How did the COVID-19 pandemic change the career identity and academic success of computer science students?

Impact of COVID-19 pandemic on mental health

The impact of COVID-19 extends beyond physical health. The emotional distress caused by the events regarding the pandemic will have an effect on generations exposed to the COVID-19 pandemic (O’Connor et al., 2020; Pfefferbaum & North, 2020; Huang et al., 2020; Cucinotta & Vanelli, 2020; WHO, 2023; WHO, 2024). The effects of the pandemic on mental health are complex and multi-layered (Talevi et al., 2020): the unpredictability of the pandemic’s progression, disrupted everyday lifestyle and social distancing left an undeniable impact on the well-being of humanity and life in general. According to a meta-analysis conducted by Salari et al. (2020) the pandemic has impacted both public health and psychological and mental disorders. Several studies have highlighted the psychological effects of the outbreak, noting an increase in depression, anxiety, suicide risk, and post-traumatic stress (O’Connor et al., 2020). The psychological consequences of lockdown and social distancing were found to be more significant than the impact of getting sick with the virus and the effects of the infection on physical health; underscoring the importance of addressing mental health in regard of the pandemic (Holmes et al., 2020; Gao et al., 2020; Taylor, 2022).

Impact of the Covid-19 pandemic on mental health amongst university students

The COVID-19 pandemic has profoundly impacted students’ and academic professionals’ education pathways and career development. Research highlights (Smith et al., 2022) that the abrupt shift to online learning and teaching created significant disruptions, particularly in identity formation and professional growth. Studies show varying degrees of adaptation among students and staff. People experienced challenges such as increased stress, workload, and difficulty in delivering certain core subjects effectively. A study found that universities rapidly shifted to online teaching during the COVID-19 pandemic and it significantly disrupted academic roles and identities (Smith et al., 2022). They examined the effect on the identity of British university staff members using the identity theory (n = 33). Three main themes emerged: identity disruption, sensemaking, and nostalgia for pre-pandemic practices. The study shows the importance of directly hearing from academics amidst growing concerns over academic identity during a time of stress and insecurity. The findings suggest that the upheaval in teaching practices led to a significant need for adaptation and academic roles reshaped, a new emphasis on online pedagogies and interactions with students and peers. Crick and their team (2020) examined the impact of “emergency remote teaching” on CS educators in the UK, relying on data from a large-scale survey (N = 2,197) conducted shortly after the lockdown in March of 2020. The survey included 214 CS educators from various educational settings. These educators generally held more positive attitudes towards the shift to online learning, teaching, and assessment compared to their peers in other disciplines, with consistent perceptions across schools, colleges, and higher education institutions. Despite recognising the opportunities presented by the focus on digital skills, educators raised concerns about increased workload, effective pedagogy, and job security. Specific issues for computer science included the challenge of adequately teaching core subjects like mathematical foundations and programming, and the impact on different types of formal assessments and evaluations. To sum up, academic staff experienced shifts in their roles and pedagogical practices.

Other studies had analysed the phenomenon from students’ perspective: Chowdhury et al. (2022) found that university students struggle in developing successful careers after graduation. The COVID-19 pandemic has intensified these struggles, leading to increased stress and depression. The study aimed at evaluating the prevalence and factors associated with depression and stress related to career perspectives of university students from Bangladesh during the pandemic (n = 516). The findings of the online survey revealed that 80.2% of the students experienced mild to severe depression, and 77.3% reported low to moderate stress. Students who believed that delaying graduation due to COVID-19 would reduce their job prospects were more likely to be depressed. Another study found that amid the COVID-19 pandemic, medical schools globally faced disruptions, thus significant adaptation in curriculum delivery were necessary that impacted the professional development of medical students (Findyartini et al., 2020). Using a phenomenological approach, the study analysed 80 reflections, considering students’ year of study, gender, GPA, and educational stage (preclinical or clinical). Three main themes emerged: students’ adaptation processes and coping strategies during the pandemic, adjustments in learning processes, and their perceived roles as medical students amid the crisis. The findings underscored the implementation of adaptive learning tools by students and highlighted potential shifts in socialisation processes crucial for professional identity formation due to pandemic-related disruptions. The study emphasises the role of medical schools in supporting students’ adaptations and fostering their professional identity amidst such unprecedented challenges.

Early-career researchers, especially those from marginalised groups, reported disproportionate negative impacts on their work and mental health. While previous studies have predominantly focused on the experiences of women and parents, there has been limited attention to how the pandemic disproportionately affects academic work based on factors such as race, disability status, sexual identity, first-generation status, and career stage (Douglas et al., 2022). A research stratified random survey of early-career academics across four science disciplines (N = 3,277) to investigate socio-demographic and career stage differences in the impact of COVID-19 on seven work outcomes (Douglas et al., 2022): changes in research progress, workload, concerns about career advancement, and support from mentors, as well as work disruptions due to physical health, mental health, and caretaking responsibilities related to COVID-19. The findings reveal that scholars from marginalised and minority groups across early career stages reported more negative work outcomes due to COVID-19. Individuals with physical or mental disabilities were particularly affected across all seven work outcomes. Women, primary caregivers, underrepresented racial minorities, and first-generation scholars reported increased disruptions related to physical health symptoms and additional caregiving responsibilities compared to their more privileged counterparts. Doctoral students experienced more disruptions from life challenges than other early-career researchers, especially related to health issues, while assistant professors reported more negative changes in research progress and increased workload. These results suggest that the COVID-19 pandemic has disproportionately harmed work outcomes for marginalised and minoritised early-career researchers (Alonso et al., 2019; Auerbach et al., 2016; Dhar et al., 2020; Dunajeva, 2022; Kálmán et al., 2021; Limón-Vázquez et al., 2020; Morvay-Sey et al., 2022; Salimi et al., 2023; Son et al., 2020; Torales et al., 2020). Institutional support and curriculum design are pivotal in retaining students. High failure rate courses and the lack of academic support services are significant predictors of late dropout (Dyrberg & Holmegaard, 2019; Kiss et al., 2019; Li et al., 2023; Salazar-Fernandez et al., 2019; Shmeleva & Froumin, 2020; Vooren et al., 2022; Whitcomb & Singh, 2021).

Young adulthood and career identity

One of the main tasks during the transition to adulthood is the formation of identity across various life domains (Arnett, 2000). The concept of identity is associated with Erikson, who described it as the ability to perceive oneself as a complete entity, meaning that identity represents an achieved wholeness (Erikson, 1968).

In Erikson’s (1968) psychosocial model, eight stages of development were delineated, each characterized by a normative developmental crisis that individuals must confront during that stage. According to Erikson’s theory, adolescence is marked by intense turmoil and the search for identity, while young adulthood is characterized by intimacy, work, and establishing a family. He sets the endpoint of this stage at around 25 years old, representing the possibility of developing intimacy as a new skill and competence, contrasting with isolation as a less adaptive outcome.

Arnett (2000), however, argues that the traditional Eriksonian stages are undergoing transformation. According to him, the period of emerging adulthood, spanning ages 18 to 30, is situated between adolescence and young adulthood. This is a time when young individuals search for their vocation and stable relationships, eventually settling into adulthood with commitment and stability. Due to various societal changes, the beginning, end, and duration of life stages can shift, and nowadays, the process of becoming an adult is so prolonged that Arnett (2000), along with several other authors (Kenniston, 1968; Vaskovics, 2000; Klein, 1990; Côté, 2000, cited in Murinkó, 2010), refers to this specific period as a distinct life stage in its own right. Arnett proposes that a defining feature of emerging adulthood is the opportunity for young adults to extend the exploration that began in adolescence across both professional and personal areas. They can experiment with different careers and relationships without facing the commitments and responsibilities typical of adulthood, as they no longer consider themselves children but do not fully see themselves as adults either (Arnett, 2000).

Building on Erikson’s theory, James Marcia (1966) developed the concept of identity statuses. According to Marcia, young people go through an identity crisis during which they need to make decisions in various life domains that are crucial for their future. Marcia identified four identity statuses based on two dimensions: crisis/exploration and commitment.

Crisis/exploration refers to adolescents re-evaluating their parents’ choices, actively seeking their future possibilities, and searching for the most suitable alternative for themselves. The commitment dimension means that the individual commits to certain goals, values, and views.

Marcia’s model can be viewed as a developmental framework in which the identity statuses progress from the least mature to the most mature in the following order: identity diffusion, foreclosure, moratorium, and achieved identity.

Individuals with diffused identity are characterized by both the lack of a distinct exploratory period and a lack of commitments, so identity diffusion is considered the lowest identity status. Those in foreclosure are committed to certain values but have not experienced an identity crisis, as they adopt identity elements from parents or significant others. One step forward, we find individuals in the moratorium state, who are experiencing an identity crisis and struggling between alternatives, but no commitment has yet been made. Finally, achieved identity means that young people have gone through a period of active exploration and have committed to a profession, political, or religious view (Marcia, 1966; Kroger & Marcia, 2011, cited by Elekes et al., 2018).

According to Waterman’s (1985) research, the proportion of individuals in identity diffusion gradually decreases with age, while the proportion of those who achieve mature identity increases across all examined identity domains. However, later studies found that a person’s identity statuses can vary across different identity domains (Archer, 1988, cited by Meijers, 1998). For instance, one might have an achieved identity in political or religious domains but hold a lower status in professional domains. This is particularly true for higher education students, as university experiences can challenge previous career choices and initiate new exploration in young adulthood.

In defining career identity, Meijers (1988) built upon the explanations of earlier theorists (Holland, 1985; Vondracek, 1992, as cited by Meijers, 1998) and proposed that career identity can be understood as a network or structure of meanings. In this structure, the individual consciously connects their motivations, interests, and competencies with suitable career roles. This structure is continuously evolving due to learning processes prompted by real-life experiences (which extend beyond just work and working). Career identity is not merely the sum of these experiences but rather the integration of these experiences into meaningful or useful frameworks.

The formation of identity, however, does not occur in a vacuum; it is essential to consider socio-cultural factors and environmental influences as well (Arnett et al., 2011). In this context, as studied by Bobba et al. (2023), the COVID-19 pandemic, as a current environmental influence, has also affected the development of identity. This research explored differences in identity processes and time perspectives between university students and workers across two different time periods (pre-COVID-19, with data collected in 2017, and during the pandemic, in 2021). The participants were young adults between the ages of 19 and 29, divided into two groups: university students and workers. Workers in 2021 demonstrated a notably more stable vocational identity, marked by stronger commitment, deeper exploration, and less reconsideration compared to those in 2017. Surprisingly, students across the two cohorts did not exhibit a similar trend. The authors explain that these results could be attributed to the differential impacts of the pandemic on the lives of workers and students. As universities quickly switched to online courses, concerns over the pandemic’s immediate and long-term effects on occupational conditions became prominent among workers. Therefore, strengthening commitment to vocational identity may have served as an adaptive response to cope with the uncertainties brought about by the pandemic.

In our study, we would also like to examine whether COVID-19 affects the occupational identity of our sample (informatics students).

Career Identity and Its Impact on Dropout Rates

Career identity is central to understanding dropout rates among CS students. Career identity refers to aligning a student’s self-concept with their chosen career path, including their sense of belonging in the field, their perceived fit, and their commitment to pursuing a career in that domain. A strong career identity can enhance a student’s motivation, resilience, and persistence in the face of academic challenges.

A strong sense of belonging in the academic environment is important for the development of a solid career identity. Research indicates that students who feel integrated into their academic community are more likely to persist in their studies. Höhne & Zander (2019) found that belonging uncertainty is a significant predictor of dropout intentions among first-semester CS students. Students who do not feel they belong in the CS community may struggle to develop a strong career identity, increasing their likelihood of dropping out.

Self-efficacy, or the belief in one’s ability to succeed, is closely linked to career identity. Students who perceive themselves as competent in CS are more likely to identify with the field and persist in their studies. Peteranetz et al. (2018) highlighted that students’ computer science-related career aspirations and perceived instrumentality of CS courses significantly influence their academic achievement and retention. High self-efficacy contributes to a stronger career identity, supporting academic persistence.

Women and underrepresented minorities often face additional challenges in developing a career identity in CS due to stereotypes and a lack of role models. Höhne & Zander (2019) emphasized that female students with a threatened social identity in the male-dominated field of CS are more likely to experience belonging uncertainty and dropout. Initiatives aimed at improving gender diversity and inclusivity can help mitigate these effects and support the career identities of all students.

The alignment between students’ career aspirations and their professional identity in CS is a critical factor for retention. Kapoor & Gardner-McCune (2018) explored factors behind CS students’ desire to leave the major, highlighting that professional identity development plays a significant role. Students who can envision a clear and desirable career path in CS are more likely to persist in their studies. This includes having access to career counseling and exposure to various career opportunities in the field.

Research question

This research is crucial for understanding the long-term effects of COVID-19 on the career identity of young adults, which can inform future educational and career support strategies.

By examining how the pandemic has influenced career identity development, universities and policymakers can better support students in navigating their career paths, particularly in fields like IT where dropout rates are high. Insights from this study can help enhance educational practices and programmes by identifying the needs of students who struggled to develop a mature career identity during the pandemic. To address the limitations of current research and to develop effective, evidence-based interventions, it is crucial to conduct large-scale, comprehensive studies on student career identity. Such research should include diverse socio-demographic groups, employ mixed methodologies, and provide longitudinal data to capture the evolving nature of career development. This will ensure a precise and actionable understanding of the challenges and opportunities faced by students in their career journeys. The findings can aid in formulating policies that address the gaps in career development support that were exacerbated by the COVID-19 pandemic, ensuring that future crises do not similarly hinder student progress. The research can identify groups of students who are more vulnerable to dropping out due to an early closing identity, allowing for targeted interventions to prevent attrition and promote academic success.

By addressing these questions, the research aims to provide comprehensive insights into the intersection of career identity development and the unprecedented challenges posed by the COVID-19 pandemic.

Hypothesis

1. Dropout increased during COVID-19 pandemic among CS students. (Students who arrived in the year of Covid have a higher dropout rate than students who arrived at other times.)

2. Upon entry, the achieved identity state is significantly more typical for the career identity of graduate students than other career identity states.

3a. Achieved identity is higher for pre-Covid students than for post-Covid CS students.

3b. For post-Covid students, the moratorium identity is higher than it was before Covid.

3c. For post-Covid students, the diffuse identity is higher than it was before Covid.

3d. The rate of early closures is lower in the pre-COVID population than in the post-COVID population.

Participants

The survey was conducted among first-year computer science BSc students. The total sample consisted of 3653 participants, of whom 510 were female, 3114 were male, and 29 had no gender data. The average age of the participants was 19.36 years. Informed consent was obtained from all participants in written form prior to their participation in the study.

Procedures

The data were collected using an online questionnaire package in the context of a university course on the integration of first-year undergraduate students into the university, under anonymity, during the autumn semesters between 2016–2023.

For the hypotheses, the sample was divided into two groups, pre-Covid (students from the 2016 to 2019 cohorts, 1794 students) and post-Covid (students from the 2020 to 2023 cohorts, 1858 students).

IBM SPSS Statistics version 28 was used for the analysis and MS Excel for descriptive statistics.

Questionnaire

Occupational Identity Scale (OIS, 1987)

The questionnaire included in our analysis is the 28-item Melgosa Occupational Identity Scale (OIS, 1987). The Hungarian adaptation of the scale is attributed to Zsuzsanna Andrási (1995).

The test consists of statements related to the identity states of achievement, diffusion, moratorium, and foreclosure, which respondents evaluate on a five-point scale. Accordingly, the scale includes four subscales corresponding to these identity states: Achievement (7 items), Moratorium (8 items), Foreclosure (7 items), and Diffusion (6 items).

The reliability analysis of the entire sample (Cronbach’s alpha: 0.864) and the individual subscales indicates that they are reliable (Achieved Identity: Crα = 0.777, Moratorium: Crα = 0.763, Foreclosure: Crα = 0.842, Identity Diffusion: Crα = 0.641).

Data analyses

An independent sample t-test was used to compare the means of two groups: pre-Covid19 pandemic and post-Covid19 pandemic cohort. Levene’s test was used to test for homogeneity of variance. The comparison of our means of several variables was tested by one-sample analysis of variance, and for variables that did not meet the criterion of normal distribution, a Kruskal-Wallis test was performed. In addition, we used descriptive statistics when examining the students’ final scores.

Interpretation of results

As a result, our first hypothesis was confirmed. We found that, contrary to the trends of the previous cohorts, the drop-out rate for students who enrolled during the Covid year (2020) increased slightly (10.28%) over half of the program period (three semesters) but then decreased again for the following two cohorts (2021: 5.45%, 2022: 5.01%). This result aligns with the literature (e.g. Höhne & Zander, 2019).

The results (H3a) showed that students in the pre-Covid cohort (18.93 points) had significantly higher scores on achieved identity than those in the post-Covid cohort (18.36 points) (t₍₃₃₇₆₎ = 3.130, p = 0.002), confirming our hypothesis. Career identity achievement (H3a) was assessed using an independent samples t-test on tr-and post-Covid samples. The findings reinforced that students in the pre-Covid cohort had significantly higher achieved identity scores than their post-Covid counterparts (t₍₃₃₇₆₎ = 3.130, p = 0.002). Unlike other literature sources (Bobba et al., 2023), which found no significant difference between the 2017 and 2021 student samples in terms of vocational commitment, in-depth exploration, and reconsideration of commitment—suggesting that there was no substantial change in these factors among students before and after COVID—our results indicate a significant difference in career identity measured before and after COVID, with higher levels of achieved identity among students in the pre-COVID population. Among the possible explanations, it has been suggested that, before COVID, students had more opportunities to explore career-related experiences and participate in programs influencing career choice in person (e.g., open days, expos). In many countries, the transition to online education in public schooling was not seamless, and in some regions, it further widened the gap for certain social groups (e.g., Arnett et al., 2011; Bobba et al., 2023).

Interpretation of career identity status

The degree of career identity moratorium status (H3b) was examined by comparing pre- and post-Covid samples. The results showed that students in the pre-Covid (22.67 points) cohort scored lower on identity moratorium than those in the post-Covid cohort (23.05 points) (t₍₃₃₇₆₎ = −1.938, p = 0.053). Although this result is not statistically significant, it is at the trend level. Therefore, our hypothesis was confirmed. The degree of career identity diffuse status (H3c) was also examined. The results indicated that students in the pre-Covid cohort (15.74 points) scored lower on identity diffusion than those in the post-Covid cohort (16.65 points) (t₍₃₃₇₁₎ = −5.938, p < 0.001). The result is statistically significant, confirming our hypothesis. After Covid-19 pandemic, the number of individuals with a diffuse identity increased compared to the pre-Covid period. These findings suggest that the disruptions caused by the pandemic, such as limited career exploration opportunities and uncertainties in education and the job market, may have contributed to greater identity diffusion among students. In contrast, the increase in moratorium status at the trend level may indicate that more students remained in an exploratory phase without making form career commitments.

The degree of foreclosure in career identity (H3d) was analyzed using the Mann-Whitney U-test due to the non-normal distribution of responses on the early closure scale in the pre- and post-Covid samples. The results showed no significant difference between the pre-Covid (Mdn = 1706.74) and post-Covid (Mdn = 1672.73) groups (U = 1388555, p = 0.311). Indicating that our hypothesis was not supported. While no significant differences were found in moratorium and early closure identity statuses, post- Covid participants tended to score higher than their pre. COVID counterparts. This suggests a shift in career identity patterns, particularly in the proportion of individuals with an early closure identity. Contrary to expectations, pre-COVID participants exhibited a higher rate of early closure identity than post-COVID individuals. These findings highlight the impact of the COVID-19 pandemic on career identity formation. Before the pandemic, the number of individuals with an achieved career identity was higher, suggesting that COVID-19 may have disrupted career decision making processes and identity consolidation. Furthermore, the study also explores the relationship between career identity development and dropout tendencies, emphasising the need for further investigation into how pandemic-related uncertainties influence students’ academic and professional trajectories.

Fewer students achieved a stable career identity, and also the attrition rate increased in our sample. Only diffuse identity is significantly higher post-COVID compared to pre-COVID. Students struggled to connect with their profession to a sufficient extent. For those who entered university after the COVID outbreak, the impact on the achieved identity was small, but significant, with only a slight difference in scores. However, the difference in diffuse identity was more pronounced (15.76–16.65). The specific age at which students experienced COVID (Whether they were 15, 16, or 17 years old) has minimal impact on achieved identity, with a difference of only 0.57 in scores.

Recommendations for HEI insitutions

Further research is needed to accurately interpret the effects. There are already university experiences, but starting university is a different process- one that unfolds over several years, allowing us to assess the impact of COVID-19. As students begin learning professional subjects, they start engaging with their chosen field.

There were intermittent disruptions due to COVID-19, Which served as challenges but also as opportunities for acquiring information. However, the significance of these disruptions varies across professions. The social and personal role models play a crucial role for IT professionals; the younger they were when affected by COVID-19, the more they missed out on these influences. We specifically examine the youngest cohort of 15-year-olds- those who were in the process of deciding whether to pursue studies in this field. The real issue is that the earlier someone experienced the effects of COVID-19, the more uncertain they tend to be about their career identity. This raises new research questions, as the lack of a well-formed career path and commitment has significant implications for future employment. It is strongly recommended to introduce measures not only at the university level but across all educational settings to strengthen soft skills and professional identity. Providing as much relevant information as possible about the chosen profession can positively influence students’ confidence in their decisions. The ultimate goal is to minimise these effects and prevent them from becoming a widespread generational issue.

Restrictions and suggestions for further studies

The sample is made up of students taking CS courses, which is the study’s initial restriction. The course’s greater popularity among male students is one noteworthy feature. When making inferences about the population, it is important to consider this gender disparity.

The second drawback is that each cohort’s identification scale measurements were made during their first semester of college. As a result, the longitudinal intrapersonal changes in identity could not be monitored. The identity status of first-year students prior to, during, and following the Covid-19 pandemic was determined by the current study.

The third crucial restriction is that cohorts are determined by academic year group rather than by birth date. Although a large percentage of first-year students are between the ages of 19 and 20, this age group is not the only one to enroll in college (average age: 19.36 years).

In the modern era, the question of career identity is a major concern. Reducing the rate of academic attrition and assisting young people in choosing a career route are important goals. To get around the aforementioned restrictions, more research is necessary.

Additionally, if comparable research were carried out on other university samples, the results’ interpretation would be more expansive. In order to create more potent preventive and intervention programs, our research team will keep investigating the psychological causes of academic dropout.