Households’ Socioeconomic Vulnerability Assessment Due to COVID-19 Outbreak: A Web-Based Survey in Bangladesh

Households ’ Socioeconomic Vulnerability Assessment Due to COVID-19 Outbreak: A Web-Based Survey in Bangladesh. ABSTRACT Aim: COVID-19, which can be considered a disastrous event, has created not only a public health emergency but also a major socio-economic crisis in Bangladesh. This study, therefore, aimed to assess households ’ socioeconomic vulnerability due to the COVID-19 outbreak in the country. Methods: A quantitative cross-sectional survey was performed among 404 households from different districts in Bangladesh. The socioeconomic vulnerability index (SeVI) was developed using data collected from participants via an online-based self-reported questionnaire that included demographic, social, economic, and physical characteristics as well as exposure to COVID-19. Results: The socioeconomic vulnerability index (SeVI) was calculated as an index score of 0.405 where social, economic, and COVID-19 exposure were reported to be the more impactful components of adaptive capacity, sensitivity, and exposure, respectively. Furthermore, the economic options for households were greatly limited by the consequences of the COVID-19 pandemic. Practical implications: This study may help to identify the socioeconomic issues that resulted from the COVID-19 outbreak in the country and instruct the policymakers and corresponding authorities on which areas to emphasize for policy implementation so that households become socially and economically less susceptible to the COVID-19 outbreak as well as related infectious diseases and disastrous events in the upcoming years. Conclusion: This study found socioeconomic vulnerability among Bangladeshi households. The corresponding authorities should adopt policy initiatives to minimize the socioeconomic vulnerability due to the COVID-19 outbreak in the country.


INTRODUCTION
The novel coronavirus outbreak in 2019, also called the COVID-19 pandemic, is a world health crisis that has led to an unparalleled human life and livelihood disaster, disrupted economic processes across industries, stopped public transport networks, and limited global social interactions [1]. At the end of December 2019, the outbreak of COVID-19 started in China and spread rapidly around the globe, with 2,961,025 deaths and 137,322,644 positive cases as of April 13, 2021 [2]. COVID-19 was mentioned as a global emergency by the World Health Organization (WHO), which took place on January 30, 2020, and as a global pandemic thereafter on March 11, 2020 [3], while Bangladesh announced their first three cases on March 8, 2020 [4]. A total of 691,957 confirmed cases were recorded, comprising 9,822 deaths in Bangladesh as of April 13, 2021 [5]. In order to combat the spread of COVID-19 disease, the government declared a general holiday from March 26, 2020 through the first wave. With the exception of essential services, all government and non-government institutions have been closed since the lockdown began [6]. After a step-by-step extension of those holidays with the increase of confirmed cases, all the offices were reopened to a limited extent in compliance with health rules on May 31, 2020 [7]. But, nine months later, the government was again forced to announce a lockdown on April 5, 2021 due to the second wave of the COVID-19 outbreak and further imposed a week-long strict lockdown from April 14, 2021 due to a high increase in confirmed cases and deaths [8,9]. All the government and non-government institutions, as well as public transport, were again closed throughout the country, apart from emergency services. On the other hand, all schools, colleges, and universities have remained closed since March 18, 2020, and have extended this closure till May 23, 2021 [10].
Overall, this pandemic situation is not constrained to health crisis anymore, instead, it is becoming an unparalleled crippling effect on the social and economic contexts, and it appears with prolonged effects over time [11]. The economic sector throughout the corona outbreak confronted a sharp decline and its potential impacts and consequences have now been discussed under the name "Coronanomics" [12] and some of which as well call it "black swan" [13]. The world has been experiencing an emerging "de-globalization" as a result of this pandemic, which has resulted in inter-national lockdown, halting normal product progressions, and unprecedented market and manufacturing breakdowns [14]. Bangladesh is also not exceptional from the world, rather facing this downward situation in various sectors of the country, equivalently [15]. It has already experienced an enormous economic and social instability as a result of the emerging COVID-19 pandemic [16]. For instance, the price rise in regular necessities has been noticed due to limited supplies and business owners and vendors have discontinued functioning because of countrywide lockdown and fear of infection. In addition, due to the loss of employment and wages during this situation, middle-income, lower-income, and dailywage earner households found themselves in a serious financial dilemma. They are dropped into the extremely poor with their last savings consumed [17,18]. Moreover, decrease of export earnings and income from tourism due to the enforced border constraints and extended lockdowns has already been seen as the main contributors to a coming socioeconomic shock, as millions of populations of the country are engaged in these sectors [19]. Furthermore, according to the UN report "World Economic Situation and Prospects 2021", the country is estimated to have decreased its economic growth from 8.4 percent in the 2018-2019 fiscal year to 4.3 percent in the 2019-2020 fiscal year due to the COVID-19 outbreak [20]. Thus, this indicates that the socioeconomic status of all forms of people has been greatly impacted due to this pandemic. An analysis revealed that the income of households due to COVID-19 outbreak has been decreased by 29 percent and specifically in Dhaka Metro, it has been decreased by 34 percent. Besides the economic impact, they are also adversely impacted on their social networks and physical capital. For example, as result of COVID-19 pandemic, households have lost their remittance support from the earning members and 40 percent of households reported an expansion in food spending [21]. Human Development Research Centre (HDRC) undertook another study to determine both the immediate and long-term socioeconomic effects of the COVID-19 outbreak on urban deprived populations. The findings revealed that 22.7 percent of households' children are not continuing their study since lockdown and only 2 percent have the facilities for televisionbroadcasted academic programmes or virtual classes. Furthermore, 11 percent, 54.9 percent, 81 percent, 69.3 percent, and 85.2 percent of the beneficiary households have lost their asset, jobs (permanently or temporarily), amount of savings, rent payment capacity, and food consumption capacity, respectively [22]. Aggregately, the country is experiencing a severe surge in socioeconomic crisis and a hazardous situation.
Previous research has also demonstrated that the COVID-19 pandemic has triggered a socio-economic crisis in several countries around the world, including Bangladesh. There was a study that investigated regional variations in socioeconomic vulnerabilities connected to the COVID-19 outbreak. The findings revealed that India's COVID-19 risk remained highly variable between states and union territories. The risks associated with COVID-19 in India were driven by the country's inherent demographic, socioeconomic, and health infrastructure features [1]. Another research determined the socioeconomic vulnerability of communities in the state of Ceará, which is the epicenter of the COVID-19 outbreak in the northeastern region of Brazil. The findings demonstrated the consequences of the pandemic in an economic framework dominated by the service industry, which is characterized by high levels of human contact and social interaction. Ceará is affected by the crisis as a result of a number of demographic, social, and economic factors that are unique to the region [23]. There was additional study on socio-economic vulnerability to COVID-19 in the Greater Kampala Metropolitan Area (GKMA). The most vulnerable parishes (24.5%) are located in metropolitan cities with retail malls, banks, and transportation centers. About half of the parishes in the GKMA were moderately vulnerable (47.3%), with 28.2% being lowly vulnerable [24]. Further research examined the pandemic's socioeconomic effects on households, adults, and children in low-income nations. According to that survey, 256 million people, or 77% of the population, live in homes that have lost income due to the outbreak. Food insecurity and a lack of access to medical and fundamental foods made it difficult to cope with the loss [25]. Similarly, a previous study discovered that many people in a Bangladeshi metropolitan city had lost their jobs, particularly day laborers, maid servants, and private car drivers, among other occupations. Furthermore, financial stress has been exerted on people of all socioeconomic backgrounds [15]. Likewise, a study identified socioeconomic crises like unemployment, deprivation, hunger, and social conflicts among Bangladeshis as a result of COVID-19 [17]. In a recent study, it was shown that the COVID-19 lockdown has caused untold misery and suffering to all, especially those residing in Bangladeshi low-income areas. A lot of people lost employment and enterprises. Many people cannot even manage the minimal amount of food required for a healthy lifestyle due to income reductions. They are forced to liquidate household assets, spend savings, and take out loans that they will struggle to pay back. Domestic violence has escalated due to economic hardship, frustration, and apathy. Many types of health vulnerabilities exist. and many children are dropping out of school [22].
Socioeconomic vulnerability analysis is, therefore, the requisite to take the long-term and suitable response as well as to develop adaptation strategies to COVID-19 outbreakinduced hazardous circumstances around the country. In addition, vulnerability assessment guides us in conceptualizing the distinct spectrum of components that contribute significantly to household adaptive capacities, and it determines the broad scope of initiatives used to extensively accommodate and properly assess interconnectedness between humans and their socio-physical surroundings. Moreover, recognizing a society's socioeconomic vulnerability may help to explain why the effects of a comparable catastrophe may vary from one place to another [26]. Moreover, the socioeconomic vulnerability research has the potential to improve crisis response measures by improving knowledge of catastrophic repercussions at the household level [27,28]. Although there were numerous researches have been conducted in response to COVID-19 pandemic in Bangladesh, most of them were based on mental health issues. A few studies were found on the perspective of socioeconomic issues in Bangladesh [15,17,22]. However, the main point which encouraged the authors to conduct this research that no study was found that assessed socioeconomic vulnerability to COVID-19 outbreak at the household level by using the socioeconomic vulnerability index (SeVI). To fill up this gap, this research was designed with the aim to assess households' socioeconomic vulnerability to COVID-19 outbreak in Bangladesh. This study may help to identify the contributors of socioeconomic issues that derived from the COVID-19 outbreak in the country and persuade the corresponding authorities in the development of policy measures to tackle this significant crisis.

CONCEPTUALIZATION OF VULNERABILITY
The physical or ethical vulnerability of specific societal groups or communities to possible risks or losses induced by catastrophic occurrences is referred as social and economic vulnerability [29]. Crisis like the COVID-19 pandemic have an impact on global dimensions resulting in physiological, sociopolitical, economic, and social uncertainty [23,30]. Similarly, Bangladesh is experiencing an unparalleled economic and social hardship as a result of the current novel Corona virus outbreak [16,31]. Throughout this pandemic, it is unavoidable that a substantial humanitarian and socioeconomic catastrophe has emerged in the country, resulting in long-term effects with a variety of weaknesses which are probably to produce the harshest outcomes. However, the concept of vulnerability has been taken into account by a number of assessments in the globe [1,23,24,32]. In addition, there are also various techniques for measuring vulnerability that take into account socioeconomic factors for particular hazards [33][34][35], but this study has employed the dimensions of the intergovernmental panel on climate change (IPCC) framework to measure socioeconomic vulnerability according to the domains of [27]. This assessment has considered the engagement among demographic, social, economic, physical, and exposure to COVID-19 in developing the vulnerability index. Adaptive capacity, sensitivity, and exposure (see Table 1 for details) are the three dimensions by which vulnerability will be described. Considering COVID-19 outbreak in Bangladesh, the term 'adaptive capacity' here refers to a summation of capacities and social capital among households' in a community or country that can help mitigate the socioeconomic implications of COVID-19 outbreak. At the same time, sensitivity, in this context, focuses on the aspects underlying economic and physical predisposition to COVID-19 outbreak. Similarly, exposure refers to individuals, residents, or other aspects present in the impact areas, which are thus adversely affected as a result of significant COVID-19 outbreak risks. Finally, vulnerability is the extent to which the socioeconomic system of a household is vulnerable or unable to comply with the consequences of COVID-19 outbreak. We used a range of 0 to 1 for developing the socioeconomic vulnerability index where 0 indicates lower vulnerability and 1 indicates higher vulnerability.

Survey Settings and Study Design
This was an online-based quantitative cross-sectional survey study using primary data. This study utilized primary data because primary data are data collected specifically for a research problem using processes tailored to the specific research problem. Primary data collection adds original insights to the current social knowledge base. Moreover, the broader research community may be able to utilize content generated by different scholars [36]. Data collection was conducted among the households of different districts in Bangladesh. Data was gathered from 10th to 28th June of 2021. The responses were extracted through an online self-reported survey questionnaire (using the google survey tool-Google Forms), employing a convenience sampling technique. Since this study was based on primary data, we used web-based data collection rather than publicly accessible and nationally representative data for this analysis. In addition, this study used a web-based survey methodology because it was challenging to conduct a physical survey during the COVID-19 pandemic. A pretesting survey was conducted prior to the questionnaire design. After that, expert consultations were used to finalize the questionnaire. In order to validate the participants' understanding of the questionnaire items and to Percentage of households with access to community hygiene Were there any arrangements for handwashing at a specific place in your area from the government or private or own initiatives due to the coronavirus outbreak? 1=yes 0=no [22] fix the issues in wording, the questionnaire was also tested on a small sample of randomized internet users. The questionnaire was first written in the Bengali language and then converted to English for preparing the study report. The questionnaire was circulated through Facebook, Messenger, LinkedIn, WhatsApp, email, and other networks of the authors. We narrowed our survey to eight divisions of the country in order to prevent the data from being skewed or biased towards a certain region by using stratified sampling. The questionnaire was then circulated by the authors and our known networks, each of whom resided in one of the eight divisions. Each of them was instructed to collect a minimum of 50 responses from their particular division. It was secured by the responses to a question such as "From which division are you from?" which was included in the introduction section of the questionnaire along with the consent agreement. The main survey questions from the questionnaire have been presented in Table 1.
However, we received a total of 415 responses where only two participants did not give their consent to participate, as well as nine responses that were not usable as a result of missing and confusing cases. Finally, 404 respondents were included in the analysis. The main respondents of this study were the household heads of Bangladesh. However, as this was an online-based study and most of the household heads were not experienced in using online platforms, we have relaxed the participation inclusion criteria. Thus, the eligibility to participate in the survey included any adult member of a household using the internet, able to understand the aims of the research, and prepared to participate voluntarily. The purpose of the study was explained in the first section of the online questionnaire. Each of the participants had read the purpose carefully and had voluntarily given consent to participate.

Development of Socioeconomic Vulnerability Index (SeVI)
The SeVI was created using a combined indicator-based model that included three key dimensions: adaptive capacity, sensitivity, and exposure [27,37]. All those dimensions were divided into five components: demographic, social, economic, physical, and exposure to COVID-19. The index collectively constituted several characteristics of an individual component in accordance with the COVID-19 outbreak in terms of numerical meaning, and therefore reflected the household's condition in reference to these components. The SeVI was developed based on five components, including 26 indicators (for more information, see Table 1). Each indicator was standardized as an index value as they were calculated on different scales. To produce an index score for each indicator for Bangladesh 'i', we employed the given Equation 1, which was adapted from a UNDP life expectancy index [27,38]. Developed for the purposes of this questionnaire BDT: Bangladeshi Taka (Currency); WASH: Water, sanitation, and hygiene where, Xi is original value of indicator for the household/country, Xmax is the highest value of indicator for the household/country, and Xmin is the lowest value of indicator for the household/country. Thus, the indicator index generated numerical values which signified the relative vulnerability status of the country (collected by an aggregated response of households). The numerical values ranged from zero to one for each indicator.
Once each contributing indicator accumulated the index score, the average index score of all indicators within the same component was taken into account to determine the component vulnerability score through Equation 2.
where DOi is the component-scores of vulnerability index 'i', SISk is the standardized mean score of each indicator within the component (here k is the number of indicators within the concerned component indicated in Table 1).
After accomplishing the component value of vulnerability, we continued with the dimension value of vulnerability through Equation 3 to Equation 5. In this study, dimensions were taken into assessment as adaptive capacity, sensitivity, and exposure to COVID-19 outbreaks following the framework of IPCC [39].
where j, l, and m indicate numbers of components under adaptive capacity, sensitivity, and exposure (given in Table 1), respectively, while 'i' represents the whole country, Bangladesh. Utterly, Equation 6 has determined the socioeconomic vulnerability index of the country 'i'.
where AC means adaptive capacity. We hypothesized that the interaction among the three vulnerability factors was undetermined and determined simply by the geographical features of the area in which they were found. However, we also hypothesized that SeVI had a direct connection with the system's exposure and sensitivity, and an inverse linkage with its adaptive capacity [40]. As a result, when calculating the index, we utilized an inverse value of adaptive capacity (1 minus dimension score). In this study, the SeVI was ranged from 0 to 1, where 0 indicates lower vulnerability and 1 indicates higher vulnerability. The SeVI was developed by utilizing descriptive statistics. The frequency and percentage of the scheduled variables were analyzed by employing descriptive statistics. Microsoft Excel 2019 and the Statistical Package for Social Science (SPSS) 20.0 Windows version were used to process and analyze the data. Finally, the output was interpreted as the final report of this study.

Ethical Considerations
Prior to starting filling out the questionnaire, participants provided their consent and remained anonymous. All respondents were informed about the aim of the study in the introduction section of the questionnaire. The dataset has confirmed the anonymity and confidentiality. Furthermore, this study was reviewed and ethically approved by the Khulna University Research Cell, Khulna-9208, Bangladesh (Reference number: KUECC-2021/06/21).

RESULTS
First, we have outlined the socio-demographic profile of the participants ( Table 2). We have then examined the key outcomes from the component (Figure 1) and dimensionbased vulnerability analysis (Figure 2). In two segments, component-based and dimension-based, we have reported findings of statistical assessment for the socioeconomic vulnerability of this study.

Socio-Demographic Profile
The prevailing socio-demographic characteristics of the participants in this sample have been listed in Table 2. The trend of age indicated that the household heads in the preceding five and a half decades observed substantial shifts in different issues in the country. The country had a higher prevalence of male-headed families. The average size of the household was 4.72 members, which is marginally greater than the national average (4.06 people) [41]. At the same time, a large proportion of households were affected by a loss of consistent income (68.3%), a loss of wealth and savings (69.3%), an absence of availability of community hygiene (53.2%), and the presence of chronically ill individuals (48.0%).

Demographic vulnerability
The vulnerability component score of demographic indicators was noted as 0.411 (details in Table 3). The findings revealed that households were the most vulnerable demographically, with 63.6 percent of solely dependent people (children and older adults). Furthermore, 59.3 percent of households were demographically vulnerable to the pandemic due to their urban location. At the same time, household heads had an average age of 54.94 years, with an index score of 0.483, where 34.9 percent of the household heads age was identified as over 60 years. Moreover, the index score of household size was found at 0.389, where 23.3 percent of households had more than 5 members in their family. The lowest demographic vulnerability was identified as the gender of household heads, where only 10.6% of the households were female-headed.

Social vulnerability
The social vulnerability index was measured as 0.416 (details in Table 3). Indicators of this component demonstrated that the highest social vulnerability was listed for the years of schooling of the household heads, with an index score of 0.585, where the average schooling year of the household heads was found to be 12.28 years. In contrast, the lowest proportion was recorded for those households that received assistance from community leaders (13.1 percentage). Similarly, 54.0 percent of households were identified as socially vulnerable as a result of not borrowing money from friends, family, or neighbors. On the contrary, the indicators of the social component demonstrated that 29.0 and 53.2 percent of the households had access to CBOs and local organizations as well as community hygiene (arrangements for handwashing at a specific place in the community from government/private/own initiatives due to the coronavirus outbreak), respectively.

Economic vulnerability
We found the highest component-based vulnerability index for economic issues where the index score of this component was calculated as 0.547 (details in Table 3). The study findings indicated that the highest economic vulnerability was identified for the participating households without any government employee as well as the loss of their wealth and savings (69.3 percent). Our study noted about 68.3 percent of households were economically vulnerable without stable income. In the same way, this study results listed about 59.9

Physical vulnerability
The vulnerability index of the physical component was determined as 0.244, which was the lowest component-wise vulnerability score of this study (details in Table 3). The households with chronically ill people (48.0 percent) were identified as having a high vulnerability indicator. In contrast, the lowest physical vulnerability noticed for the households WASH (access to safe water and/or soap) behavior, which was found highly satisfactory, where only 3.7 percent of the responding households had no access to WASH facilities. Simultaneously, 31.4 percent of the sample households were found to be physically vulnerable while living in rented housing. Furthermore, in our study, 26.5 percent of households were physically vulnerable due to a lack of access to quality treatment. Moreover, our study outcomes indicated that 20 percent of the households were physically vulnerable without having enough technological support for their children's online education. Furthermore, our research managed to find that 16.6 percent of participating households experienced physical vulnerability due to food insecurity over the last year.

Exposure to COVID-19
The index score of exposure to the COVID-19 component was found to be 0.246 (details in Table 3). The highest COVID-19 exposure was found in households with members who were afraid of COVID-19 (78 percent). On the contrary, our results demonstrated that studied households had lower exposure to COVID-19 with the satisfactory practices of the national guidelines regarding the COVID-19 pandemic in the country, where 10.6 percent of households did not follow the national guidelines. Similarly, our findings showed that most of the household's members had proper knowledge about the symptoms of COVID-19, including 14.1 percent of households that were not well-known about COVID-19 symptoms. In a similar way, our study reported 19.1 percent of households with infected persons and 1.2 percent of households with a dead person.

Dimension-Based Vulnerability
Following the IPCC dimensions together with exposure, sensitivity, and adaptive capacity, this study measured the dimension-based vulnerability (Figure 2).

Exposure
In this study, exposure was encircled by exposure to the COVID-19 component (details in Figure 3). The results demonstrated that the index of exposure was quantified as 0.246, which was the lowest dimension-wise vulnerability score (Figure 2). The contributing indicators of this dimension indicated that households with members afraid of COVID-19 increased their exposure to COVID-19 ( Table 2).

Sensitivity
Sensitivity covered the economic and physical components in this study (details in Figure 3). The sensitivity index score in the study included an index score of 0.396 (Figure 2). The sensitivity was heightened by the worsening economic situation of households, particularly those without a government employee, as well as the loss of wealth and savings ( Table 2).

Adaptive capacity
Adaptive capacity also comprised two components, including demographic and social components (details in Figure 3). The highest dimension-wise vulnerability score was evidenced for adaptive capacity with an index score of 0.428 (Figure 2). The adaptive capacity of households was underscored by the demographic profile, which was primarily characterized by the predominance of solely dependent populations ( Table 2).

Overall Vulnerability Score
This study also measured the overall socioeconomic vulnerability score by following the IPCC dimensions together with exposure, sensitivity, and adaptive capacity. The overall socioeconomic vulnerability index was determined as 0.405 (details in Table 3). A supplemental graph (Figure 3) was also constructed to demonstrate the volume of the relevant indicators (individually) across Bangladesh and inside the specified dimension using a scale of 0 to 1 (more vulnerable if the score is close to 1).

DISCUSSION
The continuing novel Corona virus (COVID-19) outbreak has produced an unparalleled economic and social catastrophe in Bangladesh, despite the fact that the COVID-19 outbreak is deemed a public health disaster [16]. In both countryside and metropolitan regions, COVID-19 induced lockdown has intensified the detrimental effects on job opportunities, household earnings, and livelihood. Simultaneously, the social situation of a plethora number of households across the country has deteriorated as a consequence of employment and income loss during the pandemic. To deal with the difficulties, many families curtailed their consumption of food and sought assistance from their savings, relatives, and governments [31]. Overall, there is a sharp rise of socioeconomic vulnerability has been observed in the country. This study therefore aimed to assess socioeconomic vulnerability due to COVID-19 outbreak in Bangladesh. According to the authors' best knowledge, this is the first study in Bangladesh that measures socioeconomic vulnerability due to the COVID-19 outbreak by utilizing a SeVI following the domains of [27]. The results of the statistical analysis of socioeconomic vulnerability were presented in this study in two sections: component-based and dimensionbased. The findings showed the socioeconomic vulnerability index (SeVI) with an index score of 0.405, while the dimensionbased index had index scores of 0.246, 0.396, and 0.428, respectively, for exposure, sensitivity, and adaptive capacity. Furthermore, social, economic, and COVID-19 exposure were demonstrated as more influential components of adaptive capacity, sensitivity, and exposure, correspondingly. Additionally, the economic options of the households were severely constrained due to the implications of the COVID-19 outbreak.
A major demographic crisis was observed in this study due to the COVID-19 outbreak, which had a significant impact on boosting the socioeconomic vulnerability in Bangladesh. Several demographic indicators were included to assess socioeconomic vulnerability in this study. Among them, the presence of solely dependent populations, urban-based households, and the age index of the household heads were identified as the top three index indicators. This outcome may occur for several reasons. Because previous research indicated that a greater dependency ratio and the extreme aged population (<5 or >65 years) maximized the chances of social and economic vulnerability in a disaster situation [27,42]. Evidence also suggested that areas with a larger proportion of elderly people were more likely to experience serious illnesses or mortality than those with a younger population. Those aged 60 and over tended to have a higher risk of being very ill from COVID-19. Moreover, disease transmission could be accelerated in urban areas due to population density [43]. Therefore, all of these demographic issues mainly contributed to fostering the socioeconomic vulnerability in this study.
Simultaneously, a considerable increasing trend in social vulnerability was documented among the households in this study, highly attributed to the household heads' schooling years. Our study found that households with lower-educated heads were vulnerable to COVID-19 outbreaks. However, literature shows that education can contribute to mitigating the harmful effects of an emergency situation in both direct and indirect approaches. Directly or indirectly, formal education is seen as the major means by which people gain information, skills, and competencies that enhance their adaptive capacity, including cognitive skills, problem skills, better knowledge, and risk perception [44]. Moreover, welleducated people are more knowledgeable about the catastrophic risks [45] and more inclined to plan for crises [46]. Furthermore, a better educational level may lower vulnerability indirectly through a variety of ways and means, including poverty reduction, improved socioeconomic position, and increased social capital. As a consequence, it is logical to anticipate that when confronted with catastrophic hazards, educated citizens, households, and communities are increasingly empowered and adaptable in their reaction to, preparedness for, and recuperation from catastrophes [44]. For the time being, all of these estimates were opposite in this research since the household heads were not well educated enough, and so suffered from an unprecedented socioeconomic vulnerability due to the COVID-19 pandemic. At the same time, a growing amount of literatures promoted the relevance of social network in crisis response, in particular cases starting from natural catastrophes [47] to pandemics [48,49]. But social capital is insufficient when it might offer partial support to households or communities. Rather, maximum gains can be obtained by full social capital support including bonding, bridging, and linking [50,51]. Anyway, evidence suggested adverse impacts resulting from disparities in bonding, bridging, and linking which ultimately reduced the community resilience. For instance, several nations are competing with each other to tackle the ongoing COVID-19 situation. The national government, corporate enterprise, and international players are being survived to supply personalized protection materials such as N95 protective suits and face shields due to the lack of bonding, bridging, and linking among them [50]. Yet, there was an inadequacy of social capital in this study, notably due to the limited access to CBOs or local organizations. In addition, over half of the households in the survey did not have access to community hygiene, indicating a linkage and bridging gap. As a result, these types of social crises exacerbated the country's socioeconomic fragility.
In this study, economic disruptions of households were identified as having the highest level of vulnerability as contributors. According to this study's findings, households without a government employee as well as those with the loss of their assets and savings were the most vulnerable in terms of economic instability. Such an outcome resulted because the previous report noted that Bangladesh has experienced two forms of job loss in every sector except government jobs as a result of the coronavirus pandemic: temporary lockdowninduced joblessness and persistent loss of employment. That study also added that between 12 and 17 million people have lost their jobs as a result of the country's two-month shutdown. Consequently, they utilized their assets and savings to deal with the reduction in income [31]. Furthermore, our research found that households with unstable income and single income earners were economically susceptible in the midst of the pandemic. Such outcomes are noticeable because a substantial number of households have already lost their income as a result of the COVID-19 pandemic-induced employment loss, which ultimately triggered income instability among them [31]. For this reason, households with only one income person may experience economic vulnerability in the country.
Physical vulnerability was another component that contributed to the development of socioeconomic vulnerability. But physical vulnerability was demonstrated to be lower in intensity in this study when compared to demographic, social, and economic vulnerability. However, households with chronically ill people were found to have the maximum level of physical risk in this study. According to scientists, COVID-19 disease is considered to be more harmful to those with chronic diseases, so this observation seemed expected [52]. Additionally, in a prior study, households having a family member with a chronic disease were identified as a key physical vulnerability indicator for developing socioeconomic vulnerability to hazard-related risks [27].
Similarly, COVID-19 exposure was an important component that contributed to the development of a socioeconomic vulnerability index in this study, with households with afraid family members ranking first. This result was significant since an earlier study indicated that the COVID-19 outbreak fostered fear among the Bangladeshi people, leading to socioeconomic crises such as unemployment, hardship, starvation, as well as social conflict [17].
Likewise, this study found higher level of adaptive capacity comparing exposure and sensitivity. Literature suggests that exposure, sensitivity, and adaptive capacity all play a role in determining vulnerability. A lower adaptive capacity in comparison to exposure and sensitivity adds to a high level of vulnerability (top). Higher adaptive capacity, on the other hand, serves to mitigate the consequences of exposure and sensitivity, which in turn helps to minimize vulnerability (bottom) [53].
To reiterate, social, economic, and COVID-19 exposure were supposed to be the most influential components of adaptive capacity, sensitivity, and exposure, respectively in this study. Furthermore, as a result of the COVID-19 outbreak, the economic options of the households were severely constrained. The outcomes of this research may provide guidance to decision -makers and other relevant agencies on where to focus policy implementation efforts in the next years so that households become less socially and economically vulnerable to COVID-19 outbreaks and associated hazards and disasters.

LIMITATIONS AND FUTURE DIRECTIONS
To begin with, the survey respondents used the internet to participate in the study, indicating that their socioeconomic status is greater than the general population. As a result, the findings' generalizability was hindered. Second, a typical weakness was the study's cross-sectional design. Thus, determining the type of influence was challenging, and we were restricted from drawing causal conclusions from our findings. Then, the study sample was small which did not represent the whole situation of the country. Furthermore, this analysis was based on self-reported responses regarding experiences with the COVID-19 outbreak that could not be supported by qualified data enumerators or experts. Longitudinal, face-to-face survey, and district-specific further research with a larger and dynamic sample in consideration of socioeconomic vulnerability issues among the same population are therefore strongly suggested.

CONCLUSION
In response to the COVID-19 outbreaks in Bangladesh, we performed a cross-sectional survey to investigate and measure the socioeconomic vulnerability status of Bangladeshi households. This study aimed to develop the socioeconomic vulnerability index for Bangladesh. This index (SeVI) also measured the component-based and dimension-based socioeconomic vulnerability. The findings of our study indicated that the SeVI was a manageable and viable technique that captured the vulnerability situation of Bangladeshi households. Overall, the socioeconomic vulnerability is prevalent among the households of Bangladesh due to the COVID-19 outbreak. The economic possible options for the households were greatly limited by the consequence of the COVID-19 pandemic. These phenomena were illustrated in the SeVI of this research. The index scores of SeVI indicators demonstrated which components were more responsible for developing the vulnerability. As a whole, the findings of this study may instruct the policy-makers and corresponding authorities where to place emphasis in policy implementation so that households become socially and economically less susceptible to COVID-19 outbreaks and related hazards and disastrous events in upcoming years. Also, SeVI will improve crisis response interventions by improving understanding of catastrophe consequences at the household level [27,28].