant as it suggests that students’ mental well-being is intricately linked to their academic performance. The observed increase in academic procrastination behavior and decreased interest in academic life among anxious students (N. Ammar et al., 2020) underscores the need to explore how anxiety translates into tangible changes in behavior and performance. The literature highlights specific vulnerabilities among students. For example, rural, low-income female students are reported to be more affected by mental health issues related to COVID-19 (Dangal and Bajracharya, 2020). The finding that students in higher-level classes experienced higher levels of anxiety (Huckins et al., 2020) and the steady increase in anxiety levels (Van Wyk, 2020) emphasize the dynamic nature of anxiety within the educational context. These insights can guide the current research in examining how different student demographics respond to and cope with COVID-19-related anxiety.

A critical aspect to consider is the documented negative influence of COVID-19 anxiety on academic performance (Madero Gómez et al., 2020). This link between mental health and academic outcomes underscores the importance of addressing anxiety as a determinant of students’ educational trajectories. A Research by Brouwer et al. (2021) points to the relationship between anxiety, depression, and suicidal thoughts among students. This highlights the severity of mental health issues stemming from COVID-19 anxiety and underscores the urgency of addressing this aspect in the current study. Baloran (2020) suggests that students with a high level of awareness about coping strategies experience lower anxiety levels. This raises questions about the effectiveness of interventions and coping mechanisms in mitigating the adverse effects of COVID-19-related anxiety. The challenges related to class uncertainty, the transition to virtual classes, and the lack of infrastructure (Tan, 2021) emphasize the role of external factors in exacerbating students’ anxiety. These factors may have direct implications for the study’s objectives, especially in understanding how students’ external environments contribute to their anxiety and academic behavior.

In conclusion, the cited literature provides a robust foundation for the current research by highlighting the multifaceted nature of COVID-19 anxiety among students, its impact on academic behavior and performance, and the need to consider various demographic and environmental factors. This critical discussion underscores the significance of investigating the long-term consequences of COVID-19 anxiety on students’ educational experiences and achievements, further justifying the current study’s objectives and scope.

Academic behavior

Academic behavior refers to the manner, posture, or ethical behavior that an individual encompasses in their academic life (Faize and Husain, 2020). The measurement of academic behavior is a complex task, and it involves many layers. The study considered the significant aspects of academic behavior such as ‘academic integrity,’ ‘learning habits’, ‘growth mindset’ and ‘academic satisfaction.’ The previous literature (Ferrinho, 2020; Lashley et al., 2020; Saddik et al., 2020 etc.) defined these components and found a relationship with academic performance.

Academic integrity

Academic integrity is a fundamental concept in education, representing the obligation to uphold honest and ethical behavior within the academic realm (‘Arbor Day Academic Integrity Survey Report’, 2014). It underscores the moral imperative that students acquire knowledge and skills through honest means, free from cheating or unethical practices. This ethical foundation is essential for the credibility and quality of education. The COVID-19 pandemic has introduced significant disruptions to academic integrity, particularly during the transition to online learning modalities. Studies such as Jamalpur et al. (2021) and Kecojevic et al. (2020) have documented a surge in cases of academic dishonesty, ranging from cheating during online exams to plagiarism in assignments. These findings underline the challenges educators and institutions face in maintaining academic integrity in the virtual learning environment.

The shift to online assessments during the pandemic has sparked discussions around approaches like online invigilation to combat cheating (Sheen et al., 2020). While these measures aim to preserve academic integrity, they also raise ethical concerns and questions about their effectiveness. Exploring students’ perspectives on such surveillance-based approaches could provide valuable insights into their impact on academic behavior. The pandemic has ushered in a host of changes in academic settings. Researchers have noted significant score disparities between online and classroom exams (Chaves et al., 2021; Eshet, Steinberger, and Grinautsky, 2021), an increase in academic outsourcing and file sharing (Charissi, Tympa, and Karavida, 2020), and a general uptick in academic misconduct (Papadimos et al., 2020). These changes necessitate a thorough examination of how alterations in academic practices have influenced students’ behavior and their adherence to academic integrity.

Stress appears to play a pivotal role in the rise of cheating behavior among students, particularly in the context of the pandemic-induced shift to online learning (Kulal and Nayak, 2020; Günther, 2021). The psychological toll of the pandemic and the challenges of adapting to new learning environments may contribute to academic misconduct. Understanding the interplay between stress, psychological factors, and academic behavior is essential for crafting targeted interventions. Lastly, the responsibility for upholding academic integrity is not borne solely by students. Günther (2021) highlights that multiple stakeholders, including students, educational institutions, teachers, and government bodies, share this responsibility. Effective solutions to address academic integrity challenges should involve collaborative efforts from all these parties. This perspective underscores the importance of a holistic and multifaceted approach to preserving the ethical foundations of education.

Learning habits

Learning habits are a crucial component of academic behavior, encompassing study attitudes and habits that profoundly influence students’ academic performance (Tran et al., 2020). These habits reflect how students approach learning, organize their study routines, and engage with educational materials. Their significance lies in their role as determinants of educational success, making them a subject of considerable importance for both educators and researchers. The COVID-19 pandemic has wrought significant changes in students’ learning habits, as evident from the findings of studies like Lawrence et al. (2021). The sudden shift to online learning modalities disrupted established study routines, leading to a widening learning gap. Reduced study hours and a decline in note-taking habits suggest the challenges students faced in adapting to remote learning environments. The transition to online classes also brought forth challenges in maintaining a productive learning space, with limited one-on-one interaction with teachers (Balhara et al., 2020).

One notable revelation from the literature is the positive relationship between COVID-19 anxiety and decreased study hours, subsequently affecting academic achievement (Jamalpur et al., 2021). This link underscores the critical role of students’ psychological well-being in shaping their learning habits, especially in the face of unprecedented disruptions. The long-term implications of altered learning habits post-pandemic are a matter of concern. Zhang, Taub, and Chen (2021) estimated a significant reduction in learning gains, raising questions about how persistent changes in learning habits might impact students’ educational trajectories in the years to come.

Furthermore, the digital divide has exacerbated inequalities in education. Students from low-income backgrounds, unable to access essential technology like computers and Wi-Fi, face severe challenges in participating in online classes (Zhang, Zhou, and Xia, 2020). These inequalities not only hinder their ability to maintain effective learning habits but also exacerbate educational disparities. Another worrying trend is the increased time spent on social networking sites during online classes (Riva, Wiederhold, and Mantovani, 2021). This diversion from educational materials and reduced engagement pose challenges to effective learning habits.

In conclusion, the literature demonstrates the profound impact of the COVID-19 pandemic on students’ learning habits. These changes, characterized by disruptions, inequalities, and shifts in student engagement, have implications for academic performance and the long-term educational outcomes of students. Understanding these dynamics is essential for crafting targeted interventions to support students in adapting to new learning modalities and ensuring their continued academic success.

Growth mindset

Motivation to excel academically and cultivate a growth mindset are central to shaping students’ academic behavior, and this has been underscored by the existing literature. As elucidated by Yeo et al. (2021), the pursuit of academic excellence is a powerful driving force, and educational institutions, such as schools and colleges, play a pivotal role in nurturing and sustaining this motivation among students. However, the COVID-19 pandemic has introduced unique challenges to this dynamic, particularly as traditional classroom settings shifted to online platforms. The repercussions have been profound, making it increasingly difficult to gauge and address students’ mental well-being and academic stress, as highlighted by Ho et al. (2020) and Sarah et al. (2021).

The transition to online learning during the pandemic has not only posed challenges to student motivation but has also altered the educational landscape. Sarrot (2020) points out that many educators, in their quest to adapt swiftly, may have focused primarily on completing syllabi, potentially overlooking the broader educational dimensions encompassing mental abilities, reasoning skills, and communication skills. This shift in priorities has significant implications for students’ overall development and their academic behavior. One particularly noteworthy concept that emerges from the literature is the notion of a growth mindset. Snipes and Tran (2017) highlight its importance by showcasing how a growth mindset can mitigate academic stress and contribute to improved academic achievement. This mindset not only fosters academic well-being but also instills confidence in students to navigate challenges, reducing feelings of depression and anxiety, as demonstrated in studies by Hanisa (2020), Bowyer (2021), and Fatimah, Rajiani, and Abbas (2021).

In a critical discussion, it is evident that the COVID-19 pandemic has reshaped the educational landscape, requiring innovative approaches to sustain motivation and a growth mindset among students. Educators must adapt to new teaching paradigms while prioritizing students’ mental well-being and the cultivation of essential skills. Moreover, the intricate interplay between motivation, a growth mindset, and academic achievement necessitates a deeper examination, accounting for various contextual factors and individual student experiences. By critically analyzing these aspects, the current research stands poised to contribute valuable insights into the evolving dynamics of motivation, mindset, and academic behavior within the unique challenges posed by the COVID-19 pandemic. These insights can inform strategies for maintaining motivation, nurturing growth mindsets, and enhancing academic performance in the face of ongoing educational disruptions.

Academic satisfaction

Academic satisfaction, as delineated by Aristovnik et al. (2020), encompasses students’ contentment with various facets of their academic environment. It extends beyond subjective feelings and holds a significant influence on students’ overall mindset and academic performance. This satisfaction is intricately linked to how students perceive their educational settings, encompassing elements such as classes, teachers, exams, and more. Importantly, the literature has consistently underscored the profound impact of academic satisfaction on students’ academic achievements, forming a symbiotic relationship between contentment and success. The existing body of research accentuates the strong correlation between academic satisfaction and academic performance. Studies conducted by Kumar, Nayar, and Bhat (2020), Li (2020), and Peinado-Guevara et al. (2021) converge on the idea that students who harbor satisfaction in their academic pursuits are more likely to excel academically. It is evident that contentment within the educational environment acts as a potent motivator, shaping students’ positive mindsets and bolstering their pursuit of academic excellence.

However, the COVID-19 pandemic has introduced unique challenges, altering the landscape of academic satisfaction. Students’ expressions of missing the informal classroom interactions, the camaraderie of classmates, and the dynamics of traditional in-person learning underscore the pandemic’s impact. As illuminated by Cipriano et al. (2020) and Akcil and Bastas (2021), the abrupt shift to online classes has been met with feelings of boredom and decreased concentration. These factors have collectively led to a decline in academic satisfaction, ultimately linked to reduced academic performance. In a critical discussion, it is discernible that the pandemic has not only redefined the parameters of academic satisfaction but has also revealed its intricate relationship with academic performance. The traditional markers of satisfaction, such as classroom experiences and teacher interactions, continue to be pivotal. However, the swift adoption of online learning platforms has introduced novel challenges in maintaining satisfaction levels. The ability to adapt to virtual classrooms, coupled with the potential loss of engagement, now plays a substantial role in shaping academic satisfaction.

Moreover, it is essential to appreciate the bidirectional nature of the relationship between academic satisfaction and performance. Satisfaction can serve as a catalyst for improved performance, while declining performance can conversely erode satisfaction levels. This cyclic interplay emphasizes the need for a comprehensive approach to addressing both factors in tandem. The current research can make a valuable contribution by shedding light on the unique ways in which the COVID-19 pandemic has impacted students’ academic satisfaction. Furthermore, it can delve into strategies aimed at preserving or enhancing academic satisfaction within the context of ongoing disruptions, recognizing its pivotal role in motivating students and shaping their overall educational experience.

Academic performance

The literature on measuring academic performance offers a spectrum of approaches, reflecting the multifaceted nature of assessing students’ educational achievements. Some studies, such as those by Aristovnik et al. (2020), Cipriano et al. (2020), Faize and Husain (2020), and Sarah et al. (2021), adopt a conventional approach by relying on traditional metrics like exam scores, grade point averages, or percentages to gauge academic performance. These metrics have long been considered fundamental indicators of a student’s educational success. However, an emerging body of literature, represented by studies like Sarrot (2020) and Trammell et al. (2021), argues for a broader perspective on academic performance. In this view, academic achievement encompasses not only exam results but also factors like participation in extracurricular activities, which contribute to a well-rounded educational experience. The inclusion of extracurricular activities recognizes that holistic development goes beyond academic scores and encompasses other facets of a student’s growth.

The COVID-19 pandemic has posed distinctive challenges to assessing academic performance. With the suspension of extracurricular activities and the cancellation of exams in some universities, the conventional methods of evaluation were disrupted. Consequently, your study adopts a pragmatic approach by asking students to self-rate their academic performance over the past year. This approach acknowledges the exceptional circumstances and seeks to understand the pandemic’s influence on academic achievement, with academic behavior as a potential mediator. The literature demonstrates that extended disruptions in educational continuity, such as long summer vacations, absenteeism, and weather-related college closures, have historically been associated with negative impacts on academic performance (Aristovnik et al., 2020; Sheen et al., 2020). However, the COVID-19 situation presents a more complex challenge, as it has affected not only academic activities but also outdoor pursuits, thereby hampering students’ overall development (A. Ammar et al., 2020).

Moreover, the pandemic’s adverse effects on students’ reading abilities and the estimated time required for them to catch up on their subjects, as suggested by Sheen et al. (2020), underscore the potential long-term repercussions of disrupted learning. This is further highlighted by Chhatre et al. (2020), who indicate that academic achievement gaps can lead to higher dropout rates in rural areas. The study also draws attention to the broader socio-economic implications, including increased income disparity, lower economic growth, higher unemployment, and a rise in crime rates, if effective measures are not taken by the government. In a critical discussion, it is evident that the COVID-19 pandemic has disrupted traditional methods of measuring academic performance, necessitating innovative approaches. The broader conception of academic achievement, which considers extracurricular activities and holistic development, aligns with the changing educational landscape. Furthermore, the long-term consequences of disrupted learning, as highlighted by the literature, emphasize the urgency of addressing these issues comprehensively, not only for individual student well-being but also for broader societal outcomes.

Uncertainty of classes

The disruption caused by the COVID-19 pandemic has been acutely felt in educational institutions, triggering uneasiness and anxiety among students, teachers, and management alike. Dewi and Wibowo (2020) and Yeo et al. (2021) highlight the palpable sense of disruption in regular classes and the postponement of exams, contributing significantly to academic stress. Moreover, this disruption has, in some cases, eroded students’ interest in academic pursuits. This finding underscores the psychological toll that class uncertainty can exert on students. The uncertainties surrounding academic schedules have been particularly challenging for international students studying abroad, as pointed out by Suwidiyanti and Anshori (2021). Visa issues, delays in the academic year, and the logistical challenges of online learning have compounded their difficulties. Meanwhile, domestic students have grappled with financial burdens, as Li (2020) notes how many were obligated to pay hostel rent despite being unable to reside on campus due to the shift to online classes.

Adding to the complexity, some private educational institutions initiated online admission processes for the subsequent academic year, as observed by Arora, Chakraborty, and Bhatia (2021). This created emotional and financial strains on a significant number of students, further highlighting the multifaceted challenges posed by the pandemic. In a critical discussion, it is evident that the uncertainties surrounding class schedules during the COVID-19 pandemic have had far-reaching consequences. These uncertainties have taken a toll not only on students’ emotional well-being but also on their academic pursuits. The challenges faced by international students underscore the global ramifications of these disruptions, while the financial burdens experienced by domestic students have added another layer of complexity.

Conceptual model

With a critical analysis of the previous literature, the present study hypothesized a conceptual model to explain the relationship between study variables. The model explains the casual relationship between the independent variable (COVID-19 anxiety and the uncertainty of classes) and the dependent variable (academic performance) mediating through academic behavior. The model (Figure 1) was developed to understand whether COVID-19 anxiety and uncertainty of classes affect the students’ academic behavior, thereby affecting their academic performance. To depict this concept, the following hypotheses were framed:

Figure 1. COVID-19 anxiety and uncertainty of classes on academic performance through academic behavior.

H1 - Path of first Hypothesis; H2 - Path of Second Hypothesis; H3 - Path of third Hypothesis.

H₁: COVID-19 anxiety has a significant impact on students’ academic behavior.

H₂: Uncertainty of classes significantly impacts the academic behavior of students.

H₃: There is a significant relationship between the academic behavior of students and their academic performance.

Research design

The present study is explanatory (causal) and explains the causal relationship between the study variables. The study was carried out from 27^th August 2021 to 10^th March 2022 in the Coastal Karnataka region of Karnataka. The present study is explanatory and explains the causal relationship between the study variable. The study was carried out in the Coastal Karnataka region of Karnataka. The study area comes under Mangalore University’s jurisdiction and consists of 194 colleges providing undergraduate and postgraduate courses. In the study area, approximately 35,000 students are pursuing their graduate and postgraduate degrees.

Instrument and variable measurement

In this study, COVID-19 anxiety, one of the independent variables, was measured using a well-established Likert scale comprising six statements adapted from Baloran (2020). These statements sought to gauge students’ levels of anxiety related to the pandemic. Respondents were asked to express their degree of agreement or disagreement with each statement on a five-point Likert scale, with options ranging from “1” indicating “Strongly Disagree” to “5” signifying “Strongly Agree.” Subsequently, individual scores for COVID-19 anxiety were calculated by summing the responses to these six items, yielding a total score that ranged from 6 to 30. Higher cumulative scores represented higher levels of anxiety experienced by the respondents in relation to the COVID-19 pandemic.

The second independent variable, the uncertainty of classes, was assessed using a set of five newly developed items. These items aimed to capture students’ perceptions and sentiments concerning the uncertainties surrounding their academic classes. Similar to the COVID-19 anxiety scale, participants were asked to express their agreement or disagreement with each of these items using a five-point Likert scale, where “1” denoted “Strongly Disagree,” and “5” indicated “Strongly Agree.” Individual scores for class uncertainty were determined by summing the responses to the five items. The total score for this variable ranged from 5 to 25, with higher scores indicating a greater impact of class uncertainty on the students.

Furthermore, the mediating variable, academic behavior, was assessed by considering four distinct factors: academic integrity, learning habits, growth mindset, and academic satisfaction. Each factor was evaluated using a series of statements. Respondents provided their feedback on these statements using the same five-point Likert scale, where “1” represented “Strongly Disagree,” and “5” represented “Strongly Agree.” Separate scores were computed for each of the four academic behavior factors by summing the responses to the relevant statements. Higher scores within each factor indicated a stronger presence of that specific aspect of academic behavior.

Lastly, academic performance, the dependent variable, was evaluated through a single direct question, inviting respondents to self-assess their academic performance. Participants used a five-point rating scale to convey their evaluation, with options ranging from “5” denoting “Excellent” to “1” signifying “Extremely Poor.” This straightforward assessment allowed respondents to subjectively rate their academic achievements within the provided scale.

These meticulously designed measurement scales facilitated a comprehensive and nuanced evaluation of the variables under scrutiny, affording valuable insights into the interplay between COVID-19 anxiety, class uncertainty, academic behavior, and academic performance within the context of this study.

Reliability and validity

In the first stage, the items generated by undertaking an organized literature review and construct were recognized. In the second stage, these were discussed with experts and academics for the content validity of the questionnaire. In the third stage, pre-testing was conducted by taking 15 students at Mangalore University Campus to know the level of understandability of the questionnaire among students. At stage four, the reliability test was conducted to know the internal consistency of the item scale using Cronbach alpha and the results are depicted in Table 1.

Table 1 explains the result of reliability analysis for scale data, and it exhibits that COVID-19 anxiety (I=.703), uncertainty of class (I=.758), academic integrity (I=.845), learning habit (I=.676), growth mindset (I=.689), and academic satisfaction (I=.787) were found reliable (I>0.6).

Participants and selection criteria

Students pursuing postgraduate and undergraduate courses from the Mangalore university jurisdiction were considered as participants. The researcher visited administration office of Mangalore University and approached Registrar (Evaluation) to get details of the students. With help of given details and from each college’s website, the researcher was able to develop sampling frame. Later a simple random sampling technique (Lottery Method) was used to select sampling unit. Taro (1967) sample size formula used to determine sample size for our study. A total of 480 students of regular course (Minimum of 2 years course) were invited to fill out the questionnaire from 10 ^th November 2021 to 18^th February 2022.

Data collection setting

Data was collected using a structured questionnaire. The researcher distributed the questionnaire through email, WhatsApp, and telegram app of students (personal interview was not possible due to COVID-19 protocols). We collected contact details for further clarification and follow-up. Data was collected from April 10, 2022 to July 25, 2022. Collected data were entered into SPSS 20 and missing data was removed. Later, data were coded and used for analysis. Some repetitive questions were used to identify the bias responses from the students and such responses were removed from the final data. Data analysis and interpretation was undertaken from 2^nd August 2022 to 10^th September 2022.

Ethical considerations

The adopted questionnaire was submitted to the academic integrity and ethical committee of Kingdom University and got approved as per the research policy and procedure of the university council 20^th November, 2022. We were not able to collect prospective ethical approval due to lockdown in the institution. Therefore, for this study we obtained retrospective ethical approval from Academic Integrity and Ethics Committee, College of Business Administration, Kingdom University, Bahrain on 20^th November 2022 (CBA/17/22). The researcher asked for the oral permission from students before distributing questionnaire and informed them the purpose of the research. The respondents gave their full consent without force from anyone. Subsequently, in order to protect the rights and privacy of the respondents, all forms of private data (Name and contact details) were acquired will remain confidential.

Data analysis and interpretation

SPSS version 20 was used for data entry and analysis. After entering data into SPSS, data screening and data transformation were made for accurate analysis. In the data analysis phase of this study, two distinct statistical methods were employed to investigate the research objectives. First, the One Sample t-test was utilized to assess whether the sample mean values of the key variables—COVID-19 anxiety, uncertainty of classes, academic behavior, and academic performance—significantly deviated from hypothesized or predetermined values. This statistical test helps ascertain whether these variables exhibit statistically significant variations from the expected or baseline values, providing insights into the unique characteristics of the study’s sample in relation to these variables.

Secondly, the study employed multiple regression analysis within the framework of Structural Equation Modeling (SEM). SEM is a robust and advanced statistical technique that delves into the complex relationships and causal pathways among multiple variables simultaneously. Within the SEM framework, multiple regression allows for a thorough exploration and quantification of the relationships among COVID-19 anxiety, uncertainty of classes, academic behavior (mediating variable), and academic performance (dependent variable). It goes beyond mere correlation analysis by enabling the testing of a hypothesized model that elucidates how these variables interact and influence each other. Specifically, SEM can unveil whether COVID-19 anxiety and class uncertainty have direct or indirect effects on academic performance through their impact on academic behavior, providing a comprehensive understanding of the underlying mechanisms and effects in the study.

Demographic profile

Demographic details of the respondents help to understand the diversity of students in respect of their gender, age, the field of study, and level of study. The researcher took utter care to ensure that each group had equal proportions in numbers. The demographic details of the respondents are depicted in Table 2.

Table 2 demonstrates that the majority of the respondents were female (69.4%), and 65.20% of the respondents were aged between 18–22. Most of the respondents (37.5%) were to pursuing their degree/diploma course, and 60.21% of the respondents were studying commerce and management. The analysis of the demographic details shows that female students are more interested in the enrolment of higher education. The study revealed that the commerce and management courses have the highest Gross Enrolment Ratio (GER) than science and arts courses.

COVID-19 anxiety among students

The COVID-19 situation altered every aspect of human life, especially mental health. Students felt fear and depressed in this pandemic situation which adversely affected their academic life (Meo et al., 2020). At this juncture, the researcher was interested to know the level of COVID-19 anxiety among students. The total responses were converted into the score and grouped into four levels. If the mean value was less than 2, then it was treated as ‘extremely low level of anxiety’, if the mean value was between 2.1 to 3, then it was treated as ‘low level of anxiety’, if the mean value was between 3.1 to 4, then it was treated as ‘high level of anxiety’, if mean value was more than 4, then it was treated as ‘extremely high level of anxiety.’ The analysis of the COVID-19 anxiety among students is depicted in Table 3.

Table 3 explains that students had a high level of anxiety about being infected (mean (M)=3.38), and they agreed that they felt depressed (M=3.48) during the lockdown. Students agreed that they spent more time on reading or watching COVID-19 information (M=3.56), and they also agreed that they were feeling anxiety and insomnia during the lockdown period (M=3.08). The majority of the students strongly agreed that they get too conscious in social gatherings (M=4.21); the overall mean and standard deviation are 3.6167±.64333, which indicates that the students had a high level of COVID-19 anxiety. The result of the one sample t-test exhibits that the opinions given by respondents are significantly different from the mean value; since the p-value is less than 0.05, we can conclude that the results are statistically significant.

In the analysis of students’ responses to statements related to COVID-19 anxiety in Table 3, the calculated effect sizes, represented by Cohen’s d, were used to quantify the magnitude of differences observed when comparing the sample mean to the hypothesized mean of 3. The results indicate varying levels of COVID-19 anxiety among students during the lockdown. Notably, for statements such as “Feeling depressed” (Cohen’s d=0.772) and “Spending more time on reading or watching COVID-19 information” (Cohen’s d=0.827), students reported moderate to large effect sizes, signifying substantial deviations from the hypothesized mean. Conversely, “Feeling anxiety and insomnia during the lockdown period” exhibited a smaller effect size (Cohen’s d=0.154), suggesting a relatively minor difference from the hypothesized mean. Additionally, “Getting too conscious in social gatherings” displayed a large effect size (Cohen’s d=1.643), indicating a significant departure from the hypothesized mean, highlighting pronounced social anxiety in such situations. These effect sizes offer a nuanced understanding of the magnitude of students’ COVID-19 anxiety, complementing the significance tests and providing valuable insights into the impact of the pandemic on their emotional well-being.

Students’ perception on uncertainty of class

Suspension of schools and colleges is the only option to prevent COVID-19 from spreading among students on campus, but this decision has made significant changes in the students’ academic behavior (Nguyen et al., 2020; Tran et al., 2020). To know how the uncertainty of classes affected the perception of students on their academic life, the researcher collected opinions of students and the results are portrayed in Table 4.

Table 4 indicates that students are strongly agreed that frequent closure of schools and colleges affected the academic behaviour of students (M=4.39) and suspension of classes and postponement of exams adversely affected their learning interest (M=4.39). The majority of the students strongly agreed that irregular classes made them unfocused in the academic field (M=4.33), and they also agreed that they are comfortable with online classes (M=3.98). The overall mean and standard deviation are 4.2458±.68438, which indicates that students have a negative perception (since all statements are negative) about the uncertainty of class and its influence on their academic life. The result of a one sample t-test indicates a p-value less than.05; therefore, it can be concluded that the results are statistically significant.

In the assessment of factors affecting students’ academic behavior during the COVID-19 pandemic, calculated effect sizes, represented by Cohen’s d, were used to quantify the magnitude of differences observed when comparing the mean scores of specific statements to the hypothesized mean of 3. The results reveal substantial variations in the impact of these factors on academic behavior. Notably, statements such as “Suspension of Classes and Postponement of exam adversely affect learning Interest” (Cohen’s d=1.383) and “Overall” (Cohen’s d=1.869) displayed large effect sizes, indicating significant departures from the hypothesized mean and emphasizing the adverse influence of these factors on students’ academic behavior. Conversely, “Not comfortable with online classes” exhibited a moderate effect size (Cohen’s d=0.854), suggesting a comparatively smaller yet notable impact. Other statements, including “Frequently closure of Schools and Colleges affect Academic behavior of Students” (Cohen’s d=1.538), “Missing friends and Teachers during lockdown” (Cohen’s d=1.135), and “Irregular classes made unfocused in the academic field” (Cohen’s d=1.767), also displayed substantial effect sizes, underscoring their significance in shaping students’ academic behavior amidst the pandemic. These effect sizes enhance the interpretation of the results, providing valuable insights into the varying degrees of influence these factors exert on students’ educational experiences.

Academic integrity of the students during the COVID-19 pandemic

Academic integrity helps the students develop responsible behavior and honesty in their work (Aucejo et al., 2020). To know whether difficulties and challenges during the COVID-19 situation altered the academic integrity of the students, the researcher analyzed academic integrity and the results are shown in Table 5.

Table 5 indicates that students agreed that their classmates were copying from another student or source during an online test (M=3.41) and that they join the online class but do not listen to the class (M=3.95). Most of the students strongly agreed that the COVID-19 situation made students lazy, Dishonest, and irresponsible (M=4.09), and that they copy online sources to write an assignment without giving citations to the original author (M=3.86). The overall mean and standard deviation are 3.8297±.88585, which indicates that students agree that the COVID-19 situation negatively influenced their academic integrity and results from a one sample t-test supported (p<0.05) the significance of the result.

In the examination of factors related to academic integrity and behavior in the context of online learning during the COVID-19 pandemic, effect sizes were calculated using Cohen’s d to gauge the magnitude of differences between the mean scores of specific statements and the hypothesized mean of 3. These effect sizes shed light on the extent to which these factors influenced students’ behavior. Notably, “Joining the online class but do not listen to the class” displayed a substantial effect size (Cohen’s d=0.920), indicating a notable departure from the hypothesized mean and underscoring the prevalence of this behavior among students during online classes. Similarly, “Covid situation made students lazy, Dishonest, and irresponsible” exhibited a significant effect size (Cohen’s d=0.994), emphasizing the impact of the pandemic on students’ attitudes and conduct. Additionally, “Copy from online sources to write their assignments without citation” (Cohen’s d=0.809) and “Copying from another student or source during an online test” (Cohen’s d=0.368) also displayed noteworthy effect sizes, indicating the challenges posed by academic dishonesty and lack of citation practices in the online learning environment. These effect sizes provide valuable insights into the behavioral changes observed in students during the pandemic and contribute to a deeper understanding of academic integrity issues in online education.

Students’ learning habits during the COVID-19 pandemic

Undoubtedly, the COVID-19 pandemic situation would create unequal learning opportunities depending upon access to technology and family support; this would completely change the learning habit of students (Rahman et al., 2020). With this intention, the researcher gathered information about how students’ learning habits were affected by the COVID-19 pandemic and the analysis is depicted in Table 6.

Table 6 exhibits that students agreed that they clearly remember their subject’s contents after lockdown (M=3.37) and can appropriately concentrate on their studies (M=3.38). The majority of the students agreed that their study hours are decreased in the last year (M=3.97), and they were not able to write proper notes from online classes (M=3.81). The overall mean and standard deviation are 3.7396±.59568, which indicates that students agreed that the COVID-19 situation altered their learning habits, and the one-sample t-test proved the significance (p<0.05) of the result.

In evaluating the impact of the COVID-19 pandemic on students’ learning habits and concentration, effect sizes were calculated using Cohen’s d to quantify the magnitude of differences between the mean scores of specific statements and the hypothesized mean of 3. Notably, “Reduction in Group discussion, Debate and Doubt session” exhibited a substantial effect size (Cohen’s d=1.283), indicating a significant departure from the expected mean and underscoring the marked reduction in interactive academic activities during the pandemic. “Decrease in the study hours” also displayed a notable effect size (Cohen’s d=1.087), emphasizing the considerable decrease in study time. Additionally, “Writing notes from online class” (Cohen’s d=0.688) showed a significant effect size, highlighting changes in students’ note-taking practices. These effect sizes offer valuable insights into the alterations in learning habits and concentration levels among students in response to the challenges posed by the pandemic, contributing to a deeper understanding of the evolving academic landscape during these times.

Students’ growth mindset during COVID-19 pandemic

A growth mindset is essential to keep the students motivated towards learning, and it makes them keep hope in their careers (Molock and Parchem, 2020). With this intention, the researcher analyzed whether the COVID-19 situation influenced students’ growth mindset, and the results are shown in Table 7.

Table 7 indicates that students agreed that the COVID-19 pandemic does not alter their academic goal (M=3.45); instead, it gave them more time to concentrate on their passion (M=3.42). Most of the students agreed that the lockdown period gave them more learning choices (M=3.54) and got a break from academic stress (M=3.64). The overall mean and standard deviation are 3.5135±.79866, which indicates that students have a highly positive (since all statements are positive) mindset about academic life, and the result of the one sample t-test proved the significance of the result.

In examining the influence of the COVID-19 pandemic on students’ academic behavior and personal attitudes, effect sizes were calculated using Cohen’s d to gauge the magnitude of differences between the mean scores of specific statements and the expected mean of 3. Notably, “Break from academic stress” displayed a substantial effect size (Cohen’s d=0.547), indicating a significant deviation from the anticipated mean and highlighting the relief students found from academic pressures during the pandemic. “Overall” academic responses also showed a noteworthy effect size (Cohen’s d=0.656), suggesting a collective shift in students’ attitudes and behaviors during these challenging times. Other factors, such as “Choices of learning during lockdown” (Cohen’s d=0.495) and “Impact on Academic Goal” (Cohen’s d=0.405), demonstrated moderate effect sizes, underlining notable shifts in students’ academic aspirations and learning preferences. These effect sizes provide valuable insights into how students adapted and coped with the disruptions caused by the pandemic, shedding light on their evolving academic perspectives and personal pursuits.

Students’ academic satisfaction during the COVID-19 pandemic

Students’ academic achievement cannot be measured merely with the yardstick of grade points or marks, and it is important to consider students’ academic satisfaction to complete the process (Aristovnik et al., 2020). Therefore, the researcher asked the students to state their level of satisfaction in various aspects of academic settings such as teachers, management, family, etc. The analysis of students’ academic satisfaction during the COVID-19 pandemic is depicted in Table 8.

Table 8 indicates that students are satisfied with classes taken during the lockdown period (M=3.08), teachers’ support during lockdown (M=3.65), and infrastructure facilities they had to attend online classes (M=3.15). Most of the students agreed that they are satisfied with measures taken by schools and colleges to uplift the academic career of the students during pandemic situations (M=3.48). Most of the students strongly agreed that they are satisfied with family support in their pursuance (M=4.01) and agreed that students are satisfied with academic performance (M=3.37). To find out the level of satisfaction, the researcher calculated the mean score ranging from 1 to 5. If the mean value was less than 2, then it was treated as very low satisfaction; if the mean value was between 2.1 to 3, then it was treated as low satisfaction; if the mean value was between 3.1 to 4, then it was treated as high satisfaction; if mean value was above 4, then it was treated as very high level of satisfaction. The overall mean and standard deviation are 3.4583±0.77391, which indicates that students had a high level of academic satisfaction in the last academic year.

The effect sizes (Cohen’s d) were computed to assess the magnitude of differences between the mean scores of various statements and the expected mean of 3. Notably, “Family support” exhibited a substantial effect size (Cohen’s d=1.140), underscoring the considerable deviation from the expected mean and highlighting the crucial role of family support during the pandemic. “Teachers’ Support during the lockdown period” also displayed a moderate effect size (Cohen’s d=0.587), indicating notable differences in students’ perceptions of teacher support. Additionally, “Overall” responses demonstrated a noteworthy effect size (Cohen’s d=0.652), signifying a collective shift in students’ perspectives on various aspects of their educational experience during the challenging times of the pandemic. Other factors, such as “Measures taken by Schools and colleges” (Cohen’s d=0.447) and “Academic performance” (Cohen’s d = 0.327), indicated moderate effect sizes, reflecting discernible shifts in students’ satisfaction and academic performance perceptions. These effect sizes provide valuable insights into the multifaceted factors that shaped students’ educational experiences during the COVID-19 pandemic, emphasizing the significance of support systems and overall satisfaction with online learning and academic performance.

Academic performance of students during the COVID-19 pandemic

As earlier cited, the measurement of academic performance is a real challenge. The traditional school of thought purely measures based on their marks or grade obtained in written exams. However, modern schools of thought also add other academic aspects such as extracurricular activities to measure academic performance (Rodriguez et al., 2020; Suwidiyanti and Anshori, 2021). Moreover, in the last year, universities cancelled most of the extracurricular activities and cancelled exams for some students. So, it is not easy to consider both schools of thought as a yardstick to measure academic performance. Therefore, in this study the researcher directly asked the students to rate their academic performance in the last academic year, and the result is depicted in Table 9.

Table 9 indicates that the majority of the students (51.5%) have an average level of academic performance and 24.4% of the students have an above average level of academic performance. The study revealed that 11.9% of the students are below the average level of academic performance, and only 7.5% of them perform excellently in their academics. The overall mean and standard deviation are 3.18±.908, which indicates that students are at the above-average level of academic performance.

Hypothesis testing

To find out the causal relationship between the study variable, a hypothesized conceptual model was developed and represents three hypotheses. To test each hypothesis, proper statistical tools were used, and the result of each hypothesis are tabulated in Table 10.

Table 10. Regression beta coefficients and hypotheses test result.

Hypothesis	Casual relation	Estimate (β)	R 2	R	P-value	Remarks
H1	Covid Anxiety ➔ Academic Behavior	-0.07	0.155	-.394	.000	Supported
H2	Uncertainty of Class ➔ Academic Behavior	-0.23	0.087	-.296	.000	Supported
H3	Academic Behavior➔ Academic Performance	0.37	0.107	.328	.000	Supported

The analysis of the first hypothesis (H₁) showed that COVID-19 anxiety significantly described the variability in academic behaviour (AB) of the students (R²=0.155, p<0.05), indicating that high COVID-19 anxiety negatively impacted on academic behaviour. Hence, H₁ explaining the relation between COVID-19 anxiety and academic behaviour was accepted (R=-0.394), as COVID-19 anxiety (β=-0.07, p<0.05) significantly and negatively impacted the academic behavior of the students. The analysis of the second hypothesis (H₂) exhibits that the uncertainty of class significantly contributes to the variation of academic behaviour of the students (R²=0.087, p<0.05), indicating that a high level of uncertainty of classes negatively affects the academic behaviour. Hence, H₂ signifying the relation between the uncertainty of classes and academic behaviour of the students was accepted (R=-0.296), as uncertainty of classes (β=-0.23, p<0.05) significantly and negatively impacted the academic behavior of the students. Likewise, the test result of the third hypothesis (H ₃) showed a positive correlation between the academic behaviour and academic performance of the students (R=0.328, p<0.05), which explains that students with a high level of academic behaviour achieved high academic performance. Hence, H₃ implicating the relation between academic behaviour and academic performance was accepted, as academic behaviour (β=0.37, p<0.05) significantly influenced the students’ academic performance. The result of a causal relationship between study variables is explained with the help of a path analysis and is depicted in Figure 2.

Figure 2. Path analysis of the model.

Source: Data Analysis. AI - Academic Integrity; LH - Learning Habit; GM - Growth Mindset; AS - Academic Satisfaction.

Figure 2 indicates that a 1% increase in COVID-19 anxiety decreases academic behavior by 0.07%, which means that increases in COVID-19 anxiety reduce the academic integrity, learning habits, growth mindset, and academic satisfaction of the students. Likewise, a 1% increase in uncertainty of class decreases 0.23% of academic behavior, which means frequent closure of schools and colleges adversely affects the students’ academic behavior. Later, the researcher evaluated the impact of COVID-19 anxiety and uncertainty of class on academic performance. Here the researcher identified both the direct effect (COVID-19 anxiety and uncertainty of class on academic performance) and the indirect effect (through academic behaviour). The standardized total (direct and indirect) effect of COVID-19 anxiety on academic performance is -.026. Due to both direct (unmediated) and indirect (mediated) effects of COVID-19 anxiety on academic performance, when COVID-19 anxiety goes up by one standard deviation, academic performance goes down by 0.026 standard deviations. The direct (unmediated) effect of COVID-19 anxiety on academic performance is.000, which means that COVID-19 anxiety does not have a significant direct impact on academic performance. The standardized indirect (mediated) effect of COVID-19 anxiety on academic performance is -.026.

Similarly, the standardized total (direct and indirect) effect of the uncertainty of class on academic performance is -.085. Due to both direct (unmediated) and indirect (mediated) effects of the uncertainty of class on academic performance, when the uncertainty of class goes up by one standard deviation, academic performance goes down by 0.085 standard deviations. The direct (unmediated) effect of the uncertainty of class on academic performance is.000, which means that the uncertainty of class does not directly impact academic performance. The indirect (mediated) effect of the uncertainty of class on academic performance is -.112. Due to the indirect (mediated) effect of the uncertainty of class on academic performance, when the uncertainty of class goes up by 1, academic performance goes down by 0.112. The relationship between academic behaviour and academic performance (H ₃) indicates that when academic behaviour increases by one standard deviation, academic performance increases by 0.37, which explains that an increase in academic integrity, learning habits, growth mindset, and academic satisfaction helps to increase the academic performance of the students. The model fit indices such as goodness of fit index (.98), adjusted goodness fit index (0.92), root mean square error approximation (0.040), chi-squared fit statistics/degree of freedom (1.803), and comparative fit index (0.92) supported the present model.

Practical/Policy Implication

The findings of this study underscore the critical need for targeted interventions and support systems to address the devastating impact of COVID-19-related anxiety and uncertainty on students’ academic behavior and performance. Educational institutions should prioritize the development and implementation of mental health resources, offering counseling services to help students cope with stress and anxiety. Additionally, efforts should be made to bridge the digital divide by providing necessary technological resources and support for students who face barriers to remote learning. Flexibility in academic policies, clear communication about changes in course structures, and proactive measures to alleviate workload pressures can contribute to a more supportive learning environment. Recognizing the unique challenges posed by the pandemic and implementing practical solutions will be essential in ensuring the well-being and academic success of students during these challenging times.

Underlying data

Figshare: Covid Anxiety and Uncertainty of Classes _ Devastating effect on Academic Behavior of Students (Responses) - Form Responses 1.csv. https://doi.org/10.6084/m9.figshare.21286266.v1 (Rahiman & Kulal, 2022).

This project contains the following underlying data:

Data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).