The novel coronavirus disease 2019 (COVID-19) outbreak was declared a pandemic quickly because of its rapid transmission rate and high morbidity and mortality. As of April 19, 2022, COVID-19 has infected more than 24,000,000 individuals and has caused more than 28,000 deaths in South Korea (hereafter Korea) [1]. Fear and anxiety have surged in people because of the uncertainty about the return to a normal social life, the high infectivity rate of COVID-19, the rapid spread of its variants, and the limited availability of effective

treatment [2]. The COVID-19 pandemic appears to continue and exercises a long-term and profound impact on individual health. Therefore, researchers and clinicians have expressed concerns about the potential adverse effects of COVID-19 on the mental health of the general population, including university students [3-5]. Recent studies have evidenced sup-port for such apprehensions. The pandemic's impact on students' mental health is a particular cause for trepidation. For example, Cao et al. conducted a cross-sectional study in China and found that 24.9% of university students suffered from anxiety because of the COVID-19 outbreak [6]. An online survey of university students in Korea (n = 261) evidenced that 61% of students reported depressive symptoms and that their depressive symptoms intensified in tandem with the increased perception of restrictions in their social activities [7]. Other studies involving university students have found that students testified to high levels of COVID-19-related fear. This trepidation was positively associated with depression and anxiety [8, 9]. Since, many university students have been forced to shift their activities to the digital sphere, they inevitably confront unprecedented challenges because of the ongoing pandemic and its associated mandatory physical distancing measures [10, 11].

The fear of COVID-19 is a core factor for heightened stress and anxiety levels during the pandemic [3]. Specifically, people are afraid that they or their family members will become infected or sick [12]. Recent evidence suggests that the fear of COVID-19 infection could elicit anxiety reactions or exacerbate preexisting conditions related to somatic symptoms [13]. Other studies have similarly reported that the fear of becoming infected by COVID-19 could trigger an anxiety-related disorder [14, 15]. These scholarly findings indicate the importance of assessing the fear of COVID-19 so that its effects on health and mentally vulnerable populations can be predicted and controlled. A reliable screening tool to identify the specific fears of COVID-19 during the pandemic has thus become a significant public health issue, given the growing prevalence of mental health problems among university students. In fact, measures aimed at evaluating an individual’s psychological response to COVID-19 have recently begun to emerge. These instruments include the (a) five-item Coronavirus Anxiety Scale (CAS) [16], (b) four-item Obsession with COVID-19 Scale (OCS) [16], (c) 36-item COVID Stress Scale (CSS) [17], and seven-item Fear of COVID-19 Scale (FCV-19S) [18]. Ransing et al.’s review of the features of these four instruments suggested that existing instruments could achieve their maximum utility only through appropriate translation, cultural adaptation, assessment, and validation [19]. The FCV-19S, recently developed by Ahorsu et al. is the most popular of the mentioned measures because of its brevity along with the proven satisfactory psychometric properties of its different language versions. Indeed, the scale had been translated within three months of the original Persian FCV-19S being developed [20]. It has now been translated into 13 languages and empirically validated in diverse cultures.

The FCV-19S assesses fear sensed by individuals toward COVID-19 and helps healthcare providers design and implement appropriate interventions to reduce such fear [18]. In particular, the FCV-19S measures the psychological response to COVID-19 and examines associated behaviors (e.g., problematic use of the Internet, suicidal thoughts, bodily complaints, sleep-related difficulties, and psychological distress) [20]. Recent studies have confirmed the psychometric properties of FCV-19S with adequate results. However, inconsistent findings have been reported regarding the factor structure of the FCV-19S. The FCV-19S is assumed to present a unidimensional structure and its one-factor structure has been supported by research on several translated versions of the scale [21-24]. However, Iversen et al. proposed a two-factor structure in Norwegian samples to further expand such research [25]. Yet, other scholars have shown with Israeli and Spanish samples that the bifactor structure represents physiological and emotional responses [3, 26].

The FCV-19S has thus far been scantly used in the Korean context. Han et al. translated and verified the reliability and validity of the Korean version of the FCV-19S (KF-COVID-19S) among adults [27]. A single-factor structure of the FCV-19S was confirmed by the exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Meanwhile, a survey of normal adults and psychiatric patients in Korea found that FCV-19S incorporates two dimensions: physical and emotional fear factors [28]. These differences in findings suggest the need for further exploration and verification of the factor structure of the FCV-19S. Heightened mental health concerns have been noted among university students during the pandemic. Moreover, the self-perceived mental health of Korean university students has worsened over the last decades, and this situation was exacerbated during the pandemic [7]. Students sensing relatively high levels of anxiety and fear regarding infectious diseases such as COVID-19 tend to be less capable of coping with psychological distress and managing their daily lives [27]. From this perspective, an instrument such as the FCV-19S is essential for university counseling or medical services. Such a measure could help identify students with high levels of COVID-19-associated fear and lockdown anxiety early. Appropriate psychological interventions could accordingly be applied to foster their well-being [29]. The current study aimed to respond to this need by establishing the factor structure, reliability, and validity of the KF-COVID-19S in the context of Korean universities. To identify the factor structure of the KF-COVID-19S, it is necessary to use both EFA and CFA with independent datasets. Before analyses, the data were randomly split into two subsets (A and B). Exploratory factor analysis was performed on Subset A and a series of confirmatory factor analyses were performed on Subset B to test how well the model suggested by theory and the EFA fit the empirical data.

The current study purposed to evaluate the factor structure of the KF-COVID-19S in the Korean context, using a university's student population. The findings of this study contribute significantly to the evidence base pertaining to the underlying factor structure of the recently developed FCV-19S. The EFA and CFA revealed two correlated factors that corresponded reasonably to the dimensions of somatic symptoms and emotional fear when the KF-COVID-19S was examined using a student population. The instrument was originally developed as a unidimensional structure; however, evidence is accumulating vis-à-vis different structures that have been proposed and its dimensionality consensus is compromised. Nevertheless, the discrepancies in results reported in the literature are partially attributed to the use of (1) individuals with different cultural backgrounds and translated languages, (2) sample heterogeneity (e.g., children as compared with older adults) [2, 26], (3) different statistical techniques (e.g., EFA vs. CFA), or (4) sample size. The current study’s findings of a model with two correlated factors and an estimated between-factor association of.65 corroborate the outcomes of investigations with similar samples (i.e., general adults in a community) [25, 28].

Factor 1 assessed the physical or physiological symptoms of insomnia, palpitation, and sweating, and was labeled “somatic symptoms ”.Individuals may experience these physical and physiological symptoms when they sense severe concerns about contracting COVID-19 [28]. A recent study reported that panic attacks and generalized anxiety disorders triggering physical symptoms could be experienced when conditions became extremely severe during the COVID-19 pandemic [36]. Factor 2 was characterized as “emotional fear” to encompass other expressions of fear through adjectives such as “afraid,” “uncomfortable,” “nervous,” and “anxious” [28]. An online survey conducted in Korea during the COVID-19 pandemic classified 31% of the respondents as at risk for depression and categorized 23% of the participants as at risk of anxiety [37]. This result signifies that COVID-19 could cause significant psychological symptoms related to mental health. University students are particularly vulnerable to mental health problems. Campus closures, academic disruptions, and movement restrictions during the COVID-19 pandemic have caused heightened future-related anxiety and concerns in these young adults. Hence, this population requires further attention and support for its wellbeing during the COVID-19 pandemic.

COVID-19 will probably remain a significant worldwide public health problem for some time. Certainly, the development of infection-preventing measures, medical treatments, and vaccines must be prioritized. Nonetheless, the impact of COVID-19 on the mental health of students must not be overlooked, and in turn, the mental health impact on society should not be disregarded or underestimated [28]. The screening of university students for mental health difficulties should be prioritized at university counseling or medical centers [38]. In such a context, KF-COVID-19S presents as a promising instrument for campus environments. It can promptly and efficiently determine the degrees of anxiety caused by COVID-19 in a timely manner [28]. The FCV-19S was not developed as a tool for the diagnosis of mental disorders and calculating an accurate cut-off score may not be clinically meaningful. However, the KF-COVID-19S could denote an ideal measure for the sustainable large-scale screening for the severity of COVID-19-related anxiety in college student populations.

Cronbach’s alphas were computed at.78 for both somatic symptoms and emotional fear in congruence with Bitan et al.’s study [3]. An alpha value of >.70 indicates an acceptable level of reliability but some scholars contend that the recommended reliability value of the scales should be greater than.80 for some applied settings (e.g., individual assessment purposes) [39]. Hence, our results suggest that the somatic symptoms and emotional fear scales concerning COVID-19 can be used in combined forms as measured by FCV-19S to contribute to the broader clinical assessment of such psychological responses. Discriminant validity was maintained through moderate correlations between the physical and emotional fear subscales. The mean score for the current study’s sample was relatively high as can be expected because the survey was conducted during the peak of COVID-19 in Korea. At that juncture, the campus was shut down and the physical and social activities of students were restricted to a minimum. Hence, the results of the current study must be cautiously interpreted in light of the prevalence of COVID-19 in the country at the time of data collection.

The FCV-19S may be an appropriate instrument for the assessment of fear toward COVID-19 in vulnerable populations such as university students because of the ongoing stressors and uncertainties linked to the pandemic. University students represent a special social group with active engagement in life habits based on relationships, physical and university-related activities, travel, and communal gatherings [40]. However, the global outbreak of COVID-19 has dramatically changed the lives and relationships of students [10, 11]. Fear about being infected during the COVID-19 pandemic has caused heightened future-related anxiety and concerns in these young adults [22-24]. Therefore, multidimensional measures of psychological reactions (i.e., somatic symptoms and emotional fear toward COVID-19) experienced during the pandemic could be more apt for the Korean university context than tools offering a single-factor solution. In other words, we can use the two-dimensional factor model of the FCV-19S to assess symptoms or responses to the fear of COVID-19 in more detail.

However, a predominant issue that could derive from the one- or two-factor model dilemma is the uncertainty about the appropriate scoring procedure. Specifically, the use of two-factor scores may not be optimal in practice despite the two-factor model evincing a better fit to our data than the single-factor model. In the research context, such a difficulty would pose a serious multicollinearity problem since the factors were highly correlated [41]. Moreover, the reliability of the total scale was good; it was better than specific individual factor reliabilities. It is suggested taking all this into account, that using a single and total FCV-19S score would probably be advisable for research, practical, and clinical purposes. The utilization of the total score could help diagnose, evaluate, and monitor the severity of fear caused by COVID-19. Nonetheless, further research is needed in the future to compare the utility of the total and subscale scores in different clinical scenarios (e.g., screening versus monitoring). Healthcare providers could use the available information on how an individual experiences the fear of COVID-19 to design and implement more appropriate programs to alleviate the fear [18]. Subsequently, such programs could reduce the stigma, anxieties, and stress attached to the fear of COVID-19.

The findings of this study should be contemplated with due acknowledgment of the limitations of the investigation. First, the potential of sampling bias cannot be ruled out because the subject pool comprised a self-selected sample of students enrolled in a single university. It is unclear whether the factorial structure would equally apply to other populations (e.g., adolescents, older adults, or clinical populations). A follow-up, the re-verification study should be conducted with randomly sampled participants to confirm the results obtained by the current study. Next, the survey was administered online, and this platform could have made the measure inaccessible for some individuals. This limitation may have influenced the results of the current study, but only to some extent, given the target population of university students and allowing for the quarantine and physical distancing requirements. Another limitation this study shares with other investigations is its reliance on participant self-reporting. Such self-reporting data and their findings may be subject to external bias caused by social desirability. Future research initiatives could include a broader range of data sources. Final limitation of the study entailed the non-inclusion of other forms of validity, such as multigroup invariance and known group validity. These facets could have strengthened our current findings. Future research initiatives should also seek larger samples and account for variables such as gender, socioeconomic status, and physical and mental health state.

To conclude, the results of the present study empirically uphold the correlated two-factor structure of the KF-COVID-19S in Korean student populations enrolled in universities. Although the two-factor model offered the best fit for our data, the high correlation between the two factors, the higher reliability of the total scale, and the low reliability of the subscales make it more appropriate to compute and use the single FCV-19S score for most practical purposes. However, it may be beneficial if prospective research endeavors explicitly test this assumption.