Introduction

Since its emergence in December 2019, COVID-19 has created a worldwide health catastrophe. In 2022, the second-generation Omicron variant sharply increased the number of infected persons across the globe. According to statistics from the World Health Organization, more than 511,479,320 total cases were confirmed as of May 2022, resulting in 6,238,832 fatalities.

To fight this pandemic, countries around the world implemented quarantine and isolation policies, which successfully slowed the spread of the disease. In a transnational study, it was found that compared to the Chinese there were significantly fewer Polish respondents who wore face masks (Poles: 35.0%; Chinese: 96.8%; p < .001). Significantly more Polish respondents reported having physical symptoms resembling that of the COVID-19 infection, recent medical consultation, recent COVID-19 testing, and hospitalization (Wang et al., 2020a). The use of face masks by most people in Hong Kong has played an important role in controlling the spread of COVID-19 (Wong et al., 2020). Although air quality and ventilation experts believed wearing face masks could offer psychological benefits (Zhai, 2020), the mixed opinion, contradictory messaging, and shortage of face masks could cause public anxiety and confusion. Concurrently, the conditions of quarantine deteriorate an individual’s behavioral patterns, emotional status, and well-being (Fiorillo & Gorwood, 2020; Galea et al., 2020). Stress experienced due to the pandemic was highlighted as a mediator of the longitudinal effects of COVID-19-related lockdowns on well-being (Achterberg et al., 2021). Partial lockdown and the factors that were associated with an increased level of depression, stress, and anxiety were being single, separated, or widowed, having a higher education level, a larger family size, loss of jobs, and being in contact with individuals (Le et al., 2020), and other study has found the household income loss due to the impact of COVID-19, People holding undergraduate degrees, working in other sectors rather than healthcare, and having definite-term contract had a higher likelihood of income reduction (Tran et al., 2020). Additionally, people’s increasing intolerance for the adoption of health-protective measures for a prolonged time led to “pandemic burnout” (Moroń et al., 2021; Lau et al., 2022). It has been found that burnout caused by a pandemic can make people feel emotionally drained and affect people’s every aspect of their life (WHO, 2022). People who are stuck in pandemic burnout may lead to lower levels of motivation, feelings of helplessness, loneliness, hopelessness, depersonalization, and reduced personal achievement (Haktanir et al., 2022). Lau et al. (2022) examined the psychometric properties of COVID-19 Burnout Views Scale (BVS) with a large Hong Kong sample and found the expected positive associations of COVID-19 BVS with the presence of chronic illnesses (Hayat et al., 2022), support for “living with the COVID” policy, i.e., by reducing the social distancing measures (Ligo et al., 2021).

In this context, studies have also examined the prevalence of psychological distress in various populations during the pandemic. The survey by Wang et al. (2020a) found that more than half of the respondents rated the psychological impact as moderate to severe in China. Other countries have also assessed the psychological status of the public during the epidemic situation, the public has the same problem, making individuals may feel stress, anxiety, uncertainty, fear, and insecurity due to the COVID-19 pandemic (Zakeri et al., 2021; Magson et al., 2021). A systematic review, including 19 followed cross-sectional study design studies conducted in eight different countries, found that relatively high rates of symptoms of anxiety (6.33–50.9%), depression (14.6–48.3%), posttraumatic stress disorder (7–53.8%), psychological distress (34.43–38%), and stress (8.1–81.9%) were reported in the general population during the COVID-19 pandemic in China, Spain, Italy, Iran, the US, Turkey, Nepal, and Denmark (** model. Handbook of the psychology of self-forgiveness (pp. 87–99). Springer." href="/article/10.1007/s12144-023-04500-9#ref-CR61" id="ref-link-section-d1895677e595">2017), higher levels of self-esteem and self-confidence (Pandey et al., 2020; Woodyatt & Wenzel, 2013), greater personal growth and life satisfaction (Levi-Belz & Gilo, 2020; Woodyatt et al., 2017) and greater emotional stability and flexibility (Bem et al., 2021). In addition, there were researchers also showed that higher levels of self-forgiveness are associated with higher levels of compassion, personal growth, and life satisfaction during the COVID-19 pandemic (Beltrán Morillas & Expósito, 2021).

The present study

According to previous studies, clear relationships exist between anger, forgiveness, and subjective well-being. However, the status of these relationships may be impacted by the COVID-19 lockdown situation. Focusing on this context, this study established a moderated mediation model to investigate how trait anger affected subjective well-being in Chinese college students. Specifically, this model was designed to test the mediating role of forgiveness and moderating roles of individual conditions (i.e., lockdown and non-lockdown). Based on previous research, we first proposed the two following hypotheses:

  • H1: Forgiveness will mediate the relationship between trait anger and subjective well-being.

  • H2: The lockdown situation will moderate the mediation model.

Methods

Participants

During the period of this study, due to the second spread of Omicron, people in some cities of China avoided traveling, and could only conduct online learning and work at home or school. WeChat and QQ, two convenient communication applications popular among college students in China, were used to spread online questionnaires and collect data. Project investigators used convenient sampling to invite undergraduates and graduates who had been learning online during the spring semester in May 2022 through their social networks to participate in the questionnaires. This study conducted a questionnaire survey among 1,274 respondents in China, all of whom were asked to report on the following: (1) current health status (non-diseased or symptoms related to COVID-19, including fever, suspected infection, confirmed infection, or cured infection), (2) current living status (not under lockdown and can travel freely or under lockdown at home or hotel due to COVID-19), and (3) identity (general public or frontline worker, such as medical staff). Of all initial respondents, 218 frontline medical workers were excluded, resulting in a final sample consisting of 1,056 individuals from the general public (325 males and 731 females, mean age = 22.45 ± 12.70 years) (Table 1).

Table 1 t-test results for trait anger, forgiveness, and subjective well-being among not locked-down situation and locked-down situation.
Full size table

Throughout the investigation, established ethical guidelines were followed to a high degree. First, the Research Ethics Committee of Shanghai Normal University approved the study design. Second, when participants were sent an invitation to investigate, they were also informed of the content and purpose of the study. Finally, participants are required to agree to voluntarily participate in an anonymous online questionnaire survey before data collection, and during this period, participants can stop answering the questionnaire for any reason at any time.

Materials

Trait Anger Scale (TAS)

This study measured the individual level of trait anger using the trait anger scale (TAS) (Spielberger, 1988), which consists of 10 items across two dimensions, including temperament trait anger and reactive trait anger. All items are rated on a 4-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree), with higher aggregate scores indicating higher trait anger. The TAS has previously shown high internal consistency (α = 0.88), test-retest reliability (r = .83), and validity (Spielberger, 1988). In this study, Cronbach’s α was 0.92.

Heartland Forgiveness Scale (HFS)

This study measured the individual level of forgiveness using the Heartland forgiveness scale (HFS) (Thompson et al., 2005), which consists of 18 items across three dimensions: self-forgiveness, the forgiveness of others, and situational forgiveness. All items are rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree), with higher aggregate scores indicating higher forgiveness. In this study, Cronbach’s a was 0.76.

Subjective Happiness Scale (SHS)

This study measured subjective individual happiness using the subjective Happiness scale (SHS) (Lyubomirsky & Lepper, 1999), which consists of four items. All items are rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree), with a higher mean score indicating greater subjective happiness. The SHS has previously shown good convergent validity (0.62), internal consistency (0.86), and test-retest reliability (0.72) (Lyubomirsky & Lepper, 1999). The scale also has good structural validity for use among Chinese participants (χ2/pdf = 1.55, CFI = 0.99, TLI = 0.99, SRMR = 0.01, RMSEA = 0.02).

Results

Preliminary analyses

The correlations between variables before testing the mediation effect were calculated. As shown in Table 2, trait anger is negatively correlated with forgiveness and subjective happiness, forgiveness is positively associated with subjective happiness.

Table 2 Correlations between all variables
Full size table

Mediation analyses

After testing the correlations, the mediating role of forgiveness in the relationship between trait anger and subjective happiness was tested using PROCESS 3.4 on SPSS (Hayes, 2018).

Gender and age were controlled for in the regressions, and the mediation effect of forgiveness was examined. As shown in Table 3, trait anger negatively correlated with subjective happiness (b = -0.05, p < .001), and negatively correlated with trait forgiveness, (b =-1.01, p < .001). With the addition of forgiveness in the regression model for subjective happiness, the regression coefficient of trait anger was reduced to -0.02 (p < .001). Moreover, forgiveness positively correlated with subjective happiness (b = 0.03, p < .001). Thus, forgiveness mediated the relationship between trait anger and subjective happiness. Specifically, the direct effect was − 0.02, 95% CI= [-0.29, -0.01], and the indirect effect was − 0.03, 95% CI= [-0.03, -0.02]. The indirect effect explained a 58.86% variance of the total effect.

Table 3 Testing the mediation effect of trait anger on subjective well-being
Full size table

Moderated mediating effect test

Controlling for age and gender, the moderated mediation analysis was conducted. As Table 4 shows, the interaction (Model 1) between trait anger and the lockdown situation under the COVID-19 pandemic significantly predicted forgiveness, suggesting that the lockdown situation moderated the effect of trait anger on forgiveness. Simple slope tests suggested that, the people in the lockdown situation (coded 1, Bsimple = -1.07, t =-16.34, p < .001) and the situation of not locked-down (coded 2, Bsimple=-0.73, t=-5.37, p < .001), higher trait anger predicted lower forgiveness (Fig. 1). Nevertheless, the slope for the situation of locked down was larger than the situation of not locked down.

Table 4 Test the moderated mediation effect of trait anger on trait forgiveness
Full size table
Fig. 1
figure 1

The moderating effect of lockdown situation on the relationship between trait anger and forgiveness

Full size image

Table 4 also shows the interaction (Model 2) between trait anger and lockdown situation during the COVID-19 pandemic, which significantly predicted subjective happiness. This result suggests that the lockdown situation moderated the direct effect of trait anger on subjective happiness. Simple slope tests (Fig. 2) showed that, for individuals with the lockdown situation (coded 1, Bsimple = -0.02, t =-3.33, p < .001) and the situation of no locked-down (coded 2, Bsimple=-0.04, t=-3.99, p < .001), higher trait anger predicted lower subjective happiness. (Fig. 2). Nevertheless, the slope for the situation of no locked-down was larger than that lockdown situation.

Fig. 2
figure 2

The moderating effect of lockdown situation on the relationship between trait anger and subjective well-being

Full size image

Discussion

This study examined a model in which forgiveness was positioned as a mediator in the relationship between trait anger and subjective well-being, thus, establishing a moderated mediating model in the context of the COVID-19 lockdown situation. The results showed that anger and forgiveness were each significantly correlated with subjective well-being, with forgiveness playing a mediating role. According to the mediating model, the lockdown situation moderated the relationships between trait anger and both subjective well-being and forgiveness. In sum, these findings suggest that forgiveness mediates the relationship between trait anger and subjective well-being, while trait anger has negative predictive effects on forgiveness and subjective well-being, respectively. Meanwhile, the lockdown situation increases the negative predictive effects of anger on forgiveness and subjective well-being.

Yildiz defined forgiveness as the process of giving up a grudge and healing (2017). In the process of forgiveness, the individual gradually reduces negative emotions toward others and oneself (Menahem & Love, 2013). Both the results of this study and previous results show a negative correlation between forgiveness and anger (Barber et al., 2005; Rohde-Brown & Rudestam, 2011). As forgiveness is difficult under high-trait anger, reducing anger levels and increasing anger control should be considered as variables that can functionally increase the potential for forgiveness. If the anger and underlying primary emotions decrease, then individuals should find it easier to forgive (Sharma & Das, 2017; Malakcıoğlu, 2018). Some studies have also shown that forgiveness increases positive emotions and conditions, including love, happiness, hope, compassion, and mercy (Enright & Fitzgibbons, 2015; Wade et al., 2014). Thus, individuals can enhance their subjective well-being by transforming negative emotions and implementing forgiveness.

During COVID-19, long-term isolation eliminates the opportunity for many individuals to communicate with others, thus reducing the level of social support, and then in such an environment, students are vulnerable to a variety of negative conditions, which is likely to produce fear, social isolation, anxiety, and depression (Cao, 2020; Ma & Miller, 2021). Against this background, this study found that the lockdown situation moderated individual subjective well-being. Compared with individuals who were not under lockdown, those under lockdown with higher individual levels of idiosyncratic anger were increasingly susceptible to lower subjective well-being. Previous studies have shown that the potential to spend time outdoors may be particularly important under conditions of lockdown and social distancing (Burtscher et al., 2020; Samuelsson et al., 2020). These may include opportunities to physically and psychologically escape from the stressors of household confinement, maintain some social relationships, engage in physical activity, and develop a sense of connection with the outside world (Samuelsson et al., 2020). Therefore, the lockdown situation can impact people’s happiness. This study affected subjective well-being by moderating anger and forgiveness.

The results of this study suggest that improving individual anger levels can enhance self-happiness. Studies have reported that a high prevalence of psychological distress was associated with a longer duration of quarantine (Hawryluck et al., 2004). In particular, for those grieving from the traumatic and sudden loss of loved ones due to the COVID-19 outbreak, the inability to gain closure can result in anger and resentment (Shear, 2012). To further meet the demands of the nation in this pandemic, it would be worthwhile to provide online or smartphone-based psychoeducation about the disease outbreak, promote mental wellness and initiate psychological intervention (e.g. cognitive behavior therapy) (Ho et al., 2020). The strongest evidence-based treatment is cognitive behavior therapy (CBT), especially Internet-based CBT which can prevent the spread of the infection during the pandemic. CBT, by enhancing stress management, can also mitigate maladaptive co**s, such as avoidance, antagonistic confrontation, and self-blame. Maladaptive co** styles have been associated with worse psychological outcomes (Sim et al., 2010; Maunder et al., 2006). To date, Internet-based CBT has helped a vast majority of patients bridge geographical distance and control the cost by using some platforms (e.g., Moodle). Moreover, it has enabled the delivery of psychotherapeutic interventions via the Internet (Zhang & Ho, 2017). Systematic reviews and meta-analyses conducted on stress disorder and insomnia have highlighted the efficacy and effectiveness of CBTs which are delivered via the Internet (Soh et al., 2020; Sijbrandij et al., 2016). Therefore, Internet-based CBT is one of the most effective measures to improve the anger levels of individuals during the pandemic.

Limitations

Based on the statistical results, the lockdown situation did not have a significant direct effect on subjective well-being. In terms of sample selection, this may have been due to differences in lockdown times between respondents. With regard to the social environment, this may have been due to the balance between controlling the spread of COVID-19 through lockdown and building confidence in regional public health measures. There is also evidence that feeling socially connected and adhering to COVID-19-related restrictions during lockdown decreased the risk of mental health problems two months into the pandemic (Magson et al., 2021).

In addition, future research should examine the following limitations: first, in the study, the gender and age of participants were controlled, and frontline medical workers and epidemic prevention volunteers were excluded. However, the results may also have been influenced by other factors, such as socioeconomic status, mental health/psychological diagnoses, and marital status which should be paid attention to in the future. Second, the participants in this study were undergraduates and graduates who had been learning online during the spring semester in May 2022, which limits the generalizability of the findings to other populations. Third, the present study used convenience sampling, which may cause sampling bias and reduce the validity. In the future, random sampling should be used for a better representation of the general population. Finally, this study collected cross-sectional data, which cannot be used to establish causal relationships between trait anger, forgiveness, and subjective well-being, and the lack of pre-lockdown data leads us not to be sure that there was no similar situation before the lockdown. Thus, future studies aimed at how trait anger and forgiveness impact subjective well-being can significantly benefit from longitudinal designs.

Future directions

The COVID-19 pandemic has been an unprecedented challenge for healthcare providers. The physical and mental health risks of COVID-19 infection have been well documented, but there is a lack of research on implementing and evaluating targeted solutions to current and future treatment challenges.

According to the strategies that the WHO 2013–2030 Mental Health Action Plan proposed (2021), the upstream and downstream measures could be taken to improve mental well-being, particularly among those at high risk. Early assessment and identifying risk factors for mental well-being are key to supporting referral processes for service delivery (Matsushita & Grubic, 2022), and there is a need for mental health services to continuously and quickly adapt to unpredictable and changing pandemic demands to ensure high quality, yet safe service for patients and carers, providing remote care options through telepsychiatry and e-mental health platforms can further help address pandemic-related service disruptions (Byrne et al., 2021).

Furthermore, as Matsushita and Grubic (2022) suggested, in both systems planning and treatment decisions, it is strongly recommended to engage service users, and students who are empowered to inform the development of pathways to effective mental health support can further benefit from improved mental health literacy and positive self-development.

Conclusion

Focusing on COVID-19 conditions among college students in China, this study established a moderated mediating model with forgiveness positioned as a mediator in the relationship between trait anger and subjective well-being, while the lockdown situation was positioned as a moderator. The results found that trait anger negatively predicted subjective well-being, while forgiveness worked as a mediator between the two variables. In this context, trait anger had negative predictive effects on forgiveness and subjective well-being, respectively. Further, the COVID-19 lockdown situation moderated the influence of trait anger on both forgiveness and subjective well-being. In cases where an individual’s travel was limited, subjective well-being was more susceptible to trait anger.