Edited by: Patrícia Arriaga, University Institute of Lisbon (ISCTE), Portugal
Reviewed by: Valentina Sommovigo, University of Pavia, Italy; Marta Matos, University Institute of Lisbon (ISCTE), Portugal
†These authors share first authorship
‡These authors share second authorship
This article was submitted to Organizational Psychology, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Social media plays a significant role in modern life, but excessive use of it during the COVID-19 pandemic has become a source of concern. Supported by the conservation of resources theory, the current study extends the literature on problematic social media usage during COVID-19 by investigating its association with emotional and mental health outcomes. In a moderated mediation model, this study proposes that problematic social media use by workers during COVID-19 is linked to fear of COVID-19, which is further associated with depression. The current study tested trait mindfulness as an important personal resource that may be associated with reduced fear of COVID-19 despite problematic social media use. The study collected temporally separate data to avoid common method bias. Pakistani employees (
The past decade has seen a rapid increase in social media usage, which has stirred debate on its potential benefits and drawbacks (
Social media infodemic has always been an issue but this has become an even greater challenge during the COVID-19 pandemic, which has made many people fearful (
Social media can help in disseminating information, which might be useful in dealing with the pandemic, but it is also linked to anxiety and depression (
The limited research available on fear of COVID-19 indicates that it is a strong predictor of mental health issues (
Although a wealth of literature on the adverse psychological outcomes of COVID-19 has been generated within a short time (
Proposed hypothesized model.
The current study relies on the conservation of resources (COR) theory (
Building upon the conservation of resources theory (
Researchers have reported an increase in social media use during the COVID-19 pandemic (
Researchers have started to realize that psychological issues related to COVID-19-need immediate attention (
As the global death toll due to COIVD-19 continues to rise, many people fear catching COVID-19 (
Another problem is the sharing of disturbing videos on social media in which nurses say goodbye to their families before leaving to treat COVID-19 patients, patient suffering, and coffins on trucks, which seems to enhance fear of COVID-19 (
Fear of catching COVID-19 may be related to mental health issues (
While medical research on COVID-19 largely focuses on treatments and vaccinations, scholars have also started to raise awareness on maintaining psychological health during this challenging time (
The current study also proposes a moderated mediation model (
The current study is quantitative. Physical contact with respondents was not possible due to the COVID-19 lockdown restrictions; therefore, all respondents were contacted online. Furthermore, only currently working employees were considered for participation in the study. This study followed the CHEERIES checklist for e-surveys and the STROBE checklist for time-lagged studies. The researchers’ institutional review board approved the study. Data for all variables were self-reported, which enhances the risk of common method bias. To minimize common method bias in these self-report measures, the data were collected in three time lags with a minimum gap of 7 days between each lag, in line with the recommendations of
Pakistan reported its first COVID-19 case on February 26, 2020 (
We used a non-probability convenience sampling technique for data collection as the total population of employed individuals in Pakistan is unknown. The authors collected the email addresses of employees working in different organizations through personal contacts and organizations’ official websites. Data were collected from 12 organizations, seven of which were universities, whereas the remaining 5 were in the IT field. Informed consent was obtained from the respondents before participation. The informed consent form explicitly mentioned that participation is voluntary and described the purpose of the study. Although no monetary benefit was given to the participants, they were promised that the survey results would be shared with them upon request. The survey was open enrollment; anyone with the survey URL could participate. We used Google Forms to implement the survey. The authors emailed the informed consent form to potential respondents. The informed consent form contained the study’s purpose and participation criterion along with the URL for Time 1. Email recipients were invited to participate in the study if they met the inclusion criterion. The researchers emailed the URL for Time 2 after a gap of 7 days to all employees who had responded at Time 1. The researchers emailed the URL for Time 3 to employees who had responded at Times 1 and 2. Respondents were able to review their responses before submitting the questionnaire. In accordance with the instructions provided in the CHEERIES checklist for conducting e-surveys, only one response was allowed per email address. This restriction was applied to avoid receiving more than one response from the same respondent. The minimum time required to complete the survey at Time 1 was 10 min, whereas the survey at Time 2 required 4 min, and the survey at Time 3 required a minimum of 6 min.
As the study was time-lagged, the scales were presented in sequence. At Time 1, respondents were asked to provide data on demographics, problematic social media use, and mindfulness. Fear of COVID-19 was measured at Time 2, and employee depression was measured at Time 3. During all three time lags, the respondents were asked whether or not they had contracted COVID-19. The survey comprised three parts: (i) an informed consent form explaining the purpose of the research and ensuring the anonymity of responses; (ii) demographic variables like age, gender, education, and work experience; and (iii) study variables as per the hypothesized model. The inclusion criteria were listed in the informed consent form. Respondents were asked to complete the survey only if they (i) were employed and currently working from home, (ii) had experienced no depressive symptoms or mental health issues before COVID-19, and (iii) had an active social media account.
Twenty-five respondents were in the high-risk group as they were aged 50 and above, but none reported being infected with COVID-19. The confidentiality of the data was fully maintained. The data file was saved in a password-protected folder to which only researchers had access. We assigned a unique I.D. to each respondent and used it to match responses during all three time lags to ensure the anonymity of respondents. We ensured that their responses would be kept confidential and would only be used for the purposes of the study. Three hundred and forty-seven respondents completed the first wave of the study. At Time 2, the 347 employees were contacted again to provide data about their fear of COVID-19, and 312 respondents provided data at Time 2. Finally, these 312 respondents were asked to fill in the questionnaire about depression at Time 3, and 267 responded. These 267 responses were included in the final analysis of the hypothesized model. Thus, the final response rate was 66.7%. The final sample size for all three waves of data collection contained no missing values.
G∗Power (version 3.1.9.4) designed by
The current study is time-lagged, which has the drawback of a lower response rate, as this design requires researchers to approach respondents more than once and most respondents fail to respond at all time lags. Several other time-lagged studies have also shown a very low response rate. For instance, a study in which data was collected in two waves showed a response rate of 49%, only as 162 out of 320 respondents filled in the survey at both time lags (For reference, see
Of the final 267 respondents, 177 were male and 90 were female. 70% of respondents were aged 21–40 years old. 92% had bachelor’s degrees or higher. 63.7% had less than three years of work experience, while the remaining 36.3% had more than 3 years of work experience (see
Respondent characteristics.
Variable | Frequency | Percentage |
Male | 177 | 66 |
Female | 90 | 34 |
21–30 years | 108 | 40.4 |
31–40 years | 80 | 30 |
41–50 years | 54 | 20.2 |
50 and Above | 25 | 9.4 |
Below Bachelor | 20 | 7.5 |
Bachelor | 82 | 30.7 |
Masters and above | 165 | 61.8 |
Less than 1 year | 70 | 26.2 |
1–3 years | 100 | 37.5 |
3–5 years | 43 | 16.1 |
5–7 years | 34 | 12.7 |
7 and above | 20 | 7.5 |
The questionnaires were adapted and distributed in the English language. The vast majority of employees in Pakistan speak English well (
Problematic social media use was measured with a 6-item scale adapted from a previously published source (
Fear of COVID-19 was measured using a 7-item scale recently developed by
The 15-item Mindful Attention and Awareness Scale (MAAS) developed by
Depression was measured using the brief 9-item Patient Health Questionnaire PHQ-9 developed by
Another recent study also used the Patient Health Questionnaire PHQ-9 with a 5-point Likert scale format to measure depression (e.g.,
Prevalence rates of depressive syndromes.
No. of respondents | Percentage | |
Minimal 1–7 | 0 | 0 |
Mild 8–15 | 8 | 3.00 |
Moderate 16–23 | 51 | 19.10 |
Moderately Severe 24–32 | 114 | 42.69 |
Severe 33–45 | 94 | 35.21 |
The Statistical Package for Social Sciences (SPSS) Research Version 21, Process macro by Hayes plugin extension, and AMOS Version 21 were used for data analysis. We confirmed that the data fulfilled all regression assumptions, including linearity, normality, homoscedasticity, multicollinearity, and autocorrelation, before testing the proposed hypotheses. The data was found to be linear, and the error terms were homogenous. Likewise, multicollinearity is not an issue that affects our data because all correlations were well below the cutoff value of 0.70. Additionally, the variance inflation factors (VIF) for all variables were well below the cutoff value of 10 (
The analysis for regression assumptions, means, standard deviations, analysis of variance, correlations, reliability coefficients, and demographic frequency distributions were all conducted using SPSS. Confirmatory factor analysis (CFA) was conducted using AMOS. We used the maximum likelihood method (ML) for estimating parameters in the CFA model. Researchers recommend ML for social sciences research involving Likert scales (
The factor loadings for each variable were checked to confirm the convergent validity of the study variables. The factor loadings were greater than 0.4 for all items, which shows that the items load strongly on their respective latent variable. The average variance extracted (AVE), maximum shared variance (MSV), and composite reliability (CR) were calculated to test discriminant validity. Additionally, a four-factor CFA was conducted to further confirm the discriminant validity. For this purpose, we examined the values of model fit indices, including model chi-square (χ2), degrees of freedom (df), comparative fit index (CFI), incremental fit index (IFI), Tucker Lewis index (TLI), and the root mean square error of approximation (RMSEA).
Model 7 of the Process macro was used to test the moderated mediation model. The number of bootstrapped samples was set to 5,000, and a 95% confidence interval was specified. The current study utilized a bias-corrected method for constructing confidence intervals.
In line with
Confirmatory factor analysis and alternative models
Model | χ2 | df | χ2/df | CFI | TLI | IFI | SRMR | RMSEA |
One factor (Combine all variables into one factor) | 2921 | 629 | 4.64 | 0.51 | 0.48 | 0.51 | 0.15 | 0.12 |
Two factor (Combine “PSMU and EM” and “FOC and DEP”) | 1952 | 628 | 3.11 | 0.71 | 0.70 | 0.72 | 0.11 | 0.09 |
Two factor (Combine “SMU and FOC” and “EM and DEP”) | 2270 | 628 | 3.61 | 0.65 | 0.63 | 0.65 | 0.13 | 0.10 |
Three factor (Combine PSMU and EM) | 1622 | 626 | 2.59 | 0.78 | 0.77 | 0.79 | 0.10 | 0.08 |
Three factor (Combine FOC and DEP) | 1394 | 626 | 2.22 | 0.83 | 0.82 | 0.83 | 0.08 | 0.07 |
Three factor (Combine PSMU and FOC) | 1421 | 626 | 2.71 | 0.83 | 0.82 | 0.83 | 0.06 | 0.07 |
Three factor (Combine PSMU and DEP) | 1434 | 626 | 2.91 | 0.82 | 0.81 | 0.83 | 0.07 | 0.07 |
Three factor (Combine EM and FOC) | 1621 | 626 | 2.59 | 0.78 | 0.77 | 0.79 | 0.11 | 0.08 |
Reliabilities, convergent and discriminant validity, descriptive statistics and intercorrelations.
S. N | Variable | AVE | MSV | CR | 1 | 2 | 3 | 4 | 5 | 6 | 7 | ||
1 | PSMU | 3.53 | 0.80 | 0.54 | 0.09 | 0.81 | |||||||
2 | FOC | 3.46 | 0.76 | 0.51 | 0.21 | 0.84 | 0.38** | ||||||
3 | DEP | 3.23 | 0.76 | 0.52 | 0.20 | 0.86 | 0.41** | 0.45** | |||||
4 | EM | 3.22 | 0.86 | 0.59 | 0.16 | 0.94 | −0.22** | −0.27** | −0.12** | ||||
5 | Gender | – | – | – | – | – | 0.04 | 0.09 | 0.01 | 0.05 | |||
6 | Age | – | – | – | – | – | 0.01 | 0.07 | 0.09 | –0.07 | –0.02 | ||
7 | Experience | – | – | – | – | – | 0.00 | 0.07 | 0.15* | –0.01 | 0.12** | 0.45** | |
8 | Education | – | – | – | – | – | 0.09 | –0.02 | 0.03 | 0.01 | –0.05 | 0.11 | 0.30** |
Conditional process analysis.
Unstandardizedβ | SE | LLCI | ULCI | |
PSMU → FOC | 0.37** | 0.06 | 0.25 | 0.48 |
EM → FOC | −0.15** | 0.05 | –0.25 | –0.10 |
PSMU x EM → FOC | −0.15** | 0.06 | –0.27 | –0.10 |
FOC → DEP | 0.34** | 0.06 | 0.22 | 44 |
PSMU → FOC | 0.26** | 0.05 | 0.15 | 0.36 |
PSMU → FOC → DEP | 0.12* | 0.03 | 0.07 | 0.20 |
EM Low -1 SD (2.35) | 0.17 | 0.04 | 0.10 | 0.25 |
EM Mean (3.22) | 0.12 | 0.03 | 0.07 | 0.20 |
EM High + 1 SD (4.09) | 0.07 | 0.03 | 0.02 | 0.16 |
–0.05 | 02 | –0.11 | –0.02 |
Before testing Hypothesis 3, on problematic social media use during COVID-19 and mindfulness were mean-centered by employing Model 7 of
Employee mindfulness as a moderator in the relationship between problematic social media use and fear of COVID-19.
Summary of hypothesis results.
H. No | Hypothesis | Support |
H1 | Problematic social media use during COVID-19 is positively linked to employee depression. | Supported |
H2 | Fear of COVID-19 mediates the relationship between problematic social media use during COVID-19 and employee depression. | Supported |
H3 | Employee mindfulness moderates the relationship between problematic social media use and fear of COVID-19 such that the relationship will be weaker in the case of higher employee mindfulness and strong in case of lower employee mindfulness. | Supported |
H4 | Employee mindfulness moderates the indirect effect of problematic social media use during COVID-19 on employee depression via fear of COVID-19 such that the indirect effect will be weaker in case of high mindfulness and stronger in case of lower mindfulness. | Supported |
Despite the potential drawbacks of problematic social media use during COVID-19, there is little research on its association with adverse psychological outcomes, particularly mental health issues, among employees (
This study further tested the moderating role of mindfulness on the relationship between problematic social media use and fear of COVID-19, and the conditional indirect effect of problematic social media use on employee depression via fear of COVID-19 when mindfulness is high vs. low. The data supported the moderation and moderated mediation hypotheses. This shows that employees with a higher level of mindfulness experience less fear of COVID-19 despite excessive social media use. Hence, they also report experiencing a lower level of depression than employees with a lower level of mindfulness. A few existing studies have come to similar results (
Despite the potential benefits of social media, problematic use is associated with a chain of negative outcomes that gain momentum over time, ultimately leading to serious outcomes. This study investigated the negative outcomes associated with problematic social media use in the context of a life-threatening pandemic. The study also introduced mindfulness as an important trait to help in dealing with external stressors. This study suggests that employees should refrain from excessive social media use due to its association with negative health outcomes. The study results support the conservation of resources theory in the context of the pandemic by supporting the notion that stressors like excessive and problematic social media use act as a threat to employee resources and are therefore associated with negative outcomes like fear. These negative outcomes then gain momentum and magnitude, and could ultimately take the form of more intense and negative outcomes such as depression. However, personal resources like mindfulness may help employees and protect valuable resources after exposure to such stressors and assist in further resource gains.
The current study adds to the limited body of knowledge on the psychological outcomes of COVID-19. The use of social media has risen since the onset of the COVID-19 epidemic, and this study showed that problematic social media use during COVID-19 is linked to negative emotional and mental health outcomes. Thus, this study responds to the call for research on the antecedents and consequences of fear of COVID-19. It further contributes to ongoing scholarly discussion that the fear associated with COVID-19 is linked to mental health issues by studying excessive social media use during COVID-19. Additionally, this study identifies trait mindfulness as a useful personal resource by demonstrating that it may help employees to control their fear of COVID-19. Another contribution of this study is that it shines a spotlight on factors linked to depression among employees during the current pandemic. This needs to be addressed quickly, as mental health issues might impede progress and can adversely affect the overall operations of an organization.
This study also offers important insights that may have implications for practitioners. Managers could recruit a mental health professional to offer free-of-cost consultations to employees. These consultations may also be provided online, as it is still not safe to meet in person due to the high risk of infection. This could help in resolving employees’ mental health issues. Managers could also increase employees’ awareness about COVID-19, particularly its symptoms and preventive measures so that they are not misled by the misinformation available on social media. One way of doing this is to share research and reports from credible resources on employee email or WhatsApp groups, as accurate information can bust many of the myths linked to COVID-19 and reduce employees’ fear of getting COVID-19. Managers could also conduct regular mindfulness training sessions for their employees, as mindfulness is linked to not only reduced fear of COVID-19 but also lower symptoms of depression and various other mental health issues. Mindfulness training experts could be hired to provide online sessions to employees in their homes.
The findings of the current study could also have implications for policymakers. Policymakers may start campaigns to enhance public awareness of the potential drawbacks of excessive social media use. Government authorities may also wish to create official pages on different social media platforms where people can get accurate information on COVID-19.
The current study should be seen in light of its limitations. It investigated the problematic use of social media as a whole rather than the use of any specific platform. Future studies may collect data on the use of specific platforms and their comparative impact on the mental health of users, as each platform has a different user base. This study did not investigate differences in fear of COVID-19 between employees working from home and the office during the lockdown. Multiple organizations required their employees to continue working from the office, including banks and telecom firms. It would be worth studying the difference in fear of catching infection among employees working from home and working from the office in future studies. As another limitation, the scale used to measure problematic social media use did not include the option “Never,” even though some participants might have never “felt an urge to use social media more and more.” Thus, future studies should use a scale containing the “Never” option.
The data of the current study were collected at three different time lags to address common method bias, but this method also has its drawbacks. First, respondents’ social media use, fear of COVID-19, and depression may vary across different time lags. Future studies may wish to collect data for all the variables at all three time lags to compare variations in social media and depression at different time points. Furthermore, this study only highlighted one mental health outcome linked to problematic social media use and fear of COVID-19, namely depression. Future studies might also test for the association between problematic social media use and other psychological health outcomes, such as anxiety, hypertension, and negative emotions, etc. It would also be fruitful to study the link between mental well-being and problematic social media use during COVID-19. This study identified only one dispositional resource, namely mindfulness, which weakens the negative association between problematic social media use during COVID-19 and fear of COVID-19. Future studies might also introduce other personal and situational resources linked to reduced negativity during COVID-19. For instance, it would be useful to test the moderating role of psychological capital, self-efficacy, and family support. Finally, this study only collected data from currently employed individuals, excluding unemployed individuals. Future studies might also seek to collect data from students and unemployed individuals, especially those who lost their job during the pandemic.
Another limitation of this study is the modification of the rating scale for measuring depression. The original scale for depression ranged from 0 and 4; we modified it to have a 1–5 range to maintain a uniform scale for all of the other variables, which were measured on five-point Likert scales ranging from 1 to 5. The Patient Health Questionnaire (PHQ-9) scale for measuring depression severity makes it possible to calculate the prevalence of minimal to severe depression. The cut-off values for minimal to severe depression for the 5-point Likert scale we used might differ slightly from the original measure. This is a potential limitation that could affect the results for depression severity. Future studies might therefore wish to use the original PHQ-9 scale to measure depression.
COVID-19 is reducing humanity’s economic, physical, social, and now psychological resources. The novel coronavirus continues to infect people worldwide, with the WHO issuing a recent warning that this pandemic is not over and the world will have to face more devastating outcomes. Although social media can help us stay connected with the world in this time of isolation, it is very important to ensure its moderate and controlled use to avoid spreading fear of COVID-19 and prevent further depressive symptoms in people. Building and maintaining psychological and mental health is crucial for preventing adverse outcomes linked to problematic social media use during COVID-19. Mindfulness practices might help in relaxing tension in a difficult environment and may also help in strengthening people’s mental ability to deal with the fear of COVID-19 and other mental health issues.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by the departmental Ethics Approval Committee at The University of Lahore (UOL). The Faculty of Management Sciences, Lahore Business School UOL Research Ethics Board, reviewed “Detrimental Health Outcomes of Social Media Usage during COVID-19 Outbreak: The Moderating Role of Mindfulness” research proposal and considers the procedures, as described by the applicant, to conform to the University’s ethical standards and university guidelines. Moreover, the participation in the survey was voluntary, and study participants were first explained about the details of the project. It was assured to them that their responses will be kept in strict anonymity and will be reported as aggregate results. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.
All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.