key: cord-1052953-vk80tbn3
authors: Collins, Francesca E.
title: Measuring COVID-19-related Fear and Threat in Australian, Indian, and Nepali University Students
date: 2021-01-27
journal: Pers Individ Dif
DOI: 10.1016/j.paid.2021.110693
sha: dec86ce8d0b273c9f2fbddd8329e7444ef6aaf80
doc_id: 1052953
cord_uid: vk80tbn3

The present study describes the adaption and validation of a brief measure of contagion-related fear and threat in Australian, Indian, and Nepali university students in Australia at the height of the first wave of the COVID-19 pandemic. Adapted from Ho, Kwong-Lo, Mak, and Wong’s (2005) SARS-related fear scale, the Contagion Fear and Threat Scale (CFTS) was rapidly adapted to capture the experience of COVID-19 pandemic-related fear. The factor structure and validity of the 6-item scale were established among Australian (n=154), Indian (n=111), and Nepali (n=149) university students studying in Australia in May-June 2020. Factor analysis revealed two 3-item factors in the Australian student sample: Fear of Infection and Existential Threat. These factors were confirmed in the Indian and Nepali student samples and mirror those found by the Ho et al. (2005) in their original instrument. The convergent and discriminant validity of the full CFTS, Fear of Infection, and Existential Threat scales are indicated via correlations with established measures of depression, anxiety, stress, subjective wellbeing, and religiosity. Differences in the performance of the Fear of Infection and Existential Threat scales are considered in terms of the respective objective and subjective nature of the constructs.

The Australian lockdown commenced shortly before the author was to commence a broad investigation into the psychological wellbeing of domestic and international students in Australia. The author believed an exploration of psychological wellbeing at this time would be incomplete without direct examination of students' psychological responses to contagion.

Interest in psychological responses to contagion has increased in the early 21st century in the wake of Severe Acute Respiratory Syndrome (SARS), Middle East Respiratory Syndrome, and Ebola outbreaks (World Health Organization, n.d.) . Compared to other health threats, contagion engenders fear because of its imminence (i.e., its spread is inevitable), invisibility (i.e., its presence isn't apparent), and transmissibility (i.e., it spreads easily and without discrimination; Pappas et al., 2009 ). This contagion-related fear is heightened by perceptions of the speed of transmission and exposure to contagion-related misinformation (Amin, 2020; Pappas et al., 2009) .

A search of the pre-2020 literature revealed a dearth of quantitative measures of contagionrelated fear. Research emerging from the most recent comparable epidemic, the SARS epidemic of 2002-2004, mainly used semi-structured interviews and general measures of fear, trauma, and psychological wellbeing (Ho et al., 2005; Stuijfzand et al., 2020) . SARS research also tended to focus on the experiences of healthcare professionals (Ho et al., 2005;  theoretically unrelated constructs (i.e., religiosity). Convergent validity was further examined in terms of via examination of factor loadings and average variance extracted.

Participants were domestic and international university students studying in Australia between 5 May 2020 and 7 June 2020, a period of nation-wide COVID-19-related selfisolation and social restriction in Australia. Data from three culturally distinct university student groups were analyzed: Australian domestic students; Indian university students studying in Australia; and Nepali university students studying in Australia. All participants were aged 18 or over and were fluent in written and spoken English as required for admission to Australian universities.

The Australian group comprised 154 Australian domestic (i.e., citizens and permanent residents of Australia) university students. They ranged in age from 18-60 years with a mean age of 25.45 years (SD = 9.04). The Australian group included 99 women (64%), 50 men (32%), 3 participants who identified as non-binary (2%), and 2 participants who did not indicate their gender (2%). will be assigned to SARS wards). In the present study, the factor originally labelled 'Insecurity' was renamed 'Existential Threat' to better reflect the non-specific sense of threat to one's existence, rather than fear, described in these items.

For the present study, the six items making up Ho et al.'s (2005) fear of infection and existential threat factors were retained, providing a focused, 6-item measure. Each item included the stem, The current COVID-19 pandemic makes me…; the name of the disease or outbreak can be changed as necessary. Participants responded to each item on a 4-point Likert scale, as used in the original study, where 0 = Definitely false and 3 = Definitely true.

A short form of the 42-item Depression, Anxiety and Stress Scales (DASS; Lovibond & Lovibond, 1996) , the Depression, Anxiety and Stress Scales-21 (DASS-21) is a 21-item scale comprising three 7-item subscales measuring symptoms of depression (e.g., I felt that I had nothing to look forward to), anxiety (e.g., I felt I was close to panic), and stress (e.g., In non-clinical populations, DASS-21 total scores and subscale scores are typically at the very low end of the scoring range. DASS-21 total scores have been reported between 8.9 -11.51 (Crawford et al., 2011; Sinclair et al., 2012) . Mean subscale scores have been reported ranging from 1.57 -5.03 for Depression, 1.44 -4.51 for Anxiety, and 1.79 -7.27 for Stress (Lee, 2019; Praharso et al., 2017; Sinclair et al., 2012; Zanon et al., 2020) .

The 

The 5 

Analyses were carried out in SPSS v.26 (descriptives, normality, correlations, reliability, validity, exploratory factor analysis) and AMOS v.27 (confirmatory factor analysis).

Significant Shapiro-Wilk statistics suggested all six contagion-related fear items were nonnormally distributed, however, examination of the Q-Q plots indicated all distributions were

normal. An exploratory factor analysis was conducted on the Australian data in SPSS v.26

with Maximum Likelihood and Promax rotation. A determinant value of 0.16, greater than the required 0.00001 was obtained and, for all six items, tolerance values were greater than .2

and VIF values less than 10, indicating an absence of multicollinearity (see Table 1 . The other variables reported in this study (i.e., depression, anxiety and stress, subjective wellbeing, and religiosity) were selected prior to the emergence of the pandemic so were not chosen specifically for their proposed relationship with the contagion-related fear items.

However, depression, anxiety, stress, and subjective wellbeing scores were expected to be associated with (and, thereby, demonstrate the convergent validity) contagion-related fear items as they represent negative, and in the case of subjective wellbeing, positive affective states. As religiosity is theoretically unrelated to affective state (Koenig and Bussing, 2010), it was used to demonstrate the divergent validity of the contagion-related fear items.

To determine factor structure of the six contagion-related fear items, an exploratory factor analysis was carried out on the Australian group data (n = 154 All items correlated significantly with one another at ≥.3, again, with the exception of this item. Any gain in internal consistency achieved by the removal of this item would be offset by the reduction of one factor to, at best, an uninterpretable two items. For this reason, all items were retained, and the six items deemed suitable for factor analysis.

The exploratory factor analysis was carried out with Maximum Likelihood extraction and Promax rotation, as recommended by Carpenter (2018) = 3.19, p = .53, and was retained for further investigation.

Confirmatory factor analysis of the two-factor solution was conducted using with Maximum Likelihood extraction on the data from the Australian, Indian (n = 111; KMO = .77; Bartlett's Table 2 ). The two-factor model demonstrated excellent goodness of fit across all three groups (see Table 3 ). Multigroup CFA revealed measurement invariance across the three groups at the configural and metric levels but not at the scalar level (see Table 4 ). Table 3 . CFA goodness of fit statistics for the two-factor model in Australian, Indian and Nepali university student groups. Note: x 2 = chi square; df = degrees of freedom; CFI = comparative fit index; RMSEA = root mean square error of approximation; CI = confidence interval; Δ = change; * p < .05.

Correlations for CFTS total and factor scores and the DASS-21 total and subscale scores, Subjective Wellbeing scores, and DUREL scores are shown in Table 5 .

Across all three groups, mild to moderate correlations were found between the CFTS total, Fear, and Threat scores and DASS total, Depression, Anxiety, and Stress scores with the exception of Fear and Anxiety and Fear and Stress in the Nepali sample. Across all groups, Subjective Wellbeing was mildly to moderately negatively associated with CFTS total and Threat scores. Subjective Wellbeing was also negatively associated with Fear in the Australian group. Across all groups, CFTS total, Fear and Threat scores were unrelated to 

It is notable that, across all groups, Fear scores were higher than Threat scores and Threat scores were more strongly related to DASS-21 total and subscale scores and Subjective

Wellbeing scores than were Fear scores. This may be explained by the nature of the two CFTS subscales. The object of Fear of Infection is specific and external to the self (i.e., COVID-19 infection). Existential Threat, however, has no specific object, rather it is an unanchored and subjective experience of threat to one's own existence, albeit triggered, in this case, by the COVID-19 pandemic. The moderate relationship between Threat and DASS total, Depression, Anxiety and Stress scores can be understood as capturing the subjective and non-specific character of these constructs (Lovibond & Lovibond, 1996) . Similarly, Subjective Wellbeing represents satisfaction with a range of life domains rather than a specific domain such as Standard of Living or Personal Health.

As all participants were studying in Australia at the time of data collection, features specific to the Australian experience of this first wave of COVID-19 are likely to have moderated CFTS full, Fear, and Threat scores. At the time, restrictions were swiftly implemented, actively enforced, and broadly adhered to in Australia (Australian Government, 2020).

Furthermore, Australian infection rates were low by global standards, all participants had access to socialised, high-quality healthcare, and the capacity of health system was not at risk of being overwhelmed (Australian Government, 2020).

Futhermore, data were collected around six weeks after Australia's first wave peak of daily cases, when new infections were typically in the single digits. At that time, COVID-19 hadn't taken a hold in India and Nepal which were still weeks away from their first wave peak of daily infections (Roser et al., 2020) . So, while COVID-19 was recognized (Ho et al., 2005) as J o u r n a l P r e -p r o o f Journal Pre-proof a significant threat in Europe and the Americas during May and June 2020, the threat was objectively less for Australian, Indian and Nepali students in Australia and their families in India and Nepal.

Participant age may also have affected CFTS full, Fear, and Threat scores. In Australia, although people in their 20s (90% of the current sample) are more likely to contract the disease, people over the age of 70 are far more likely to die from the disease. Compared to older age groups, the present sample may not have perceived COVID-19 to be lifethreatening (covid19data.com.au, 2020; .

The CFTS should be administered to people from a range of age groups and countries at different stages of pandemic to determine its psychometric performance across the lifespan and across pandemic waves. Translation of the CFTS is encouraged to determine whether the scalar measurement noninvariance found here is an artefact of speaking English as an additional language.

Finally, the psychometrics of the English and translated CFTS should be explored alongside those of the other recently developed measures of COVID-19-related psychological responses in relation to both COVID-19 and other impending epidemics such as measles and polio (World Health Organization, 2020) . Each of these measures has been devised for a distinct purpose; it is anticipated the more COVID-focused measures will have limited usefulness beyond the current pandemic.

Coronavirus (COVID-19) in Australia. Coronavirus (COVID

Percentile Norms and Accompanying Interval Estimates from an Australian General Adult Population Sample for Self

Mental health and international students: Issues, challenges and effective practice

Fear of Severe Acute Respiratory Syndrome (SARS) Among Health Care Workers

Subjective wellbeing and 'felt' stigma when living with HIV

Personal Wellbeing Index

Personal Wellbeing Index: A Cross-Cultural Measurement Invariance Study Across J o u r n a l P r e -p r o o f Four Countries

Australian Normative Data-Subjective Wellbeing

Confirming the Tripartite Structure of the Duke University Religion Index: A Methodological Approach

The convergent, discriminant, and nomological validity of the Depression Anxiety Stress Scales-21 (DASS-21)

Manual for the Depression, Anxiety, Stress Scales. The Psychology Foundation of

Rasch analysis of the Personal Wellbeing Index

A mixed-methods approach to understanding the role of religion and spirituality in healthcare provider well-being

Stressful life transitions and wellbeing: A comparison of the stress buffering hypothesis and the social identity model of identity change

The health and wellbeing of international students at an Australian university. Higher Education

Coronavirus Pandemic (COVID-19)

Testing measurement invariance of the Depression, Anxiety, and Stress Scales (DASS-21) across four countries

Psychometric Evaluation and Normative Data for the Depression, Anxiety, and Stress Scales-21 (DASS-21) in a Nonclinical Sample of U

COVID-19 increases risk to international students' mental health. Australia urgently needs to step up. The Conversation

Running head: COVID-19 FEAR IN UNIVERSITY STUDENTS 28

Reducing distress in university students: A randomised control trial of two online interventions

Assessing the usefulness of the God Questionnaire

Psychological Impact of an Epidemic/Pandemic on the Mental Health of Healthcare Professionals: A Rapid Review

Development and initial validation of the COVID Stress Scales

One hundred days of the coronavirus crisis

Charting the COVID-19 spread in Australia

The Measurement of Subjective Wellbeing: Item-Order Effects in the Personal Wellbeing Index-Adult