1. Introduction
Typhoons are considered to be extremely devastating natural hazards worldwide. There are seven to eight typhoons on average landing on the coastline areas of southeastern China each year, which makes China one of the countries that were hit most frequently by typhoons [
1,
2]. Between 1980 and 2012, there were nearly 300 typhoon landfalls, and 61% of the landmass of China was affected by typhoons at various degree [
3]. As a major natural disaster in China, typhoons have caused enormous property and human losses [
4,
5]. The influence of typhoon disasters permeated through all aspects of people’s activities who live in the typhoon landing regions. Typhoons do not merely destroy infrastructures and facilities, but also cause serious threats to people’s health and life [
6]. In addition to immediate injuries and deaths caused by the disaster, typhoon severely influences public health [
7,
8].
Studies have reported that the impact of typhoon disaster depends on the intensity and strengths of the typhoon, which is also linked with the precautionary concerns and knowledge of local residents [
9]. Lack of attention on disaster warnings might lead to significant damages and bring risks of typhoon related injuries even facing a low-risk typhoon, which suggests a need to pay attention to typhoon disaster prevention, regardless of the intensity of storms [
10,
11].
Knowledge of the public’s awareness about risks from natural disaster is of vital importance to effective risk communication strategies [
12,
13]. In recent years, many studies put an increasing emphasis on the public’s knowledge and risk perception of natural disaster. Studies in China suggests that risk perception of disaster among residents in typhoon-prone area is higher than other people. In addition, residents’ coping behaviors were positively associated with risk perception and knowledge of preparedness [
9,
14]. In southern Thailand, a study found that people with previous experience have high risk perceptions which were expressed as a heightened fear of a typhoon and a tendency to take threatened disasters seriously [
15]. Residents’ attitudes and risk perception are correlated with knowledge, which have an influence their coping behaviors [
16]. Study also suggests that many people are aware of coming cyclones, but are unaware of their intensity and where they would make landfall, and there is also inadequate knowledge on how to prepare for a cyclone [
17].
Previous studies have indicated that residents’ awareness and coping behaviors of typhoon disaster was related to their socio-demographic characteristics, especially educational status and economic conditions [
18,
19]. Rural residents are more vulnerable to typhoon disaster because they live in a comparatively less-developed and impoverished areas, their major concern is the impact on their livelihood, which may lead to insufficient protective measures for themselves [
20]. Study also shows risk perception and knowledge of typhoon among rural residents is lower than urban ones, which make them vulnerable [
21,
22]. Hence, a better understanding of rural residents’ risk perception and knowledge was conducive for government to develop customized strategy. There is also a need to explore the unknown factors influencing rural residents’ coping behaviors.
It has been reported that among rural residents, the KAP-gap (knowledge–attitudes–practice gap) hampers their ability to takes valid adaptive measures [
23]. People may have some knowledge about the coming disaster but a lower risk perception could influence subsequent likelihood to adopt actions for preparation [
24]. Most people in disaster-prone regions know they should make preparations, but research suggests that very few actually do. In some cases, while residents generally felt prepared for a typhoon disaster, it has been found that there was in fact still insufficient preparation [
25]. Even people who had experienced typhoon many times might not take adequate typhoon preparation actions, which increased population vulnerability [
26]. A study in Iran showed that the knowledge of people regarding disaster preparedness was encouraging, but the translation of knowledge into effective attitudes and appropriate practices was weak [
27]. There is a gap between residents’ perception or knowledge and behaviors. This research project further explores the extent of public perception and knowledge of typhoon disaster, as well as the ability of coping behaviors in a rural area exposed to typhoons.
Among the southeastern regions of China, Zhejiang province is one of the very high-risk provinces most frequently affected by typhoons [
28]. Super Typhoon “Chan-hom” landed in the middle coast of Zhejiang Province in July 2015, affecting 1.9 million people with a direct economic loss of almost 6 billion RMB, forcing 1.06 million people to evacuate [
29]. The objective of this study was to investigate local rural residents’ risk perceptions, attitudes, knowledge, and behaviors related to typhoon. We have also examined the factors influencing participants’ coping behaviors before typhoon. This study will provide useful information for policy makers and other stakeholders to reduce losses from typhoon disasters in rural residents. The study also provides the opportunity to test theories that guide the disaster risk management decision making in typhoon affected regions, and add to the body of knowledge on the subject.
2. Materials and Methods
2.1. Study Design and Setting
A cross-sectional study was conducted in July 2015, 10 days after the typhoon “Chan-hom” landing on the southeast coastal area of Zhejiang. Ninghai, one of the counties affected by “Chan-hom” was selected as study site. Ninghai is a county with multiple typhoons every year and has contingency plans for typhoon disasters. Cluster sampling was used to choose three villages in Ninghai. The population of the three villages ranged from 1400 to 1900 and the maximum distance between these villages and coastlines is no more than 10 km. We calculated the sample size for this study according to the formula of survey sample size calculation [
30], as well as reviewed the relevant articles and the assumption that 60% of people might adopt adaptive behaviors during typhoon. To estimate this proportion with a 95% confidence interval, 522 participants are required. To allow for a 10% non-response and for 1% missing data, the lowest required sample size is 587.
2.2. Study Participants
The targeted study population was local residents of the three villages. The criteria for the participants to be included were: (1) local resident or had lived in the village for at least one year; (2) in the village during the typhoon; and (3) 15 to 74 years old. All eligible residents were invited to participate in the study. Migrant workers, people with mental diseases, and unconscious persons were not included.
2.3. Data Collection
A structured questionnaire was developed after a review of the literature in combination with consultations with local CDC experts who are familiar with the situation of local residents and typhoon. The questionnaire collected the following information: (1) demographic characteristics, such as gender, age, race, educational level, occupation and marital status; (2) source of typhoon warning information and risk perceptions of typhoon (all perception questions had closed-ended responses using different five-point Likert-type answers, ranging from rank 1 to rank 5, with a higher score indicating a higher perception level); (3) attitude towards typhoon disaster prevention; (4) knowledge and individual behaviors to cope with typhoon disasters, including a question about awareness of typhoon warning signal; (5) disaster insurance and government’s measures on evacuation.
Interviewers were trained staff at the local Centers for Disease Control (CDC) and graduate students from Shandong University, who were familiar with field interviews and received training before the survey. A quality control person was assigned to each of the seven groups of interviewers. All information was collected by face-to-face interviews at participants’ homes. Thirty-three people refused to participate in our study. The main reason being the perception of receiving no benefit from the study and/or a waste of their time. A total of 683 participants were interviewed with a response rate of 95.4%. There were 24 (3.5%) questionnaires having too much missing information and errors, which were excluded before being entered into the computers. We included 659 valid questionnaires in our analysis.
2.4. Statistical Analysis
Data were entered into Epidata version 3.1. Statistical analysis was conducted with the Statistical Package for Social Sciences (SPSS) version 21. Descriptive statistics were used to describe respondents’ demographic characteristics as well as their perceptions, attitudes and behaviors. Chi square or Fisher’s exact texts were used to test for the relationships between demographic variables and other categorical variables. The mean and standard deviation (SD) was used to describe the score of risk perception. Factors with significant associations (
p < 0.15) from the univariate analysis were then analyzed by complementary log–log (CLL) regression for the reason that the Yes/No responses of dependent variable were asymmetrically distributed [
31,
32], to further explore how risk perceptions and attitudes may influence residents’ coping behaviors (Adopt or Not Adopt: there are many questions regarding coping measures in our questionnaire. If participants did not have any behavioral change in response to the typhoon, then the individual was defined as not adopting coping behaviors).
2.5. Ethical Considerations
The Ethics Review Committee (ERC) of Public Health at Shandong University has reviewed and approved the proposed use of human subjects in this survey (No. 20120501). Written informed consent was obtained from each participant (where participants are children, their parents) after they were informed that their participation was voluntary and their refusal to participate would have no negative consequences. All data were kept anonymous and confidential throughout the study.
4. Discussion
In this study, we investigated the risk perception, attitude, knowledge, and behaviors related to typhoon among vulnerable residents in Zhejiang province which will be useful in preparing for future catastrophic events. Our results show that respondents had a high-risk perception of property damage, while the risk perception of health damage and life threat caused by typhoon was not enough. Majority of participants did not receive any formal education about typhoon disaster prevention prior. This could be a factor in lack of appropriate actual coping behavior. Although the majority. Although majority of them claimed they paid attention to typhoon disaster prevention and knew how to adopt measures, the actual coping behavior before and during typhoon was inadequate (refer to
Table 4).
In our study, most residents identified television as the main ways to get information before typhoon (83.5%), the result was similar with previous studies in China [
14,
28], Florida [
26], and New Jersey [
34]. TV is a major and convenient source of information among rural residents. A good use of TV to release information by the government plays a major role in helping people to notice the typhoon warnings. As for the most favorable way to get information, the leading three methods were TV media, SMS and house-to-house notification. This result was consistent with a cross-sectional study in Bangladesh, which indicated that mass media and communication between neighbors were important information source [
18]. The young people were prone to choose SMS, while the elder people choose house-to-house notification. Considering the age structure of rural residents, we suggest expanding the door-to-door notification by village committee or local government in typhoon-prone rural areas.
With regard to the risk perception before typhoons, participants in our study showed a high concern on property damage, while an insufficient concern on health damage and life threat regardless of residents' gender, age or educational level. This result was similar with the previous study in New Jersey which indicated that the greatest concerns of residents before the typhoon was possessions and property damage, while during and after the disaster, their major concerns were personal health [
34]. In our study, only about half of participants affirmatively express that typhoon might pose threat to their health (50.7%) or life (42.2%), and less than half of the participants (37.4%) expressed a fear of typhoons. Similarly, according to a survey in 2009, 45% participants did not realize that they were vulnerable to disaster [
35].
Our results showed that there was no difference between residents’ risk perception, coping behaviors and their socio-demographic characteristics which is inconsistent with previous studies [
14,
18]. The main reason could be that we conducted our survey in rural areas only, the majority of respondents were in their middle age or older, and over half of the participants had primary and lower level of education, including the 30.1% that were illiterate. In addition, we investigated three adjacent villages where residents may share similar experience about typhoon.
Residents have paid great attention to typhoon disaster prevention, and most residents claimed that they knew how to prepare for a typhoon event. However, from our findings, their knowledge about adaptation measures was still inadequate. Most participants thought that shutting the doors and windows could reduce losses, but there were a significant proportion of participants who did not think the other measures such as keeping a watchful eye on typhoon warning, preparing emergency articles including foods and drugs, leaving the dangerous areas, keeping away from low-lying areas could work. The awareness rate of some critical issue like typhoon warning signals was low. The majority of participants claimed that they adopted coping behaviors before and during typhoon. However, only a fraction of participants actually responded when specific measures were asked. There was a gap between what they claimed and what they did. In our study, the result manifested that most residents received no typhoon preventive education, and the majority of the respondents expressed that they were willing to take part in the educational course. Previous study also suggested that compared with persons with general preparedness knowledge, persons with new knowledge and updated information were more likely to respond to typhoon effectively [
16]. As a result, we suggest that the government should conduct more educational activities and make updated information available for rural areas through multiple easy-to-follow forms.
Zhejiang is one of the most economically developed coastal provinces, the insurance industry in Zhejiang province has made a great contribution in the insurance indemnity of relief work. Ningbo city had been also developed as a comprehensive insurance pilot area a few years ago [
36,
37]. However, the rate of disaster insurances purchased was low (26.7%), and the main reason that people did not purchase insurances included: (1) Most respondents did not know disaster insurance; (2) Not necessary to purchase; and (3) Financial reasons. In rural areas, typhoon might cause damage to their farmland or fish pond, which are the basic livelihood guarantee of the greatest majority of residents (three-quarters of the residents are farmers and fishermen). Improving publicity and promotion of affordable disaster insurance among rural residents especially farmers is also a pressing matter.
Our study found that residents with more preparation time, a higher risk perception of health damage and life threat were more likely to take measures, so did people who were scared of typhoons. In addition, our study also found that residents with a clear understanding of typhoon warning signals would be more likely to adopt measures than others. Although residents may get information from TV to know the intensity and arrival time of typhoon to take responses, a better understanding of the typhoon warning signals also contributes people to adopt effectively coping measures to some extent, especially when there is blackout caused by a typhoon. While previous studies suggested that the perception of a high risk of injury is important in individual’s evacuation decision making in the face of disaster [
38,
39], this survey found that residents with a higher perception of health damage and life threat would adopt less measures. The survey did not provide data that explains why and, accordingly, further research is necessary.
In general, residents in costal rural areas have a high risk perception of property damage while that of health damage and life threat need to be improved. Most people thought they were aware of how to adopt measures to reduce losses, they were in fact still less knowledgeable and their coping behaviors were inadequate. This indicates an urgent need for appropriate targeted strategies and a necessity to develop educational and other suitable information activities to eliminate the gap. Public anti-disaster capacity building is linked with a strong awareness of disaster prevention. Only when residents have fully realized the influences and harmfulness from typhoons are they able to take the corresponding preventive measures seriously. Our findings will be useful for developing strategies to make adaptations to minimize the health and property loss.
5. Limitations
Our study has some limitations. First, our survey was conducted 10 days after the typhoon, people’s attitude and risk perception may change slightly before and after the landfall of the typhoon, and the difference could not be compared. Typhoon may improve the perception and awareness to some extent. During our investigation, we tried to use objective indicators as many as possible in questionnaires and explicitly asked participants face-to-face about their attitude and perception before typhoon to control this bias. Secondly, there were many young farmers going out for work, and over half of the participants were older than 50, thus the results might not represent the situation in total population. However, migrant workers are also a common phenomenon in rural areas in China, thus we believe that our participants are a representative sample of those who stay in the village and the results can still provide implications of how rural communities might cope with typhoons.