A Social-Ecological Model for Pandemic Management: A Study of Beppu City during COVID-19

Abstract

The COVID-19 pandemic has necessitated an examination of the interplay between government policies and initiatives, public perceptions, and social capital in Beppu City. Using a mixed-methods approach, encompassing a systematic literature review and a comprehensive analysis of survey data, this study investigated the relationships among these variables in Beppu City. Data from 223 participants were analyzed using the structural equation modeling technique implemented in SmartPLS. Findings reveal significant correlations between government policies, social capital, and public perception, highlighting the importance of proactive policy design and community engagement. The study proposes a Social-Ecological Model for Pandemic Management (SEMPM), illustrating how government policies shape public perception and social capital, which in turn influence the effectiveness of pandemic response initiatives. The SEMPM provides a roadmap for future crisis management, emphasizing adaptive strategies, effective communication, and cross-sector collaboration to build resilience and enhance community response to crises.

1. Introduction

In the face of global health crises such as pandemics, communities and governments are confronted with a multifaceted challenge that extends beyond public health alone. As illustrated by the recent novel coronavirus pandemic (hereafter referred to as COVID-19), effective management of such crises necessitates a comprehensive understanding of several interacting factors, including government policies and initiatives, the underlying social capital within the community, as well as people’s perception of the risk of the disease.

Amidst significant uncertainty, such as COVID-19, governments are implementing a range of policies, from preventive measures like mass testing and vaccination drives to crisis management strategies such as lockdowns and financial support. The effectiveness of the policies significantly influences public perception of risk and adherence to those policies, ultimately impacting the success of their implementation [1]. Risk perception refers to individuals’ beliefs about the likelihood and severity of contracting a disease like COVID-19. This perception is shaped by various factors such as personal experiences, media portrayals, and trust in authorities [1,2]. When the public trusts the government and perceives its policies as effective and fair, compliance is more likely, resulting in better outcomes. Therefore, the government’s approach to policymaking and communication is important for shaping public perception of the virus, ensuring successful implementation [1,2].

Social capital, defined as the collective value of social networks and the associated norms of reciprocity and trustworthiness [3] - significantly influences adherence to health measures and shapes risk perception [4,5]. It facilitates quick and accurate information sharing, promotes compliance with health guidelines, and impacts risk perception. Communities rich in social capital can more effectively mobilize collective action against disease spread, exhibit greater resilience during crises, and are more likely to trust and follow government guidelines [4,5]. Moreover, Government policies that support social capital can help alleviate the adverse consequences of pandemics [6].

People’s perceptions of the pandemic can shape their attitudes, beliefs, and behaviors related to the virus and adherence to its mitigation efforts. Personal experiences, information from the media and government, and social and cultural norms shape those perceptions [7,8,9].

While existing research sheds light on the critical roles of government policies, social capital, and public risk perception (hereafter, pandemic management determinants) in pandemic response [1,2,4,5,6, 7,8,9], a gap remains in understanding how these factors interact and influence each other within a specific community context. Our study aims to address this gap by investigating the interplay between these factors (see Figure 1) to shape Beppu City’s response and the overall impact of COVID-19. To understand these relationships, we developed the Social-Ecological Model for Pandemic Management (SEMPM), which illustrates these factors and their interactions. The proposed SEMPM model is informed by the results of a structural equation modeling analysis that would be conducted to validate the relationships between the constructs within the framework.

Figure 1: Global Health Crisis management (authors’ own work)

1.1 Research questions:

The research questions this study seeks to answer are:

1. 1 How do government policies and initiatives, public perception of the virus, and social capital interact to influence the impact of COVID-19 in Beppu City?
2. 2 Which specific government measures were effective in managing public perceptions and reducing the pandemic’s impact on Beppu residents?
3. 3 To what extent were those government measures effective?

To answer these questions, the study presents a series of hypotheses exploring the relationships between the pandemic management determinants through the lens of the Social-Ecological Model. These hypotheses will specifically investigate how government policies influence public risk perception, how social capital moderates adherence to these policies, and ultimately, how all three factors collectively impact the virus spread and the city’s well-being.

The study integrates the Social-Ecological framework into a comprehensive model that illustrates the deep relationships among government policies, risk perception, and social capital, elucidating their combined influence on shaping Beppu City’s response to and overall impact of COVID-19.

1.2 Theoretical framework: Social Ecological Model

The social-ecological model was initially proposed by Urie Bronfenbrenner as an ecological systems theory [10] and was later adapted by McLeroy to promote health-related behavioral change [11]. This framework emphasizes the interconnectedness of factors at various levels, influencing individual behaviors, including (1) Individual (e.g., fear of COVID-19), (2) Interpersonal (e.g., information sharing), (3) Organizational (e.g., enforcement of preventive measures), (4) Community (e.g., community participation), and Policy (e.g., government regulations) (see figure 2).

Figure (2) Social-ecological model for COVID-19 management adopted from McLeroy et al. (1988) [11]

Social-Ecological Model provides a valuable framework for understanding the multifaceted influences on health behaviors, considering the interplay between individual, interpersonal, organizational, community, and public policy factors. During pandemics, the Social-Ecological Model can be particularly useful in analyzing how these different levels of influence impact health behaviors and outcomes, helping in understanding the role of social determinants and the surrounding environment in shaping responses to health crises, thereby informing strategies for enhancing community resilience and sustainability.

In the COVID-19 context, existing studies have primarily focused on specific aspects such as vaccine uptake and mask usage. For example, studies such as those by Latkin et al. (2021) [12], Naidoo et al. (2023) [13], Lun et al. (2022) [14], and Twersky et al. (2023) [15] have demonstrated the applicability of the Social-Ecological Model in examining vaccine intentions, uptake rates among healthcare workers, and barriers to vaccination acceptance. Limited studies adopted a broader perspective by examining a wider range of preventive factors and behaviors beyond vaccine and mask usage. For example, Jang (2022) [16] and Vilme et al. (2022) [17] explored various preventive measures such as wearing masks, hand handwashing, social distancing, and adherence to public health guidelines utilizing the Social-Ecological Model to understand the multilevel factors associated with these behaviors, highlighting the significance of adopting a comprehensive approach to understanding and promoting preventive behaviors during COVID-19.

1.3 Significance of Research

Building on existing research, this study uses the Social-Ecological Model to examine the interplay of individual, social, and policy factors in Beppu City’s response to COVID-19. It explores how social capital moderates the impact of government policies and public risk perception, hence influencing specific actions and future pandemic responses. The study provides insights into policy design by examining how government actions influence public perception and adherence, aiming to guide the development of more effective measures and policies in future crises. It highlights the significance of the Social-Ecological Model in identifying the connections between societal and environmental factors during pandemics and how social capital, fostered by strong social networks, can mitigate pandemic impacts and lead to a more resilient community response.

2 Methodology

This study employed a mixed-methods approach, utilizing both qualitative and quantitative methods to investigate the factors —pandemic management determinants—that influenced Beppu City’s response to COVID-19. A mixed-methods approach is particularly useful when investigating complex phenomena [18], allowing for a detailed exploration of pandemic management. The qualitative methods, including a systematic literature review to identify key themes and develop hypotheses, provided a broad understanding of pandemic management determinants and their roles in shaping responses. Additionally, the quantitative survey tested these hypotheses, offering empirical evidence on how these factors influenced the pandemic response.

Beppu City, known for its relatively diverse population and its status as a major tourist destination [19, 20], was selected as a case study due to its unique context. This openness to international influences and a tourism-driven economy offers a valuable opportunity to explore the interplay between pandemic management strategies and those unique characteristics.

2.1 Qualitative Data Collection and Analysis (Literature Review)

A systematic search using academic databases and search engines with relevant keywords identified journal articles exploring the pandemic management determinants and their role in shaping pandemic responses. Thematic analysis of the reviewed literature identified recurring patterns and themes (see Appendix A). These themes informed the development of 18 hypotheses about the relationships between the pandemic management determinants.

2.2 Quantitative Data Collection and Analysis (Survey)

Following the literature review and hypothesis development, a quantitative survey was administered to test the hypotheses. The survey was developed based on the hypotheses derived from the literature review, using adapted scales to measure government policy evaluations, pandemic perception, and social capital. A pilot study was conducted to ensure the survey’s validity and reliability, leading to minor revisions for clarity. To maximize participation and capture diverse perspectives, the survey was distributed using a mixed mode, both online and paper-based, to accommodate participants’ varied preferences and accessibility in Beppu City.

Due to limitations in resources and time, a convenience sampling technique was employed to select participants for the survey [21]. The inclusion criteria for participants were individuals aged 18 years or older and residing in Beppu City during COVID-19. Our goal was to recruit at least 384 participants to represent the city’s population of approximately 113,387 citizens as of September 2022 [19]. This sample size was determined based on a common rule of thumb for achieving a margin of error of 5% with a 95% confidence level, assuming a simple random sample and a response distribution of 50% [22]. However, due to limited access, our final sample included 225 participants.

The main analysis involved conducting the structural equation modeling using SmartPLS to test the proposed relationships between the constructs in our Social-Ecological Model for Pandemic Management (SEMPM) framework. This analysis allowed for the assessment of the hypothesized relationships among the variables and the evaluation of the overall fit of the model [23].

Descriptive statistics were calculated to summarize the demographic characteristics of the sample and provide an overview of the key variables. The model was estimated using maximum likelihood estimation, and goodness-of-fit indices were examined to assess the model’s fit to the data. Additionally, path coefficients and significance levels were examined to determine the support for each hypothesis.

3 Systematic Literature review and Hypothesis development

In this section, the authors will systematically review the literature and derive hypotheses following the steps outlined in Figure (3).

Figure (3) Systematic Literature Review and hypotheses development framework

3.1 Comprehensive Review of Literature

Global health crises and pandemics have become increasingly important topics in recent years, particularly in light of the ongoing COVID-19. Addressing these crises requires a multifactorial approach that takes into account a wide range of factors, including government policies, risk perception, social capital, and the interaction of these factors. This section will examine the interplay between these factors during health crises, with an emphasis on the lessons drawn from COVID-19.

3.1.1 Government Policies’ impact on Public Perception of risk:

Government policies have played a significant role in shaping public perceptions of risk during COVID-19 [24], including preventative actions like travel restrictions and lockdowns, as well as crisis management tactics such as resource allocation [1]. The intensity of these measures can significantly alter public perception. For instance, strict lockdowns may amplify the perceived threat of the virus, prompting a more cautious public approach [1,24].

3.1.2 Policy Communication Influence on Risk Perception:

Transparent communication of government policies is essential for building public trust during health crises, increasing awareness about preventative measures, and serving as a template for future crisis management [25,26,27]. Timely, clear, and accurate data can prevent misinformation, leading to well-informed populations that are better equipped to counteract the impact of viruses [26,27], and foster a positive public opinion, subsequently minimizing the pandemic’s adverse impacts [26]. Consistent delivery of promises and guidelines by authorities enhances public trust and cooperation, potentially increasing adherence to preventive measures [25,26,27]. Collaboration with reputable institutions and media outlets is essential for extending the reach of accurate information [27]. Even amidst uncertainties, transparency doesn’t undermine public trust [26,27]. The public’s perception of governmental efficacy significantly influences their trust levels and risk perception. Trust-building policies, such as providing essential resources during pandemics, enhance trust and social [25,26,27]

3.1.3 Government initiatives impact on Public Perception of risk:

Financial considerations also play a role in shaping perceptions. Specifically, when there’s a lack of trust in the government’s ability to manage the pandemic’s financial challenges, feelings of financial insecurity become prevalent [28]. Moreover, direct interventions, such as the transparent and equitable distribution of essential supplies like food, not only build public trust but also serve as essential factors in mitigating longer-term recovery efforts and promoting positive perceptions of pandemic response [26]. The consistent provision of such material aid further enhances adherence to pandemic measures like lockdowns, reducing the need for individuals to venture out and acting as a safety net during challenging times [26].

Government-provided financial and material support during COVID-19 plays important roles in shaping both current responses and prospective strategies. Perceptions of effective financial assistance, targeting the economic hardships arising from lockdowns, can set a precedent for similar approaches in future crises, emphasizing the value of addressing the immediate economic needs of individuals and businesses [29]. Similarly, material aid, encompassing essentials like food, that effectively caters to the basic needs of the populace during lockdowns, may serve as a blueprint for subsequent crisis management [29].

Financial relief initiatives, especially those tailored to the specific requirements of diverse demographic groups, are instrumental in fostering a favorable public opinion of crisis management. This sentiment resonates more profoundly when direct support is channeled to the most vulnerable, reinforcing trust and confidence in the government’s pandemic response [26]. Concurrently, the timely provision of essential goods is an undeniable influencer in enhancing positive public perceptions during such crises [26].

3.1.4 Other factors influencing public perception of risk:

Public perception is strongly influenced by the number of confirmed cases, with an increase in cases correlating to a heightened perception of risk [30]. Additionally, individuals’ direct experience with the virus, whether personally or through close acquaintances, tends to amplify this perceived risk even further [31]. This perception is shaped not only by the likelihood of contracting the virus but also by its wider ramifications, such as disruptions to daily life [24,32,33]. Those who perceive the virus as a greater threat are more likely to take protective measures seriously.

Trust in government, science, and medical professionals, as well as personal and collective efficiency, significantly influences how the public perceives the risk associated with COVID-19. A high level of trust can lead to a heightened perception of risk, particularly when there is clear communication about the number of confirmed cases [31].

3.1.5 Social Capital Influence on Risk Perception and Government Policies:

Social capital, encompassing networks, norms, and trust that foster community cooperation [3, 34], significantly impacts how people perceive and respond to the collective health crisis [4,5,19, 35,36]. It influences adherence to health measures, risk perception, collective action, and resilience during crises [4,5,35,36]. Communities with higher social capital can better mobilize collective action and coordinate responses to collective crises like pandemics [19, 35]. These communities have shown more resilience during the pandemic, with lower COVID-19 cases and death rates [6,19, 37] and reduced anxiety and depression levels [38]. High social capital can heighten risk perception and trust in authorities, affecting the acceptance of new policies during a pandemic [36]. Communities with high social capital often perceive risks more acutely and trust authorities more, leading to better acceptance and implementation of pandemic guidelines [6]. Therefore, it plays a significant role in pandemic response by influencing individual and collective behaviors.

Furthermore, Government policies supporting social capital can help mitigate the negative effects of the pandemic [6]. However, these policies should consider the existing levels and types of social capital in different communities for effective interventions. For example, strict lockdowns can cause economic hardship, reduce access to essential goods, like food and medicine, and negatively impact livelihoods, especially in developing countries with limited social safety nets. This can lead to public non-adherence to government policies [26]. The social and psychological consequences of economic hardship and restricted access to essential goods can erode trust in authorities and undermine compliance with containment measures. Feelings of desperation and a perceived lack of alternative options can lead individuals to prioritize immediate needs over adherence to policies, particularly when social safety nets are inadequate [39].

3.2 Organizing and synthesizing findings

In this section, the authors systematically organize and synthesize the findings from the systematic literature review to develop a comprehensive understanding of the factors influencing the response to COVID-19 in Beppu City.

Identifying Key Themes:

The review identified three key themes that interconnect to influence public response during a pandemic (see Appendix A):

• ○ Government Policies and Initiatives: The type, intensity, and communication of government policies impact public perception of risk and adherence to guidelines. Effective policies and initiatives in a pandemic context are those implemented promptly and communicated clearly and consistently, leveraging existing social capital networks, promoting trust and cooperation within communities, and addressing public concerns. Conversely, poorly conceived or communicated policies can erode trust, hinder adherence to guidelines, and aggravate social tensions.
• ○ Public Perception of Risk: People’s perception of the virus is shaped by various factors, including the number of confirmed cases, personal experiences, trust in authorities, and media portrayals. A heightened risk perception is likely to encourage adherence to preventive measures.
• ○ Social Capital: The strength of social networks and the level of trust and cooperation within a community significantly influence how people respond to a crisis. Communities with high social capital tend to exhibit greater collective action, resilience, and adherence to health guidelines during pandemics.

3.3 Hypothesis development

3.3.1 Hypotheses addressing Relationships Between Base Factors (key themes):

Based on the identified themes, the literature suggests several relationships between the key factors (base factors) influencing public response (See Figure 4).

Figure 4: Theoretical model for hypotheses highlighting the Base factors.

3.3.2 The interplay Between Base Factors and Anti-Spread Initiatives:

By building on the themes and relationships between the base factors, the next step involves translating these general findings into specific hypotheses to explore the interplay between the specific anti-spread initiatives and policies and base factors (See Figure 5).

Figure 5: Theoretical model of hypotheses highlighting the relationship between specific factors and base factors

Table 1: Summary of hypotheses addressing the relationship between base factors

The interplay Between Base Factors
■ Perception of Risk and COVID-19 Impact: A higher number of cases and a more severe outbreak likely led to a heightened perception of risk. The level of COVID-19 impact positively affects People’s perception of the virus (H1)
■ Perception of Risk and Social Capital: Communities with strong social capital are more likely to perceive the virus as a serious threat. Higher social capital promotes community unity and effective collective action during pandemics. High levels of social capital positively influence people’s perception of COVID-19 (H2)
■ Perception of Risk and Effective Policies: Well-executed and communicated policies foster trust and generate a favorable public opinion towards COVID-19 management. Effective policy initiatives and their impact positively affect people’s perception of COVID-19 (H3)
■ Social Capital and COVID-19 Impact: Communities with high trust and cooperation might be better at adhering to guidelines, hence decreasing the virus’s adverse effects. Strong social capital tends to reduce the negative impact of COVID-19 (H4)
■ Effective Policies and Social Capital: Policies perceived as beneficial can strengthen trust and social cohesion within communities. Effective policy initiatives positively influence social capital (H5)
■ Effective Policies and COVID-19 Impact: Strategic and successful policy implementation can lead to a reduction in the negative effects of the pandemic. Effective policy initiatives and their impact tend to reduce the negative repercussions of COVID-19 (H6)

Anti-spread initiatives in this context refer to the policies and interventions implemented by governments aiming at curbing the effect of the virus. They include lockdown, clear communication of information, as well as financial and material resource allocation. These initiatives play a significant role in reducing transmission rates, protecting public health, and ultimately mitigating the overall impact of the pandemic.

Table 2: Summary of hypotheses addressing the relationship between base factors and Anti-Spread Initiatives

Public perciption and Anti-Spread Initiatives:

■ Public Perception and Financial Aid: Financial relief initiatives can lead to more favorable public opinions on crisis management. Providing financial aid positively affects people’s perception of COVID-19 management (H7) ■ Public Perception and Material Aid: The availability of essential supplies can fosters a positive view on how the crisis is being managed. The provision of material aid positively impacts people’s perception of COVID-19 responses (H8) ■ Public Perception and Clarity of Information: Transparent communication about the pandemic empowers the public and fosters trust in authorities. Transparent dissemination of COVID-19 information positively impacts people’s perception (H9) ■ Public Perception and Lockdown: While restrictive, when People perceives lockdowns as necessary for public health, the measures are more positively accepted. Implementing lockdown measures positively impacts people’s perception of COVID-19 management (H10)

Social Capital and Anti-Spread Initiatives:

■ Social Capital and Material Aid: Supplying necessary resources during crisis times can foster mutual trust and cooperation within communities, potentially strengthening social capital Providing material aid fosters social capital (H11) ■ Social Capital and Clarity of Information: Open and timely communication from authorities can contribute to increased trust and unity within a community, enhancing social capital. Clear COVID-19 information dissemination positively influences social capital (H12) ■ Social Capital and Financial Aid: Financial support leads to increased trust and fosters a stronger sense of community, potentially strengthening social capital. Financial aid initiatives enhance social capital (H13) ■ Social Capital and Lockdowns: Effective communication about the necessity of lockdowns can foster a sense of community responsibility and collective action, potentially strengthening social capital. The implementation of lockdowns positively influences social capital (H14)

Government policies and inititatives and COVID-19:

■ Financial Aid and COVID-19 Impact: The severity of the pandemic’s impact can strain resources, potentially hindering the consistency of financial aid initiatives. The severity of the COVID-19 impact negatively affects the provision of financial aid (H15) ■ Lockdowns and COVID-19 Impact: Severe outbreaks often necessitate stricter containment strategies, leading to the implementation of stricter lockdowns. A significant COVID-19 impact tends to necessitate stricter lockdowns (H16) ■ Material Aid and COVID-19 Impact: Providing material aid facilitates better adherence to safety measures, thereby mitigating the virus’s adverse impact. Material aid provision tends to counteract the negative impacts of COVID-19 (H17) ■ Clarity of Information and COVID-19 Impact: effective dissemination of accurate information can equip people to make informed decisions, ultimately mitigating the virus’s negative effects. Effective dissemination of accurate COVID-19 information attenuates the virus’s negative impacts (H18)

4 Results and discussion

This section Investigates the analysis of the survey data to explore the interplay between government policies and initiatives, public risk perception, and social capital. By examining the interplay between these factors, this section aims to provide a comprehensive understanding of Beppu City’s response to COVID-19 and offer insights for future crisis management.

4.1 Descriptive Statistics of the Respondents

Figure (6) illustrates respondents’ socioeconomic characteristics. Half of the respondents were aged between 18 and 24 years. Approximately 14% were aged between 25 and 44 years, while 28% were aged between 45 years and above. Notably, approximately 10% of the respondents were older than 64 years. The distribution of nationality of the respondents suggests that 47% of the respondents were foreign-national, while 53% were of Japanese nationality.

Figure 6. socio-demographics of the respondents

Figure (6) also demonstrates that 37% of the respondents are not currently employed. This fraction included retired senior citizens and dependents. 39% were part-time workers, while 24% had full-time work.

In terms of the residency status of the respondents, Figure (6) shows that our sample covered a wide range of residents of Beppu City: 21% permanent residents (PR), 23% Japanese students, 35% foreign students, 10% Japanese employees, and 12% foreign employees. Responses from this robust distribution of respondents from a wide range of age, residency, and employment backgrounds provide rich data that will allow for robust analysis and testing of the proposed hypotheses.

4.2 Evaluation of the measurement model: Confirmatory factor analysis (CFA)

This section evaluates measurement items by testing the validity and reliability of the constructs. These tests included examining the internal consistency and discriminant validity of the measurement items. Table 3 presents the basic consistency, reliability, and discriminant of the result¹. First, according to the literature, some common test statistics and their bands include: Cronbach’s alpha with a minimum value of 0.70 [40], item loadings with a minimum value of 0.50 [40], a minimum composite reliability (CR) of 0.8 [41], minimum average variance extracted (AVE) value of 0.50 [41] and lastly, the square root of the AVE of each construct must be greater than its diagonal correlation coefficients [42,43].

Table 3. Reliability and discriminant validity of measurement

Panel A. Reliability and validity of constructs
Constructs	Items		Factor loading		Cronbach’s alpha	CR		AVE
PER	Perception1		0.905***		0.744	0.804		0.678
	Perception2		0.929***
	Perception3		0.594***
SC	SC1		0.633***		0.815	0.809		0.651
	SC2		0.824***
	SC3		0.876***
	SC4		0.869***
C19Imp	Impact1		0.882***		0.876	0.884		0.801
	Impact2		0.895***
	Impact3		0.908***
GPI	Gpolicy1		0.920***		0.901	0.903		0.835
	Gpolicy2		0.920***
	Gpolicy3		0.900***
FA	Finaid1 <- FA		0.899***		0.949	0.952		0.867
	Finaid2 <- FA		0.904***
	Finaid3 <- FA		0.955***
	Finaid4 <- FA		0.964***
LD	Glockd1 <- LD		0.909***		0.811	0.815		0.841
	Glockd2 <- LD		0.917***
	Glockd3 <- LD		0.925***
C19Inf	Inform_covid1 <- C19Inf		0.827***		0.764	0.795		0.677
	Inform_covid2 <- C19Inf		0.770***
	Inform_covid3 <- C19Inf		0.869***
MA	Mataid1 <- MA		0.845***		0.899	0.923		0.769
	Mataid2 <- MA		0.927***
	Mataid3 <- MA		0.947***
	Mataid4 <- MA		0.779***
Panel B. Discriminant validity
Fornell Larker Criterion
	C19Imp	GPI	PER	SC	C19Inf	FA	LD	MA
C19Imp	0.895
GPI	-0.546	0.914
PER	-0.364	0.339	0.823
SC	-0.406	0.415	0.584	0.804
C19Inf	-0.752		0.332	0.401	0.823
FA	-0.383		0.301	0.339	0.394	0.931
LD	-0.487		0.303	0.381	0.391	0.318	0.917
MA	-0.351		0.119	0.205	0.354	0.396	0.259	0.877

Examining the estimated test statistics in Table 3 against the theoretical thresholds of the validity statistics demonstrates that all the requirements are met. For instance, all factor loadings are above the threshold of 0.50, and are statistically significant, which suggests that the measurement items reliably measure the constructs. Cronbach’s alpha for all constructs was above the threshold of 0.80. In addition, the validity of the convergent model was adequately established because the AVE for all constructs exceeded the minimum threshold of 0.50. Additionally, the composite reliability scores for all the constructs were approximately equal to or above the threshold of 0.80. Finally, discriminant validity (see panel B) shows that the diagonal elements (square root of AVE) are generally greater than the corresponding correlation coefficients between the constructs, indicating strong convergent validity.

These results provide strong evidence supporting the validity and reliability of the structural model. It also suggests that the measurement items in the structural model satisfy the theoretical condition for which further analysis of the relationships (hypotheses) between the constructs can be reliably tested.

4.3 The structural models

The estimated structural models showing the hypotheses previously presented in Figures 4 and 5 are shown in Figures (7) and (8) including path coefficients, respectively. Figure (7) demonstrates the empirical relationship between the four main constructs: perception (PER), social capital (SC), government policy/initiatives (GPI), and COVID-19 impacts (C19Imp), and Figure (8) illustrates the relationship between specific measures against the spread of COVID-19 and three base factors (perception, social capital, and COVID-19 impacts) in Beppu City. Table (4) provides detailed statistics on the path coefficients of the structural relationships and hypotheses.

Figure 7. Relationship between base factors

Figure 8. Relationship between specific measures and initiative and the main constructs

Table 4. Path coefficient

Panel A. main effect (Base factors)
	Coefficient	STDev	T statistics	P values
(1). Path Coefficient
C19Imp -> PER	-0.125	0.065	1.926	0.054
GPI -> C19Imp	-0.456	0.068	6.738	0.000
GPI -> PER	0.060	0.076	0.793	0.428
GPI -> SC	0.415	0.061	6.820	0.000
SC -> C19Imp	-0.217	0.068	3.176	0.002
SC -> PER	0.508	0.061	8.339	0.000
(2). Total effect
C19Imp -> PER	-0.125	0.065	1.926	0.054
GPI -> C19Imp	-0.546	0.054	10.120	0.000
GPI -> PER	0.339	0.072	4.713	0.000
GPI -> SC	0.415	0.061	6.820	0.000
SC -> C19Imp	-0.217	0.068	3.176	0.002
SC -> PER	0.535	0.060	8.882	0.000
(3). Specific indirect effect
GPI -> SC -> C19Imp -> PER	0.011	0.008	1.480	0.139
GPI -> SC -> PER	0.211	0.035	6.016	0.000
GPI -> SC -> C19Imp	-0.090	0.032	2.826	0.005
SC -> C19Imp -> PER	0.027	0.017	1.560	0.119
GPI -> C19Imp -> PER	0.057	0.032	1.794	0.073
Panel B. relationship between specific anti-spread measures
	Coefficient	STDEV	T statistics	P values
C19Inf -> C19Imp	-0.633	0.049	12.990	0.000
C19Inf -> PER	0.214	0.078	2.758	0.006
C19Inf -> SC	0.246	0.075	3.296	0.001
FA -> C19Imp	-0.045	0.047	0.958	0.338
FA -> PER	0.191	0.080	2.375	0.018
FA -> SC	0.172	0.075	2.305	0.021
LD -> C19Imp	-0.211	0.049	4.319	0.000
LD -> PER	0.179	0.077	2.319	0.020
LD -> SC	0.233	0.072	3.234	0.001
MA -> C19Imp	-0.055	0.048	1.150	0.250
MA -> PER	-0.079	0.074	1.067	0.286
MA -> SC	-0.011	0.062	0.172	0.863

Figure (7) shows that the Beppu residents’ perceptions of COVID-19 were positively and significantly correlated with social capital. Since the social capital construct evaluates residents’ affinity to belong to social groups (study group, ethnic, or religious group) and being in contact with and receiving constant support from such groups as well as from other organized groups, this finding demonstrates that individuals with strong community and social ties have a proper understanding or perception of COVID-19. On the other hand, perception appears to be negatively correlated with the COVID-19 impact, although the relationship seems to be weak and not significant at the 5% level. In addition, the relationship between perception and government COVID policies or initiatives was not significant.

The relationships between social capital and government policy, as well as social capital and COVID-19 impacts, are statistically positively significant and statistically negatively significant, respectively. Social capital appears to be an important channel for the smooth implementation and coordination of government policies and initiatives. The latter result demonstrates that social capital significantly cushioned the effect of COVID-19 impact on Beppu residents. This result is consistent with the literature on the role of social capital in reducing the adverse effects of disasters.

The relationship between the government’s COVID-19 policy and the COVID-19 impact is negative and statistically significant. This is intuitive and implies that respondents firmly affirmed that municipal government policies and initiatives did, in fact, reduce the impact of the pandemic on the general population of Beppu. This finding is crucial because government and state actors remain the single most important agencies in disaster management owing to their financial power and resources. Therefore, successful government policies and measures in disaster management, such as COVID-19, provide important lessons and directions for similar future disasters.

Furthermore, Table 4 Panel A (total and specific effect rows) provides more insight into both the direct and indirect relationships between perception, social capital, government policies, and COVID-19 impact. Interestingly, perception was positively and significantly correlated with government policies, which suggests that respondents who had a positive view of the effectiveness of government policies also appeared to have a proper view or understanding of COVID-19.

Most importantly, two additional inferences can be drawn from “specific indirect effect” estimates in Table 4. First, the indirect relationship between government policies, social capital, and perception (GPI -> SC -> PER) remains positive and statistically significant, which implies that social capital provides a vital and strong channel for government policies to reach deep into the community in times of disaster. Second, the indirect relationship between government policy, social capital, and COVID-19 (GPI -> SC -> C19Imp) impact remains negative and statistically significant, suggesting that government policies implemented through existing social capital structures decrease the impact of the pandemic in Beppu City. The latter findings are important because they show that government policies implemented within a community with strong social ties can successfully reduce the impact of disasters on citizens.

Figure 8 illustrates the relationship between specific measures to prevent the spread of COVID-19. The four main measures investigated are financial aid (FA), material aid (MA), lockdown (LD), and COVID-19 information sharing (C19Inf). Several inferences can be drawn from the results shown in Figure (8). First, perception was positively and significantly correlated with financial aid, lockdown, and COVID-19 information dissemination. This result implies that the perception of COVID-19 is influenced by financial aid, lockdown, and COVID-19 information sharing. Specifically, the results demonstrate that these three measures predict how respondents correctly perceive or understand the dangers posed by the pandemic. On the other hand, perception is not significantly correlated with material aid; that is, material aid does not predict respondents’ perception of the pandemic. None of the main variables is significantly correlated with material aid. One reason for this might be that material aid was not a large-scale initiative during the pandemic in Beppu.

Figure (8) further shows that social capital is positively and significantly correlated with financial aid and lockdown but has a significantly negative relationship with COVID-19 information sharing. The former result demonstrates that respondents with strong social ties favorably appraised financial aid and lockdown initiatives, while the latter result is puzzling. A plausible explanation for the negative relationship between social capital and COVID-19 information sharing is that information on COVID-19 comes from health authorities and the municipal government, and not from or within social networks, although these pieces of information can be circulated within these networks. Furthermore, owing to the very novel nature of the virus, information on prevention and the transmission mode of the virus changes from time to time as experts grapple with the science of the disease.

Finally, the relationships between COVID-19 impact and financial aid, COVID-19 impact, and information sharing are negative and statistically significant. This means that financial aid and frequent information-sharing decreased the impact of COVID-19 on the residents of Beppu City. This result highlights two important points. (1) Financial aid to Beppu citizens was effective in reducing the negative economic impact of COVID-19, especially on the household and personal income of Beppu residents. (2) Frequent updates and information sharing on the prevention and transmission modes of the virus were also effective in reducing the impact of the pandemic through citizen participation and cooperation with government anti-spread measures.

5 Developing the Social Ecological Model

The Social-Ecological Model for Pandemic Management (SEMPM) presented in this section captures the complex interplay between pandemic management determinants (see Figure 9). This model was developed based on the “social-ecological model for COVID-19 management” initially proposed in Figure 2, which outlined factors at various levels influencing individual behaviors in response to COVID-19. The structure of the SEMPM model, with its interconnected levels, builds upon the initial hypotheses derived from the literature review and subsequently tested through the structural equation modeling analysis (Figures 7 & 8), which helped in identifying these key relationships. The SEMPM consists of three interconnected levels: individual/interpersonal (public perception), community/organizational (social capital), and policy (government initiatives). The SEMPM² explores the dynamic relationships between these levels and how they influence each other.

Figure 9: Social-Ecological Model for Pandemic Management (SEMPM)

Government policies and initiatives influence public risk perception and social capital through their design, implementation, and communication. Conversely, public risk perception and social capital reciprocally impact the effectiveness and acceptance of government policies, while also influencing the adoption and outcomes of specific anti-spread initiatives.

SEMPM provides a roadmap for understanding and addressing multifaceted challenges for future crises. It emphasizes the importance of proactive policy design, effective communication, and community engagement. It also highlights the significance of adaptive strategies and cross-sector collaboration, encouraging the consideration of diverse perspectives and leveraging community strengths. By leveraging the lessons learned from COVID-19, the model empowers communities to build resilience and respond more robustly to future crises.

6 Conclusion

This study investigated the interplay between government policies and initiatives, public risk perception, and social capital in Beppu City’s response to COVID-19, emphasizing the multifaceted nature of pandemic management.

The study revealed various key findings. It identified social capital as a key factor in effective COVID-19 response. Communities with strong social ties and trust were more likely to understand COVID-19 and perceive government policies favorably. It also emphasizes the role of social capital in fostering community resilience and collective action during crises, showing how social networks and trust can influence pandemic responses and reduce negative impacts.

Moreover, the study revealed that specific measures like financial aid, lockdown, and frequent and transparent information sharing were positively associated with a reduced impact of COVID-19. This suggests that such measures can be effective tools for mitigating the effects of a pandemic. Interestingly, while financial aid and information sharing were effective in reducing the pandemic’s negative impacts, material aid did not have a significant influence, suggesting that the scale and visibility of material aid initiatives may influence their effectiveness.

Based on these results, the study proposes a Social-Ecological Model for Pandemic Management (SEMPM). This model emphasizes the interconnectedness of individual, community, and policy levels in shaping health crisis response. It shows how these factors collectively influence risk perception, adherence to health guidelines, and community resilience, providing a comprehensive framework for guiding effective pandemic management strategies. Effective pandemic management requires proactive policy design, clear communication, and strong community engagement. The SEMPM offers valuable insights for building community resilience and fostering a more robust response to future crises.

The study offers valuable insights for integrating social capital, risk perception, and government policy into future crisis management strategies, leading to communities better equipped to navigate global health challenges. By leveraging social capital through targeted policies, communication strategies, and collaborative community engagement, future pandemic responses can be more effective and equitable.

Practical and Policy Recommendations:

The experience of Beppu City in managing COVID-19 highlights several key findings that can inform practical approaches and policy decisions for future pandemic responses:

• ■ Policies should consider leveraging and strengthening existing social capital within communities. This includes supporting community organizations, fostering social connections and cohesion, empowering existing social networks to disseminate information, and supporting vulnerable populations.
• ■ To ensure clear, consistent, and transparent communication that shapes public risk perception and adherence to public health measures, governments should develop culturally sensitive communication strategies utilizing multiple channels to reach all community members.
• ■ Regular assessments of community needs and social capital levels are essential when designing effective policies. This allows for a tiered approach that addresses variations in the needs of different communities, social capital, and vulnerability.
• ■ Policy development and implementation should also consider cultural norms and communication styles, with particular attention to groups with lower social capital, like the foreign population and isolated seniors, to ensure they receive adequate support.

Limitations and future research

While acknowledging limitations in this study, we propose future research directions to address these identified gaps.

• ■ An interesting observation was a negative relationship between social capital and government-issued COVID-19 information sharing. This suggests that communities with strong social networks rely less on official channels for information. Future studies might explore this further, evaluating the type, source, and quality of information disseminated within these networks, as well as the role of social capital in filtering information during crises.
• ■ The use of convenience sampling may limit the generalizability of the findings to the broader population of Beppu City. In addition, the inability to recruit the desired sample size due to limited access to potential participants in Beppu City may have affected the representativeness of our sample and the generalizability of our results. Future research could address these limitations by using more rigorous sampling methods to improve the generalizability of the findings to the broader population of Beppu City. Additionally, future studies could use larger sample sizes and employ longitudinal study designs to establish causal relationships among variables.
• ■ While Beppu City offers a potentially interesting case study, the findings from this research may not be directly generalizable to the entire country. Future research that incorporates data from other Japanese cities would be necessary to understand the broader national picture.
• ■ The Social-Ecological Model for Pandemic Management (SEMPM) proposed in this paper has important implications for research, policy, and practice in pandemic preparedness and response:
  • ○ Researchers can validate and refine the SEMPM in various contexts to improve its effectiveness, exploring differences in community resilience, policy impacts, and behavioral responses.
  • ○ Policymakers can use the SEMPM to create integrated strategies that address community needs, enhance public health outcomes, and improve policy communication and support mechanisms.

Footnotes

See appendix B for details of the measurement items and the definitions of the constructs abbreviated in Table 3

The two-direction arrow means that there is a mutual reinforcement relationship. The one-direction arrow implies that one factor has an impact on the other.

References

• [1]  OECD. (2020a) The territorial impact of COVID-19: Managing the crisis across levels of government. Retrieved from: The territorial impact of COVID-19: Managing the crisis across levels of government (oecd.org)
• [2]   Levitt,  E. E.,  Gohari,  M. R.,  Syan,  S. K.,  Belisario,  K.,  Gillard,  J.,  DeJesus,  J.,  Levitt,  A., &  MacKillop,  J. (2022). Public health guideline compliance and perceived government effectiveness during the COVID-19 pandemic in Canada: Findings from a longitudinal cohort study. The Lancet Regional Health - Americas, 9, 100002. https://doi.org/10.1016/j.lana.2022.100002
• [3]   Putnam,  R. D. (2000). Bowling alone: The collapse and revival of American community. Simon & Schuster.
• [4]   Wu,  C. (2021). Social capital and COVID-19: a multidimensional and multilevel approach. Chinese Sociological Review, 53(1), 27–54.
• [5]   Wong,  A. S. Y., &  Kohler,  J. C. (2020). Social capital and public health: responding to the COVID-19 pandemic. Globalization and Health, 16(1), 1–12.
• [6]   Makridis,  Christos A., and  Cary  Wu. (2021). How social capital helps communities weather the COVID-19 pandemic. PLoS ONE 16 (1): e0245135.
• [7]   Liu,  S.,  Zhu,  J.,  Liu,  Y.,  Wilbanks,  D.,  Jackson,  J. C., &  Mu,  Y. (2022). Perception of strong social norms during the COVID-19 pandemic is linked to positive psychological outcomes. BMC Public Health, 22(1), 1403.
• [8]   Lee,  C. T.,  Kanji,  R.,  Wang,  A. H.,  Mamuji,  A.,  Rozdilsky,  J., &  Chu,  T. (2021). Cultural contexts during a pandemic: a qualitative description of cultural factors that shape protective behaviours in the Chinese-Canadian community. BMC Public Health, 21(1), 1897.
• [9]   Bicchieri,  C. (2020). New Social Norms During the Pandemic. Omnia. https://omnia.sas.upenn.edu/story/new-social-norms-during-pandemic
• [10]   Bronfenbrenner,  U. (1977). Toward an experimental ecology of human development. American psychologist, 32(7), 513.
• [11]   McLeroy,  K. R.,  Bibeau,  D.,  Steckler,  A., &  Glanz,  K. (1988). An ecological perspective on health promotion programs. Health education quarterly, 15(4), 351–377.
• [12]   Latkin,  C.,  Dayton,  L. A.,  Yi,  G.,  Konstantopoulos,  A.,  Park,  J.,  Maulsby,  C., &  Kong,  X. (2021). COVID-19 vaccine intentions in the United States, a social-ecological framework. Vaccine, 39(16), 2288–2294.
• [13]   Naidoo,  D.,  Meyer-Weitz,  A., &  Govender,  K. (2023). The Social Ecological Model: A Framework for Understanding COVID-19 Vaccine Uptake among Healthcare Workers—A Scoping Review. Vaccines, 11(9), 1491.
• [14]   Lun,  P.,  Gao,  J.,  Tang,  B.,  Yu,  C. C.,  Jabbar,  K. A.,  Low,  J. A., &  George,  P. P. (2022). A social ecological approach to identify the barriers and facilitators to COVID-19 vaccination acceptance: A scoping review. PloS ONE, 17(10), e0272642.
• [15]   Twersky,  S. E.,  Scott,  T.,  Taracena,  L., &  Souza,  M. (2023). Applying the Social Ecological Model to Understand Support for Home Healthcare Workers and Covid-19 Vaccine Uptake in NJ. Home Health Care Management & Practice, 10848223231214506.
• [16]   Jang,  S. H. (2022). Social-ecological factors related to preventive behaviors during the COVID-19 pandemic in South Korea. PLoS ONE, 17(3), e0266264.
• [17]   Vilme,  H.,  Akin-Odanye,  E. O.,  Sauls,  D. L.,  De Leon,  J.,  Paul,  C.,  Brown Speights,  J. S., &  López,  I. A. (2022). A social-ecological exploration of college and university students’ COVID-19 infection preventive behaviors. American Journal of Health Education, 53(4), 256–265.
• [18]   Timans,  R.,  Wouters,  P., &  Heilbron,  J. (2019). Mixed methods research: what it is and what it could be. Theory and Society, 48, 193–216.
• [19]   ABBADI,  H., &  SAWAGUCHI,  M. (2024). A Theory of Change for Managing the Challenges Posed by COVID-19 in Beppu, Japan: A Social Capital Approach. International Journal of Japan Association for Management Systems, 16(1), 29–50.
• [20]  Beppu Tourism. (n.d.). Home page. Retrieved from https://beppu-tourism.com/en/
• [21]   Etikan,  I.,  Musa,  S. A., &  Alkassim,  R. S. (2016). Comparison of convenience sampling and purposive sampling. American Journal of Theoretical and Applied Statistics, 5(1), 1–4.
• [22]   Cochran,  W. G. (1977). Sampling techniques (3rd ed.). John Wiley & Sons.
• [23]   Byrne,  B. M. (2016). Structural Equation Modeling With AMOS: Basic Concepts, Applications, and Programming (3rd ed.). Routledge. IBM SPSS Amos. (n.d.). IBM. Retrieved from https://www.ibm.com/products/structural-equation-modeling-sem.
• [24]   Pedrosa,  A.L.,  Bitencourt,  L.,  Fróes,  A.C.F.,  Cazumbá,  M.L.B.,  Campos,  R.G.B.,  de Brito,  S.B.C.S., &  Simõese Silva,  A.C. (2020). Emotional, Behavioral and Psychological Impact of the COVID-19 Pandemic. Frontiers in Psychology, 11:566212.
• [25]   Huang,  I. Y. F. (2020). Fighting COVID-19 through government initiatives and collaborative governance: the Taiwan experience. Public Administration Review, 80(4), 665–670.
• [26]   Leape,  J. (2020). IGC COVID-19 guidance note: containment strategies and support for vulnerable households.
• [27]  OECD. (2020b). Transparency, communication and trust: The role of public communication in responding to the wave of disinformation about the new Coronavirus. Retrieved from: https://www.oecd.org/coronavirus/policy-responses/transparency-communication-and-trust-bef7ad6e/.
• [28]   Barrafrem,  K.,  Tinghög,  G., &  Västfjäll,  D. (2021). Trust in the government increases financial well-being and general well-being during COVID-19. Journal of Behavioral and Experimental Finance, 31, 100514.
• [29]  OECD. (2020c). Social economy and the COVID-19 crisis: current and future roles (Policy Responses to Coronavirus (COVID-19). Retrieved from: https://www.oecd.org/coronavirus/policy-responses/social-economy-and-the-covid-19-crisis-current-and-future-roles-f904b89f/
• [30]   Schneider,  C. R.,  Dryhurst,  S.,  Kerr,  J.,  Freeman,  A. L.,  Recchia,  G.,  Spiegelhalter,  D., &  van der Linden,  S. (2021). COVID-19 risk perception: a longitudinal analysis of its predictors and associations with health protective behaviours in the United Kingdom. Journal of Risk Research, 24(3–4), 294–313.
• [31]   Dryhurst,  S.,  Schneider,  C. R.,  Kerr,  J.,  Freeman,  A. L.,  Recchia,  G.,  Van Der Bles,  A. M., &  Van Der Linden,  S. (2022). Risk perceptions of COVID-19 around the world. In COVID-19 (pp. 162–174). Routledge.
• [32]   Grignoli,  N.,  Petrocchi,  S.,  Bernardi,  S.,  Massari,  I.,  Traber,  R.,  Malacrida,  R., &  Gabutti,  L. (2021). Influence of Empathy Disposition and Risk Perception on the Psychological Impact of Lockdown During the Coronavirus Disease Pandemic Outbreak. Frontiers in Public Health, 8, 567337.
• [33]   Van Bavel,  J.J.,  Baicker,  K.,  Boggio,  P.S.,  Capraro,  V.,  Cichocka,  A.,  Cikara,  M.,  Crockett,  M.J.,  Crum,  A.J.,  Douglas,  K.M.,  Druckman,  J.N. and  Drury,  J. (2021). Using social and behavioural science to support COVID-19 pandemic response. Nature Human Behaviour, 5(5), 460–471.
• [34]  Institute for Social Capital. (n.d.). Definitions of Social Capital. Retrieved from https://www.socialcapitalresearch.com/literature/definition/
• [35]   Aldrich,  D. P., &  Meyer,  M. A. (2015). Social capital and community resilience. American Behavioral Scientist, 59(2), 254–269.
• [36]   Tatarko,  A.,  Schmidt,  P.,  Lebedeva,  N., &  Ryzhkova,  N. (2022). Social capital and the COVID-19 pandemic threat: The Russian experience. Frontiers in Sociology, 6, 957215.
• [37]   Ingraham,  C. (2021, February 11). Researchers identify social factors inoculating some communities against coronavirus. The Washington Post. Retrieved from https://www.washingtonpost.com/business/2021/02/11/social-capital-covid-spread/
• [38]   Li,  T. W.,  Lee,  T. M. C.,  Goodwin,  R.,  Ben-Ezra,  M.,  Liang,  L.,  Liu,  H., &  Hou,  W. K. (2020). Social capital, income loss, and psychobehavioral responses amid COVID-19: a population-based analysis. International journal of environmental research and public health, 17(23), 8888.
• [39]   Perry,  J. (2021). Trust in public institutions: Trends and implications for economic security. The UN library.
• [40]   Hair,  J. F. (2010). Multivariate data analysis: A global perspective, 7th ed. Pearson Education.
• [41]   Hair Jr,  J. F.,  Matthews,  L. M.,  Matthews,  R. L., &  Sarstedt,  M. (2017). PLS-SEM or CB-SEM: updated guidelines on which method to use. International Journal of Multivariate Data Analysis, 1(2), 107–123.
• [42]   Fornell,  C., &  Larcker,  D. F. (1981). Structural equation models with unobservable variables and measurement error: Algebra and statistics. Journal of Marketing Research, 18(3), 382–388.
• [43]   Fornell,  C.,  Johnson,  M. D.,  Anderson,  E. W.,  Cha,  J., &  Bryant,  B. E. (1996). The American customer satisfaction index: nature, purpose, and findings. Journal of Marketing, 60(4), 7–18.

Appendices

Appendix A. Summary of literature review thematic analysis

Categories	Sub-Themes	Patterns and Insights	Contradictions and additional notes
Government policies and initiatives	Impact on public perception of risk Influence on public adherence Financial and material support	Strict policies can cause economic hardship, reducing adherence. Financial and material support can build trust and encourage adherence. Transparency in communication is essential for trust-building.	Can policies be designed to address economic hardship while maintaining adherence?
Public adherence to government policies and public health guidelines	Policy communication influence Strict policies influence	Transparent communication builds trust and adherence. Strict policies can erode trust and adherence, especially without social safety nets. Public messaging should be culturally appropriate and consider diverse needs.	Can the balance be achieved between effective measures and adherence? Need for tailoring communication to specific audiences.
People’s perception of the risk of the disease	Government policies impact Factors influencing perception Trust in institutions	Strict policies and high case numbers can heighten risk perception. Trust in institutions and personal experience also influence risk perception. Media portrayals and misinformation can distort risk perception.	How does trust in different institutions (government, science) impact risk perception? The role of social media and traditional media in shaping risk perception.
Underlying social capital within the community	Social capital influence	Strong social capital leads to better adherence, risk perception, and resilience. High social capital can lead to better acceptance of government policies. Social capital can be eroded by economic hardship and social inequalities.	Can governments support and leverage social capital during pandemics?

Appendix B. Details of measurement items

	Panel A. Reliability and validity of constructs
Constructs	Items	Loading	Cronb. α	CR	AVE
Perception (PER)	My perception of how serious the Covid-19 pandemic was/is has changed completely over time. My perception of the Covid-19 pandemic changed completely due to my personal experience with the virus (personal experience includes those from friends or family members who caught the virus). I think the Covid-19 pandemic was/is as serious as it was portrayed by health professionals, policy makers and the news media.	0.905***	0.744	0.859	0.678
		0.929***
		0.594***
Social capital (SC)	I received adequate support and assistance from aid organizations (Government, University, Volunteers, Neighbors etc) I often participate in online/virtual community or social groups (ethnic, study group, religious, etc.) which kept occupied during the pandemic My involvement in online or virtual community or social groups helped me to follow the COVID-19 pandemic updates and regulations such social distancing guidelines) Keeping in touch with my community/social/religious and other clubs kept me safe during the pandemic.	0.633***	0.815	0.880	0.651
		0.824***
		0.876***
		0.869***
COVID-19 Impact (C19Imp)	My work/study was completely disrupted by the COVID-19 pandemic, and I lost a lot of time and income. The C0VID-19 pandemic affected my personal and family life so much that I became depressed/almost depressed I am yet to recover from the financial, emotional, physical and psychological impact of the COVID-19 pandemic.	0.882***	0.876	0.924	0.801
		0.895***
		0.908***
Government policy/initiative (GPI)	Beppu City Government Covid-19 measures and countermeasures were crucial to slow down the spread of the virus I am satisfied with the Beppu local government’s response to Covid-19 pandemic Beppu local government disseminate crucial information on Covid-19 pandemic on a timely and professional basis	0.920***	0.901	0.938	0.835
		0.920***
		0.900***
Financial aid (FA)	The financial support I received during the pandemic (adequate) lessened my financial burden/difficulties The financial support I received during the pandemic was enough to provide my basic economic needs and necessities. I am satisfied with the financial support I received during the pandemic. The financial support I received during the pandemic helped me to be emotionally and mentally stable during the pandemic	0.899***	0.949	0.963	0.867
		0.904***
		0.955***
		0.964***
Lockdown (LD)	Government lockdown/travel restrictions contributed to low transmission of the virus. Lockdown and travel restrictions were timely implemented which was crucial during the pandemic. Lockdown and travel restrictions were effective because they helped me to avoid the virus through the pandemic.	0.909***	0.811	0.913	0.841
		0.917***
		0.925***
Covid-19 information (C19Inf	I Received frequent and sufficient information about personal protection against the virus during the pandemic. The information I received about Covid-19 was helpful and I used it to protect myself from the Virus. I am satisfied with how the responsible agencies disseminate information about Covid-19 to everyone.	0.827***	0.764	0.862	0.677
		0.770***
		0.869***
Material Aid (MA)	I received the necessary material support during the pandemic. I am satisfied with the material support I received during the pandemic. The material support I received helped me to protect myself from the Covid-19 virus. The material support I received was inadequate. and did not in any way help me during the covid-19 pandemic.	0.845***	0.899	0.930	0.769
		0.927***
		0.947***
		0.779***

Appendix C. Factor loadings-PLS- structural equation modeling (Consistent)

	Original sample (O)	Sample mean (M)	STDEV	T statistics	P values
Gpolicy1 <- GPI	0.918	0.917	0.015	60.434	0.000
Gpolicy2 <- GPI	0.919	0.918	0.016	56.474	0.000
Gpolicy3 <- GPI	0.903	0.903	0.021	43.300	0.000
Impact1 <- C19Imp	0.882	0.881	0.024	36.236	0.000
Impact2 <- C19Imp	0.895	0.894	0.021	41.970	0.000
Impact3 <- C19Imp	0.908	0.908	0.013	68.878	0.000
Perception1 <- PER	0.905	0.904	0.027	34.135	0.000
Perception2 <- PER	0.929	0.929	0.013	72.055	0.000
Perception3 <- PER	0.594	0.590	0.074	8.039	0.000
SC1 <- SC	0.649	0.650	0.038	16.853	0.000
SC2 <- SC	0.821	0.820	0.029	28.406	0.000
SC3 <- SC	0.868	0.866	0.026	33.635	0.000
SC4 <- SC	0.860	0.857	0.030	28.283	0.000
Finaid1 <- FA	0.899	0.899	0.028	31.797	0.000
Finaid2 <- FA	0.904	0.904	0.025	35.619	0.000
Finaid3 <- FA	0.955	0.955	0.015	65.807	0.000
Finaid4 <- FA	0.964	0.965	0.007	129.670	0.000
Glockd1 <- LD	0.909	0.908	0.023	40.205	0.000
Glockd2 <- LD	0.925	0.925	0.013	68.506	0.000
Glockd3 <- LD	0.917	0.917	0.018	54.356	0.000
Inform_covid1 <- C19Inf	0.827	0.825	0.032	25.695	0.000
Inform_covid2 <- C19Inf	0.770	0.768	0.033	23.207	0.000
Inform_covid3 <- C19Inf	0.869	0.869	0.016	53.590	0.000
Mataid1 <- MA	0.845	0.843	0.051	16.466	0.000
Mataid2 <- MA	0.927	0.927	0.021	44.756	0.000
Mataid3 <- MA	0.947	0.948	0.014	70.011	0.000
Mataid4 <- MA	0.779	0.773	0.091	8.591	0.000