Psychometric Properties of the Questionnaire Covid-19 Stress on the Romanian Community Sample

: The objective of the present research is to adapt the 36-item COVID-19 Stress Rating Scales to the Romanian population. It is a self-report scale used to assess the stress caused by COVID-19 through six dimensions: Danger, Socio-economic consequences, Xenophobia, Contamination, Traumatic stress, Verification. A total of 160 medically qualified participants, aged 19-59 years, completed a socio-demographic data sheet and the COVID-19 Stress Rating Scales. A high internal consistency was obtained for the 6 scales of the instrument, with Cronbach's alpha coefficients in the range 0.74-0.86. Exploratory factor analysis and Confirmatory factor analysis were also evaluated. The results indicate that the Romanian version of the CSS has good psychometric properties, being a suitable instrument for the assessment of stress caused by COVID-19.

The spread of virus causing Severe Acute Respiratory Syndrome (SARS -Cov 2) generated a global health crisis with consequences in all population categories. In march 2020, the World Health Organisation (2020) declared the state of pandemic due to the speed of infection and number of deaths caused by the virus.
Health and wellbeing were affected both physically and mentally. With the extension and deepening of the epidemic situation, people's worries facing this vital threat increased exponentially, leading to a significant psychological distress. There is a multitude of critical factors contributing to the the negative impact on population's mental state, quality of life and level of anxiety associated with COVID-19 (Guan et al., 2021;Zhong et al., 2021) Major efforts, in the medical and psychological areas, were made in order to explore the implications of the pandemic in the psycho-social domain and permitting later intervention strategies. Many recent studies aim to describe the mental state of people facing the Coronavirus pandemic and the factors that might influence the mental health (Shah et al., 2021;Vindegaard & Benros, 2020). Depression, anxiety, sleeping problems, fear of becoming infected from other people leading to xenophobia, traumatic stress symptoms like intrusive thoughts and nightmares have been reported worldwide.
The medical and psychological literature is rich and oriented towards general or specific population categories i.e. students, children and adolescents, pregnant women, medical workers. Researches constantly reveal direct and indirect neuropsychiatric consequences of the pandemic (Salari et al., 2020). The high prevalence of anxiety, stress and depression within front-line workers caring for COVID-19 patients has been repeteadly confirmed (Secosan et al., 2020).
In the atempt to determine the psychological profile of subjects, various study designs and different scales and interviews have been used. The large variety of evaluation instruments is reflecting the struggle of proffessionals all over the world to find an appropriate and comprehensive modality to describe the psychological characteristics of the targeted population.
In Romania, there is also a preocupation among the researchers in the medical and psychological domain to explore the relation between the pandemic and different psycho-social disturbances both in general population and among healthcare workers. Currently available psychological instruments or newly developed scales that adress specific aspects related to the pandemic crisis -depending on the study design and objectives -were used, in the main univeristy and medical centers. Most instruments were distributed through online surveys for an easy and general access.
A study performed in 2020 (Timisoara) used the DASS 21 scale (The Depression, Anxiety, and Stress Scale) and the The Insomnia Severity Index (ISI) in order to determine if the false information avalanche related to the pandemic had psychological consequences over the frontline healthcare workers (ICU and Emergency Department, physicians, and nurses) and showed that the medical workers who declared to be affected by this kind of news were more stressed, anxious, and suffered more from sleeping problems than healthcare workers who declared that didn't take into consideration the large amount of false information related to pandemic (Armean et al., 2021).
Another research, evaluated the psychological consequences (stress, anxiety, burnout symptoms) of the pandemic in the medical students (Ahorsu et al., 2020). They used 2 scales, adapted and translated in Romanian. One of them was previously designed during the outbreak of severe acute respiratory syndrome (SARS, 2003(SARS, -2005, to adress the healthcare workers of emergency departments in Hong Kong (Wong et al., 2005). The other instrument, The Fear of COVID-19 Scale (Stănculescu, 2021) was developed in 2020 during the COVID-19 pandemic. The Fear of COVID-19 Scale is a 7 item scale that aimes to evaluate the worries about becoming sick and is validated on Romanian population.
Another study used WHO-Five Well-Being Index (Spitzer et al., 2006)to assess depression and the Generalised Anxiety Disorder Scale (Cordoș & Bolboacă, 2021) for the anxiety assessment combined with a questionaire that explored in the romanian population how often the respondents saw COVID-19 related information using Social Media channels with an interesting outcome, opposite to most already available literature: there was no correlation between anxiety and depression and the context of lockdown and excessive COVID-19-related information exposure (Taylor et al., 2020).
Both systematic research and empirical clinical observations suggest that stress and anxiety are the major psychological manifestations related to de viral pandemic (Chung et al., 2021;Taylor et al., 2020;Xiong et al., 2020).
Several rating scales were developed since the outbreak of the pandemic. Researcher's efforts are often disparate, the tools used are various and often nonspecific. In this critical context, a psychological intrument for the specific assessment of stress and anxiety related to COVID-19, that is easy to use, practical and internationally applicable is essential.
The above sources justify the approach of translation, adaptation and validation of the new scale whose use in Romanian research would ensure the collection of comparable results.
This article reports the results of factorial, exploratory and confirmatory analysis of data collected in Romania following research. A methodological, quantitative, cross-sectional study was conducted to translate, adapt, and test psychometric properties of the Covid-19 Stress Scales to the romanian language.

Objective
The aim of this study was to assess the psychometric properties and the factorial structure of the cross-cultural version of the Covid-19 Stress Scales -CSS (Taylor et al., 2020) applied in the study of translation, adaptation and validation in Romania.

Material and Procedure
Covid-19 Stress Scales (CSS) (Taylor et al., 2020)  As there is no generally accepted method for adapting a scale to another language, we used a combination of a back-translation procedure (Iliescu, 2017), i.e. one authorized translator translated from English into Romanian and another did a back-translation into English, in order to highlight the linguistic differences, the final form was decided by a team of 2 psychologists familiar with the subject covered by the text and the 2 authorized translators.

Participants
The sample size was estimated according to the rule proposed by the Consensus-based Standards for the selection of the health Measurement Instruments (Mokkink et al., 2016;Terwee et al., 2007). Thus, the sample size should be of 120 to 300 medical staff.
In this study were included the total sample (n = 160). The age of the respondents ranged from 19 to 59 years (M = 35.80, SD = 22 3.65). The study involved 2 biologist, 131 nurses, 11 medical registrars, 16 doctor (Tabel 1). By study, 66 post high school education, 48 License degree or equivalent, 40 master's degree or equivalent, 6 PhD graduate (Tabel 2). By gender, 135 of participants were women and 25 are men (Tabel 3). The sample consisted of medical personnel directly involved in the fight with Covid-19. All subjects volunteered to take part in the study.

Procedure
The inclusion criteria were as follows: (a) an adult of age 18 years or older; (b) living in Romania; (c) having provided consent to participate; (d) cadrul medical implicat in sectie Covid-19. The online survey was presented in a Google Form and was promoted by distributing the survey link via email. Participation was voluntary and respondents could choose to withdraw from the study at any point. All questions were anonymized to preserve confidentiality. Data collection was conducted from May 9 to June 16, 2021.

Results
The statistical analysis was performed with SPSS version 25 and AMOS version 26. The normality condition of the data was checked both by graphical inspection of the histograms and by calculating skewness and kurtosis indicators, verifying their conformity with the values provided in the literature for normal distribution.
The skewness and kurtosis values indicate that distributions do not deviate substantially from normality for any of the six scales. The internal consistencies alphas ranged from .87 (for the checking scale) to .95 (for the Stres_Covid19_total scale) thus showing very good reliability for all six scales. The correlation matrix of the COVID-19 Stress Scales is presented in Table 3. Correlations among scales ranged from .51 to .80, thus their intensity being similar to those reported by authors (Taylor et al., 2020). Table 4 presents means and standard deviations for all 36 items of the original scale,   The reliability of the instrument was calculated by the Cronbach's alpha coefficient. High values were obtained for the Cronbach's alpha coefficients, between 0.87 and 0.96 (according to table no. 6). These are close to the values obtained for the original instrument (0.83-0.94, Taylor et al., 2020). The correlations of the questionnaire were statistically significant, according to Table 6.

Exploratory factor analysis
To establish the factorial structure of CSS, exploratory factor analysis was used and then confirmatory factor analysis. In the first stage, the necessary conditions for performing the exploratory factor analysis were tested: the sphericity condition, respectively the adequacy condition of the sample. For testing the sphericity condition we used the Bartlett test for sphericity, which was statistically significant (χ2 (630) = 5532, p <0.001) and for testing the suitability of the sample, the Kayser-Meyer-Olkin index was used. (KMO), which had a value of 0.921, excellent for the use of exploratory factor analysis (Sava, 2011

Confirmatory factor analysis
Confirmatory factor analysis (CFA) had been performed to assess the model's fitness for purpose. The CFA as well as the estimation of the structural equation model by AMOS have been tried out. There is an investigation of the measuring model that assumes each item is solely loaded on its expected latent variable (Thompson, 2004). The measuring model, shown in the picture above, contains all of the information pertinent to their specific factor loading procedures. Following the drawing of covariance between the mistakes terms of the few redundancy elements, the analysis determined that the model in the current research study was adequate. As stated by Hair et al., a CMIN/DF ratio of 1.485 is appropriate (2010). A confirmation factor analysis (CFA) was performed to support the measurement model (Bagozzi & Edwards, 1998). Danger, SEC, Xenophobia, Contamination, Traumatic Stress, & Checking were all included in this study's latent variables, which were categorized as follows: Model fit has been evaluated using a combination of different fit indices, including the chi-square model, the IFI, the Tucker Lewis (TLI) index, the CFI, and the root mean approximate square error (RMSAA). When compared to the data from different models, the model had been an excellent fit. (CMIN/DF = 2.291, CFI =0.855, TLI (0.843), IFI (0.857), and RMSEA (0.090) are the values for the coefficients (Hinkin, 1995). In this study, the acceptable validity of a five-factor model was established by the CFAs. Furthermore, all of the items had a substantial effect on their respective latent variables. According to the findings of the CFA, discrimination is justified in this situation. The reason for this is that all elements loaded on their latent building techniques only follow the thresholds that are necessary for them to function. Once all of the units have been uploaded into their idea, the model matches the results fairly well. Because the objects put on some linked concepts are highly interconnected with one another, it lends further evidence to the convergent significance hypothesis. Furthermore, element uploading is important, accounting for more than 0.7 percent of the total (Javed & Liu, 2018

Discussions
The results of the present study indicate adequate psychometric properties of the translated CSS. Thus, a high internal consistency was obtained for the 6 scales of the instrument, with Cronbach's alpha coefficients in the range 0.74-0.86.
Factor analysis showed a 5-factor solution, similar to the original version of the instrument (Taylor et al.,202). Statistically significant correlations were obtained between all measured scales of the instrument.
The present study is the first study of translation and adaptation of the Covid-19 Stress Scales (CSS) on the Romanian population, investigating its psychometric properties. Thus, the present study also contributes to the literature by proposing a version of the scale adapted to a Romanian sample.
But the study has certain limitations. Thus, it is possible that the high gender, age and education disproportion may have influenced to some extent the scores obtained (due to certain socio-cultural considerations, it is possible that women are more prone to stress caused by Covid-19, as compared to men), thus caution is recommended in generalizing the results to certain populations. The questionnaire can be extremely useful in providing information on Covid-19 stress.