Early View
ARTICLE
Open Access

Effects of disaster education on children's risk perception and preparedness: A quasi-experimental longitudinal study

Ayse Yildiz

Corresponding Author

Ayse Yildiz

School of Business, Institute for Environmental Futures, University of Leicester, Leicester, UK

Correspondence

Ayse Yildiz, School of Business, Institute for Environmental Futures, University of Leicester, Brookfield, London Road, Leicester LE2 7RH, UK.

Email: [email protected]

Search for more papers by this author
Julie Dickinson

Julie Dickinson

Department of Organisational Psychology, Birkbeck College University of London, London, UK

Search for more papers by this author
Jacqueline Priego-Hernández

Jacqueline Priego-Hernández

School of Education and Sociology, University of Portsmouth, Portsmouth, UK

Search for more papers by this author
Richard Teeuw

Richard Teeuw

School of the Environment, Geography and Geosciences, University of Portsmouth, Portsmouth, UK

Search for more papers by this author
Rajib Shaw

Rajib Shaw

Graduate School of Media Governance, Keio University, Fujisawa, Japan

Search for more papers by this author
First published: 01 November 2023

Abstract

The aim of this study was to examine the effects of a disaster education intervention on children's risk perception and preparedness. It also sought to advance longitudinal studies, during an 18-month time period, of risk perception and preparedness by using a quasi-experimental methodology in child-centred disaster research. This study used a quasi-experimental longitudinal research design to measure the effects of disaster education on children. These effects were examined on children's risk perception and preparedness in the Van and Kocaeli provinces of Türkiye during the 18-month period, with a sample of 720 school children. Experimental and control groups were randomly allocated, controlling for age, school grade and school enrolment. The disaster education intervention was designed to improve the children's awareness of natural hazards and their knowledge of disaster risk reduction using discussion, visual materials and interactive teaching of emergency management. The results indicate that the disaster education intervention had a positive effect on children's risk perception and perceived importance of preparedness in both study locations. It also helped children to understand the risks and hazards in their living environments. More importantly, this study showed that disaster education enhanced the protective measures taken by children for disasters. This is the first study using the quasi-experimental longitudinal research design to measure the effects of disaster education on children's risk perception and the importance of preparedness. The findings are of relevance for organisations such as government departments and non-government organisations when designing or improving disaster education programmes.

Short Abstract

The aim of this study was to examine the effects of a disaster education intervention on children's risk perception and preparedness during an 18-month time period by using a quasi-experimental longitudinal research design with a sample of 720 school children. This is the first study using the quasi-experimental longitudinal research design to measure the effects of disaster education on children's risk perception and the importance of preparedness. The findings are of relevance for organisations such as government departments and non-government organisations when designing or improving disaster education programmes.

1 INTRODUCTION

There is growing consensus about the positive impact of disaster education in reducing the effects of natural disasters in communities, as acknowledged by researchers (e.g., Parham et al., 2021; Shaw et al., 2011; Shiwaku et al., 2007) and international frameworks and agendas (UNESCO, 2008; UNISDR, 2006, 2015). While disasters are hard to prevent, measures can be taken to mitigate the impacts of disasters and to reduce human and property losses. There is evidence that low awareness of natural hazards and inadequate understanding of risk can have undesirable effects on people's preparedness, response to hazard events, and recovery (Torani et al., 2019). Education is one of the crucial elements of disaster risk reduction initiatives (Shaw et al., 2011), with strategies for increasing awareness being one of the key elements of effective disaster risk reduction education (Aghaei et al., 2018).

Children are the group most affected by disasters (UNISDR, 2011), with approximately 175 million children being likely to be affected by natural disasters every year (Dyregrov et al., 2018). Children are most at risk in disasters because of their level of behavioural and psychological development, their relatively small physical size and their complete or partial reliance on adults (Zahran et al., 2008). However, when supported, empowered and given a chance, children can be better prepared for the effects of disasters (Peek, 2008) and, in some situations, they may be more resilient than adults (Bartlett, 2008). Some research indicates that involving children in the disaster management planning phases not only helps them but also benefits their families and communities in reducing disaster impacts (Haynes & Tanner, 2015; Ronan et al., 2015).

Recent research has examined the relationship between children and disasters, including different aspects associated with children's readiness and preparedness and, more generally, attention to children's role in increasing community preparedness and resilience (e.g., Anderson, 2005; Peek et al., 2018; Santos-Reyes et al., 2017; Yildiz et al., 2020). According to Amri et al. (2018), the growing research around children and disaster studies can be classified into three main knowledge areas: the integration of education and disaster risk reduction, the effects of disasters on children, and children's participation in disaster risk reduction. This article seeks to contribute to the first knowledge area, children's disaster education.

1.1 The effects of disaster education on children

Disaster education is one of the recognisable approaches for preventing and reducing the risk of disaster, especially for children. UNISDR and UNICEF (2011, p. 19) reported that knowledge and disaster education for children aims to ‘… contribute to a drastic shift … towards a more proactive preventative approach to disasters’. They also highlighted that children are the fundamental motive for the initiatives of disaster risk reduction, as the leaders of tomorrow and key agents for change. To extend this, the Hyogo Framework for Action, which is one of the important milestones for implementing disaster risk reduction, demanded urgent initiatives for government, non-government organisations and other stakeholders to employ education, knowledge and innovation to create disaster resilience at all levels (Shaw et al., 2009). Using the Hyogo Framework for Action, some countries (e.g., Türkiye, France, Russia and Georgia) have taken steps to prioritise disaster education for children in the school curriculum (Johnson et al., 2016). The 2015–2030 Sendai Framework for Disaster Risk Reduction (SFDRR) aims to include disaster education in the entire disaster management cycle and has many examples of children and youth engagement in disaster risk reduction initiatives (UNISDR, 2015). For instance, the Republic of Korea has implemented a school safety programme involving teachers and students; Japan has implemented disaster risk reduction education through both curriculum and extracurricular activities in the communities; Canada has a community-based resilience planning programme in which indigenous youth, together with elders and adults, take part in risk assessment and emergency preparedness activities that blend traditional knowledge and Western science (UNDRR, 2018).

Some important findings have demonstrated that disaster education programmes with children can foster a variety of benefits (Finnis et al., 2010; Ronan et al., 2010; Shaw et al., 2004; Yildiz et al., 2022). A series of quasi-experimental studies on disaster education showed that children who received disaster education displayed a better understanding of risk and hazards, improved household preparedness, and reduced hazard-related fears (Finnis et al., 2010; Ronan et al., 2010, 2012). Through a longitudinal study in five prefectures of Japan, Shaw et al. (2004) demonstrated that while school education helps in enhancing knowledge and perception, community education and family education enhance preparedness. While this evidence further supports the importance of disaster education, the long-term effects of it are still unclear, and several authors maintain the view that there is a minimal rigorous assessment of the efficacy of these programmes in producing intended learning and behavioural results (Johnson et al., 2014; Ronan & Towers, 2014; Toyoda et al., 2021). In a case study of school-based disaster education in 30 countries, Selby and Kagawa (2012, p. 35) concluded that: ‘assessment of student learning is the least considered and least developed element of disaster risk reduction education’. In addition, Johnson et al. (2014) indicated that measurable programme outcomes that directly connect children's learning to improvements in disaster preparedness is one of the areas that need further research.

1.2 Aim and objectives

The aim of this study was to examine the effects of a disaster education intervention on children's risk perception and preparedness. It also sought to advance longitudinal studies of risk perception and preparedness by using a quasi-experimental methodology in child-centred disaster research during the 18-month time period. The disaster education condition used here contained elements of emergency management materials and visual materials adapted from Ratiani et al. (2011), UNICEF (2012) and AFAD (2018b) (see details in Appendix 1). The hypothesis was that the disaster education condition would have a positive effect on children's risk perception and preparedness. We acknowledge the positive effects of disaster education documented in former studies (e.g., Shaw et al., 2004; Shiwaku et al., 2016) and contribute to this body of evidence by applying a quasi-experimental longitudinal design in a different location.

2 METHODS

2.1 Study area: Türkiye

Türkiye is exposed to a wide range of natural hazards due to its geological setting and its climate, notably earthquakes, floods, landslides, storms and wildfires. These disasters generally cause huge disruptions to the functionality of the affected communities, local facilities and government operations. For instance, the Kocaeli and Istanbul areas were severely damaged by the 7.4 Mw Marmara earthquake in 1999, which resulted in 17,000 fatalities and 43,953 serious injuries; 120,000 houses were damaged beyond repair, and 300,000 people were left homeless (Ozmen, 2000). Large-scale disasters occur every 7–8 years in Türkiye (Okay, 2005), and the disaster risk is increasing, with key drivers being population growth and climate change (Balzter et al., 2023; INFORM, 2020).

In the demographic makeup of Türkiye, around 27.5% of the total population consists of children who are between the ages of 0 and 17 years old (Turkish Statistical Institute, 2020): they are disproportionately affected by disasters relative to adults. A key impact is on health, with disasters exacerbating child mortality rates (Ersoy & Kocak, 2016). Another major impact is due to reduced school education because of damage to school buildings and related infrastructures. For instance, after the 1999 Marmara earthquake, 820 of 1651 schools in Istanbul were severely damaged, with many schools closed for up to four months (Yüzügüllü & Erdik, 2004).

The two study locations for this research were selected for two main reasons: first, they are both located on seismically active fault lines, and second, they have varying degrees of socioeconomic development. Both cities have experienced devastating earthquakes in the past 25 years: 1999 in Marmara (Kocaeli) and 2011 in Van (AFAD, 2018a). According to the SEGE (2011) socio-economic development ranking, Kocaeli is ranked fourth, and Van is ranked 75th among the 81 provinces of Türkiye.

2.2 Overview of the research design

The current research was designed to examine the effect of disaster education on children's risk perception and preparedness in the cities of Kocaeli and Van. Children's perceptions of risk and preparedness were measured for the 18-month period, along with their hazard adjustments for disasters before and after the disaster education intervention. The methodology had a quasi-experimental design (Cook & Campbell, 1979). The data were collected from the control and experimental groups in four time periods: (1) April–May 2018, (2) October–November 2018, (3) April–May 2019, and (4) October–November 2019. The experimental group intervention took place in October–November 2018 (Figure 1).

Details are in the caption following the image
Flow diagram of the study.

2.3 Procedure

The institutional ethical review was obtained before conducting research (University of Portsmouth Ethical Review reference number SFEC 2018-028). In the first data collection point, during April 2018, all participants were school children selected from grades 5 and 6 (11–14 years old). The age bracket of 11–14 years was selected in line with the ethics policies of the Turkish Ministry of National Education and the University of Portsmouth (Yildiz et al., 2020).

The approach to selecting samples was driven firstly by the authors and secondly by the coordination between the Turkish Ministry of National Education and the school authorities. The survey was permitted to be conducted in six out of 24 government-run schools in Golcuk (Kocaeli) and five out of 56 in Ipekyolu (Van) by the Turkish Ministry of National Education. Each school was chosen for participation based on class availability. The school administrators chose the individual teaching classes where the survey was conducted. After obtaining permission to conduct the school survey, the families of the selected children were sent the consent form and the information letter through their child's school bag. These letters explained the purpose and the nature of the survey, and requested their signed permission for their child to participate, as well as to take photographs with the full rights to publish these images. Prior to conducting this survey, children were informed about the purpose of the study and provided information on their right to choose whether or not to take part in the research. Participating children were encouraged to answer the questions and to ask for clarity if there was anything they found challenging.

Questionnaire data were collected from 815 children in the cities of Van (n = 387) and Kocaeli (n = 428). The selected sample size accords with Krejcie and Morgan's (1970) recommendation that a sample size of 384 is appropriate for populations that exceed 1,000,000. The Turkish Statistical Institute estimated the populations of Golcuk (Kocaeli) at 162,584 and Ipekyolu (Van) at 312,244 as of 2019. The total number of students in grades 5 and 6 aged 11–14 in the 11 schools selected in 2018 was as follows: 1740 in Golcuk (Kocaeli) and 2398 in Ipekyolu (Van).

The study recruited 815 participants, who were divided into two groups as similar as possible in both cities. Both groups were from the same ages, grades and schools. Experimental or control groups were randomly allocated by flipping a coin by the lead author. The disaster education intervention for the experimental group was delivered by the lead author with the observation of school teachers. The lead author conducted two pre-tests prior to the intervention, first in April–May 2018 and second in October–November 2018, in both the experimental and control groups. One week after the experimental group received the disaster education intervention in October–November 2018, the same questionnaires were applied to the same experimental group to measure the short-term effects of the intervention. Following that, six-month and 12-month post-intervention assessments were conducted with both groups. Figure 1 presents the flow diagram of the study. A total of 66 children from Kocaeli and 29 children from Van were not able to complete all of the surveys because they were not available on the day of data collection or had changed their school or class. Their previous responses were excluded from the survey.

Before the disaster education intervention in October–November 2018, children were asked about their previous exposure to disaster education (Yildiz et al., 2020). Forty-three per cent of the surveyed children in Van and 51.3% of the surveyed children in Kocaeli reported that they participated in disaster education in their school, while 2.3% of the children in Van and 5.2% of the children in Kocaeli reported that they participated in disaster education outside school. It is also important to note that children in Kocaeli experienced flooding between T1 and T2 time points in May 2018 (Yildiz et al., 2021), and there was not any other reported high-scale emergency situation or disaster in both study locations.

2.4 Intervention: disaster education

The disaster education intervention rolled out aimed to improve children's awareness of natural hazards in their living environment and disseminate disaster risk reduction information. The educational content was adapted from Ratiani et al. (2011), UNICEF (2012), and AFAD (2018b) and edited by the research team with the main difference of using more discussion, visual materials and interactive teaching of emergency management. Details of the disaster education programme are provided in Appendix 1. Before the implementation of the intervention, all the teaching materials were revised and evaluated by the research team and the school teachers to increase the effectiveness of this research. First, we created all the materials, which were then reviewed by two experts in school curriculum design and development to ensure that the materials met educational standards and were effective from a pedagogical perspective. After creating the initial materials, Turkish school teachers' feedback was gained for clarity and suitability for the participants' grade level. Based on all the collected feedback, we have made minor revisions, for instance, making the content simpler and avoiding technical terminology. In April 2018, a pilot study was conducted in two randomly selected classes of school children, in both Kocaeli and Van, with samples of 38 children and 28 children, respectively. It was also checked by the school authorities, with the research being well received by the school authorities and with positive comments from the parents of participating children.

The disaster education session was delivered by the Turkish-speaking first researcher with the observation of the principal teachers. The session lasted between 60 and 90 min, depending on the children's questions and the size of the classes, and used a PowerPoint presentation with video clips and pictures from previous earthquakes, floods, landslides, storms, wildfires with a local topographic map. In total, children were divided into two groups, control and experimental, in both study locations, Kocaeli and Van (see Figure 1 and Section 3.2 for the procedure). Figure 2 shows the infrastructure available in the participating schools.

Details are in the caption following the image
Photos of children during the disaster education intervention in Kocaeli (a, b) and Van (c, d).

2.5 Measures

This research used the Pictorial Representation of Illness & Self-Measure (PRISM) (Buchi et al., 1998; Sensky & Buchi, 2016; Yildiz et al., 2020, 2021) and close-ended questionnaires to measure risk perception, the importance of preparedness, and hazard adjustments for disaster preparedness, following the methodology of Yildiz et al. (2020, 2021). PRISM is considered an innovative technique to understand children's risk perception and the importance of preparedness. A paper and pencil version of PRISM was used, and the children were shown an A4 (210 × 297 cm) sheet of white paper with a fixed circle in the corner of the paper. Respondents were asked to indicate their perceptions of risk and preparedness for various hazards by indicating their proximity to ‘self’ on the sheet. The assumption behind the methodology of the tool is that the closer participating children locate their response to the circle representing the self, the higher they perceive the risk or need for preparedness. This technique, which has good test–retest reliability (Yildiz et al., 2020, 2021), was employed because it enables a wide range of responses, potentially providing a better understanding of children's risk perception and the importance of preparedness.

2.5.1 Risk perception

Using PRISM to measure their risk perception, children were asked, ‘Where would you put the (mentioned) hazard to reflect its threats to your life at the moment?’. They were then asked to place the ‘hazard’ on the PRISM template. The range of scores on the PRISM template was from 0 to 27 cm. The closer the hazard was placed to the self, the higher the perceived risk.

2.5.2 Importance of preparedness

Using PRISM to examine the importance of preparedness in their life, children were asked, ‘Where would you like to put “hazard” to reflect its importance of preparedness?’. They then were asked to place the ‘hazard’ on the PRISM template to show the importance of preparedness for the mentioned hazard. The range of scores on the PRISM template was from 0 to 27 cm. The closer the hazard was placed to ‘self’, the higher the perceived need for preparedness for a given hazard.

2.5.3 Hazard adjustment

Children's hazard adjustments for disasters, before and after the disaster education intervention, were measured using close-ended questionnaires (yes, not sure, no). The first questionnaire data for hazard adjustments were collected in October–November 2018 in the pre-intervention time period, with the second dataset collected one year after the disaster education intervention in October–November 2019.

2.6 Data analysis

We used SPSS, descriptive statistics, and two sample t tests to compare baseline variables between control and experimental groups. Repeated measures analysis of variance was used to analyse the changes in risk perception and the importance of preparedness over time to test the effectiveness of the disaster education intervention. A Wilcoxon signed rank test was performed to analyse the differences between measurements. A p value of less than .05 was considered significant. To measure the changes in the hazard adjustments for disasters, paired sample t tests were used.

3 RESULTS

3.1 Comparison of study variables between groups at baseline

The study variables (risk perceptions and importance of preparedness) were compared between the experimental and control groups before the disaster education intervention by two-sample t-tests for the data collected in April 2018. Results yielded no statistically significant difference between the groups for both cities.

3.2 Children's changing risk perception

Children's risk perception for earthquake, flood, landslide, storm and wildfire hazards in their living area was measured during the 18-month time period using the PRISM tool. Figure 3 shows the mean scores for the changing risk perception of the children during the 18-month study period. The closer their scores to zero, the more they perceived the mentioned hazard to be a risk.

Details are in the caption following the image
Mean scores of changing risk perception of children over the 18-month time period for earthquake, flood, landslide, storm and wildfire.

Repeated measures analysis of variance were carried out to examine whether disaster education had an effect on the children's risk perception. Table 1 shows the changes in disaster risk perception in relation to time and group (experimental vs. control). Lower mean scores indicate higher perceptions of risk (see Section 2.5). Repeated measures analysis of variance revealed that risk perception was higher in the experimental groups in Van for landslide, storm and wildfire risk and in Kocaeli for wildfire risk perception. Over time, there was a statistically significant difference in both Van and Kocaeli for the earthquake and landslide risk perception. The group time interaction was significant only for flood risk perception in Kocaeli. Although perceptions of flood risk increased in the experimental group after the disaster education intervention, a similar increase was found in the control group. Therefore, no evidence was found for the effect of disaster education on perceptions of flood risk.

TABLE 1. Changes in disaster risk perception over time for the two study locations.
Variables 1 month (T1) 6 months (T2) 12 months (T3) 18 months (T4) Group
M (SD) M (SD) M (SD) M (SD) F (p)
Van
Earthquake risk perception
Experimental 6.54 (4.97) 6.88 (5.73) 5.90 (3.29) 5.80 (3.52) 1.44 (.23) 4.78 (.00) 0.49 (.67)
Control 6.38 (5.21) 7.41 (5.83) 6.42 (4.04) 6.21 (3.16)
Flood risk perception
Experimental 9.21 (6.60) 9.56 (6.86) 8.20 (4.80) 8.85 (6.00) 3.00 (.08) 0.13 (.06) 4.61 (.13)
Control 9.66 (6.77) 10.07 (6.79) 9.15 (5.24) 9.50 (5.84)
Landslide risk perception
Experimental 10.03 (6.11) 10.59 (6.32) 9.55 (5.60) 9.13 (6.15) 4.21 (.04) 3.15 (.03) 0.20 (.89)
Control 11.04 (6.24) 10.93 (6.51) 10.28 (6.07) 9.93 (6.35)
Storm risk perception
Experimental 10.36 (6.73) 9.98 (5.87) 9.45 (5.81) 9.43 (5.87) 6.49 (.01) 1.24 (.29) 0.44 (.72)
Control 10.90 (6.84) 10.01 (6.28) 10.05 (6.47) 10.56 (6.01)
Wildfire risk perception
Experimental 10.79 (6.32) 10.32 (6.26) 9.96 (6.51) 9.33 (6.23) 6.49 (.01) 1.44 (.23) 0.75 (.52)
Control 10.93 (6.25) 11.39 (6.54) 10.61 (6.16) 10.81 (6.13)
Kocaeli
Earthquake risk perception
Experimental 6.35 (5.22) 5.13 (3.04) 5.18 (3.13) 5.39 (3.04) 2.12 (.15) 8.60 (.00) 0.39 (.72)
Control 6.65 (5.99) 5.18 (3.69) 5.43 (3.10) 6.06 (3.24)
Flood risk perception
Experimental 9.33 (6.12) 8.70 (5.07) 7.71 (4.46) 7.70 (4.75) 0.71 (.40) 1.91 (.13) 4.80 (.00)
Control 8.23 (5.70) 7.33 (4.47) 8.37 (4.98)* 8.53 (5.19)*
Landslide risk perception
Experimental 10.85 (5.77) 10.35 (6.31) 9.02 (5.70) 8.50 (6.15)* 0.00 (.96) 4.71 (.00) 2.29 (.08)
Control 10.22 (5.45) 9.43 (5.32) 9.66 (6.39) 9.49 (6.52)
Storm risk perception
Experimental 10.08 (6.58) 10.21 (6.45) 9.68 (5.47) 9.87 (6.25) 1.57 (.21) 0.09 (.97) 0.78 (.51)
Control 10.16 (6.57) 10.16 (6.18) 10.90 (5.76) 10.26 (6.21)
Wildfire risk perception
Experimental 10.37 (6.49) 9.64 (6.22) 9.55 (5.81) 9.33 (5.79) 8.11 (.00) 0.14 (.94) 2.03 (.11)
Control 9.96 (6.38) 10.57 (6.03) 11.12 (5.45) 10.81 (6.17)
  • Note: *indicates that differences from the ‘6 months’ were significant at p < .05 by Bonferroni test. Group: test of between-subject effects. Time and time × group: test of within-subject effects using Greenhouse-Geisser.

Even though the results show no significant effect of the disaster education intervention on children's risk perception, the mean scores of the post-interventions time points (T3 and T4) were lower than pre-intervention time points (T1 and T2) in the experimental groups in both studied cities. Furthermore, one-week post intervention scores (DE) reached the lowest point in each of the studied cities for earthquake, flood and wildfire risks (see Figure 3).

3.3 Children's changing perception of preparedness importance

Children's changing perception of preparedness importance for earthquakes, floods, landslides, storms and wildfires was measured during the 18-month time period. The closer they placed their score to zero, the more important it was for them to be prepared for those hazards. Figure 4 shows the changes in the importance of preparedness during the 18-month period for earthquakes, floods, landslides, storms and wildfires.

Details are in the caption following the image
Mean scores of children's changing perceptions of preparedness importance over the 18-month time period for earthquake, flood, landslide, storm and wildfire.

Again, repeated measures analysis of variance were carried out to find out whether the disaster education had an effect on the children's perceptions of importance of preparedness. Table 2 shows the changes in the importance of disaster preparedness over time. Repeated measures analysis of variance revealed significantly different changes between the experimental and control groups in the importance of preparedness for earthquakes, landslides, storms and wildfires in Van; and in Kocaeli for the importance of preparedness for landslides and wildfires. During the 18-month time period of the study, there was a statistically significant difference in Van for the importance of earthquake and flood preparedness; and in Kocaeli for the importance of earthquakes, floods, landslides and wildfires (see Table 2). There was a significant group and time interaction in the importance of earthquake, flood and wildfire preparedness in both cities, which mainly reflected the effects of the disaster education on the importance of preparation for the experimental group. Interestingly, while the importance of flood preparedness in Kocaeli did not change significantly in the experimental group after the disaster education intervention, the control group thought it was less important to be prepared for flood events over time.

TABLE 2. Changes in perceptions of preparedness importance over time for the two study locations.
Variables 1 month (T1) 6 months (T2) 12 months (T3) 18 months (T4) Group
M (SD) M (SD) M (SD) M (SD) F (p)
Van
Importance of preparedness for earthquake
Experimental 8.53 (5.21) 8.83 (6.36) 5.23 (3.45)* 6.33 (3.47)* 9.57 (.00) 12.87 (.00) 5.10 (.00)
Control 8.34 (5.13) 8.71 (6.33) 7.99 (4.94) 7.81 (5.18)
Importance of preparedness for flood
Experimental 9.20 (6.35) 9.65 (6.36) 7.08 (4.64)* 7.93 (4.38) 2.96 (.09) 3.70 (.02) 5.93 (.00)
Control 8.64 (6.49) 9.24 (6.39) 9.21 (6.01) 9.47 (5.96)
Importance of preparedness for landslide
Experimental 9.01 (6.30) 9.63 (6.29) 7.93 (4.47) 7.98 (5.69) 5.52 (.02) 1.82 (.16) 2.73 (.06)
Control 9.22 (6.59) 9.61 (6.78) 9.40 (5.36) 9.98 (6.75)
Importance of preparedness for storm
Experimental 9.93 (6.55) 9.79 (6.72) 8.49 (5.76) 8.40 (5.63) 15.28 (.00) 0.27 (.85) 2.67 (.04)
Control 10.08 (6.63) 10.03 (6.74) 10.51 (6.04) 10.93 (6.62)
Importance of preparedness for wildfire
Experimental 10.32 (6.63) 9.76 (5.34) 8.84 (6.07) 9.38 (5.56) 15.93 (.00) 0.54 (.65) 3.16 (.03)
Control 10.32 (6.80) 10.40 (5.33) 11.10 (6.13) 11.61 (6.61)
Kocaeli
Importance of preparedness for earthquake
Experimental 7.39 (4.54) 8.27 (7.73) 5.82 (3.43)* 5.98 (3.76)* 0.71 (.40) 4.04 (.01) 6.20 (.01)
Control 7.04 (4.54) 6.94 (5.11) 7.08 (5.07) 7.35 (5.34)
Importance of preparedness for flood
Experimental 10.05 (6.35)* 7.91 (5.48) 7.18 (4.24) 7.59 (5.01) 0.91 (.34) 8.85 (.00) 3.66 (.00)
Control 9.04 (6.09) 7.83 (5.20) 8.46 (5.06)* 8.59 (5.59)*
Importance of preparedness for landslide
Experimental 9.77 (5.94) 9.23 (6.20) 7.57 (5.93)* 7.92 (5.19)* 6.60 (.01) 4.49 (.00) 1.69 (.17)
Control 10.02 (6.83) 9.27 (6.08) 9.34 (5.98) 9.18 (6.02)
Importance of preparedness for storm
Experimental 9.87 (5.68) 10.42 (6.20) 8.75 (6.38)* 9.52 (6.06) 0.77 (.38) 2.25 (.08) 0.60 (.62)
Control 10.14 (7.26) 10.01 (5.98) 9.42 (6.27) 10.20 (6.42)
Importance of preparedness for wildfire
Experimental 10.47 (6.14) 10.43 (5.80) 8.27 (5.59)* 9.27 (6.10) 4.68 (.03) 3.11 (.03) 2.93 (.03)
Control 10.22 (7.04) 10.24 (5.60) 10.02 (6.29) 10.89 (6.76)
  • Note: *indicates that differences from the ‘6 months’ were significant at p < .05 by Bonferroni test. Group: test of between-subject effects. Time and time × group: test of within-subject effects using Greenhouse-Geisser.

In addition to this, the mean scores of the post-interventions time points (T3 and T4) were lower than pre-intervention time points (T1 and T2) in the experimental groups in both studied cities, and one-week post intervention scores (DE) were the lowest in each experimental group except wildfire preparedness in Koceli. That means the experimental group considered preparedness for each hazard was more important, compared with the control group, in each of the studied cities (Figure 4). This also indicates that the disaster education intervention had a lasting effect on children in both studied cities.

3.4 Children's changing hazard adjustments over time

Paired sample t tests were used to measure children's hazard adjustments before and after the disaster education intervention. In both study locations, the results had statistically significant differences in the experimental group for practising emergency drills at home, planning where to meet in an emergency, having a first aid kit, having a torch, and having a stockpile of water and food for three days (see Table 3). The disaster education intervention encouraged children to have a radio for emergencies in Van, yet it did not show the same effect in Kocaeli. There were also two significant changes in the control group in Kocaeli, for practising emergency drills at home and having a first aid kit, but the differences were not as strong as those in the experimental group.

TABLE 3. Hazard adjustment of children before and after the intervention.
Variables Van Kocaeli
Control Experiment Control Experiment
I have practised what to do in case of emergency at home t (178) = −0.87 t (178) = −10.54** t (191) = −2.16* t (169) = −7.34**
I have planned where to meet in emergency t (178) = 1.87 t (178) = −7.96* t (191) = 0.57 t (169) = −2.97**
I have a first aid kit. t (178) = 0.12 t (178) = −6.67** t (191) = −2.26* t (169) = −2.91**
I have a radio with a spare battery t (178) = −0.49 t (178) = −8.22** t (191) = −1.04 t (169) = −1.71
I have a torch t (178) = 1.17 t (178) = −6.71** t (191) = −1.59 t (169) = −3.76**
I have a stockpile of water and food for three days t (178) = −0.55 t (178) = −8.95** t (191) = −0.97 t (169) = −3.10**
I have a fire extinguisher t (178) = −0.63 t (178) = −1.87 t (191) = 0.33 t (169) = −1.30
  • Note: *p < .05; **p < .001. 1 = no, 2 = not sure, 3 = yes.

4 DISCUSSION

4.1 Children's changing risk perception

In this research, we looked at how children perceive natural hazards (earthquakes, floods, landslides, storms and wildfires) to threaten their lives, before and after disaster education intervention. The findings of this study indicate that the intervention significantly changed the earthquake and landslide risk perceptions of children in both cities. That is, children who received disaster education tended to perceive earthquake and landslide hazards as more of a threat to themselves than children who did not receive the intervention (see Table 1). A possible explanation for this is that the visual materials and interactive earthquake and landslide demonstrations might be more helpful for children to visualise their local risks and hazards than standard educational techniques. The lead author observed that when children were shown pictures of landslides and earthquakes in their living environment, their reaction was more emotive or curious than those for other hazards. It is also important to highlight that even though there were no statistically significant changes for all hazards over time, the descriptive statistics indicated that after the disaster education, children perceived the hazards to be more threatening in their lives, relative to children who did not receive disaster education (Figure 3).

In general, the control group scores remained relatively stable or showed a slight decrease (increase in risk perception), which might reflect a measurement effect; that is, the questionnaires may have caused the children to become more aware of hazards. This study's findings are consistent with Ronan and Johnston (2001, 2003), Ronan et al. (2010), and Finnis et al. (2010) who also reported that children who have been exposed to a disaster education programme showed a better understanding of risk and hazards and had more accurate risk perception. Thus, by deploying a quasi-experimental longitudinal design, this study contributes to the growing body of evidence about the positive effects of disaster education on children's risk perception.

Dennis (2005) suggested that young people need to be concerned about environmental hazards threatening a sustainable future. The findings of our research support the premise that disaster education initiatives can help children to be more engaged with environmental issues. In addition, the learning that occurs in social settings such as schools can influence an individual's risk perception and how they manage or lower their risk (Feng et al., 2017). Thus, it might be helpful to incorporate the findings of this study not only in formal education but also in informal educational initiatives for community resilience. This was also highlighted by Shaw et al. (2004): the more school disaster education programmes can be linked to wider community activities, the more advantages there may be for children, families, and society at large. This has a longer-term implication, in that disaster education provided to children has been shown to have long-lasting effects on their risk perception (Shaw et al., 2004).

4.2 Children's changing perception of preparedness importance

This research also investigated the effects of disaster education on children's perceived importance of preparedness. In general, findings indicate that the post-disaster education scores were lower than the pre-test scores. That is, after the disaster education intervention, children expressed that it was more important to be prepared for the hazards. It is clear from the results that the disaster education intervention was effective in increasing children's perceived importance of preparedness for disasters.

Our results are noteworthy because the increased level of perceptions of preparedness importance after the disaster education intervention can prompt children to reduce the risks in their living environment and potentially minimise the damage caused by disasters. Perceptions are the key element to encourage individuals to take action to reduce, avoid, adapt to, or even ignore risks (Wachinger et al., 2013). Our study provides a link between disaster education and the perception of preparedness importance; such findings can be an agent for building more disaster-resilient societies.

Another notable feature concerns the decrease in the mean scores for the importance of flood preparedness in Kocaeli, between the T1 and T2 time points (Figure 4), which indicates that those children thought it was more important for them to be prepared for floods—although their peers in Van responded differently. The reason behind this might be due to the fact that only the Kocaeli children experienced flooding between the T1 and T2 time points, with flooding during May 2018 (Yildiz et al., 2021).

4.3 Children's changing hazard adjustments

Our findings indicate that children who received disaster education tend to adopt more hazard adjustments for disaster preparedness than their counterparts in the control group (Table 3). Examples of how children took action after the intervention include practising what to do in case of emergency, having an emergency meeting point, and having a first aid kit; all of these measures' mean scores were higher after the disaster education intervention. These results clearly indicate that disaster education was effective in encouraging the children to adopt hazard adjustments. These findings are in line with the research of Ronan et al. (2001, 2010) and Finnis et al. (2004), which suggest that disaster education initiatives could lead children to be more ready and resilient for future disasters.

It is interesting that children's response to having a fire extinguisher did not change significantly before and after the disaster education intervention in both cities. Of the possible explanations for this, children might not have enough influence to encourage their families to consider access to a fire extinguisher and it might be regarded as a relatively expensive purchase for low-income families. Having access to a fire extinguisher is very important for saving lives during a fire emergency, and hence it is essential to involve children's families in considering this fact through disaster education programmes as a part of the disaster management cycle.

Another finding was that children in Kocaeli did not choose to have a radio as a hazard adjustment for disasters. This might be because Kocaeli children are more interested in using electronic devices, such as smartphones and computers, rather than using radios. However, that was not the case in Van, where the children preferred radios and where the lead author observed that children had few opportunities to use electronic devices relative to children in Kocaeli.

4.4 Limitations

Although these research findings have shown significant results for most study variables between control and experimental groups, there are some limitations. First, there might be some knowledge sharing after the disaster education intervention between the control and experimental groups, which may have impacted the control groups' perceptions. Even though the control and experimental groups are from different classes, they might have had conversations in their social times about what they have learned. Secondly, children who reported that they participated in disaster education prior to the disaster education intervention were not removed from this research. While we tried to observe changes before and after the disaster intervention, we acknowledge that this, a natural setting, is different from a laboratory setting. Children might get the information in terms of disaster awareness or disaster risk reduction as a part of their individual teaching classes under their core subjects, such as discussion about environmental problems in Science, Geography or Technology classes; that is, without specific disaster education. Moreover, it was not possible to learn the context, methods, materials and quality of disaster education children stated that they had received. For that reason, results from children who reported that they participated in disaster education were not removed from this research.

4.5 Recommendations

A number of implications can be drawn from the findings. This study illustrates that disaster education can help children to pay more attention to the importance of preparedness for disasters in the long term. More importantly, this study showed that disaster education enhanced the protective measures taken by children for disasters.

Through disaster education intervention, it was observed that visual materials were beneficial for children to understand the risks and hazards in their living environment. For example, when children were shown pictures of landslide hazards, some of the children mentioned that they had not been aware of landslides, but visual material helped them to better understand that hazard. It is thus recommended that during disaster education, visual materials should be used to help children to understand the issues and more effectively raise awareness.

When the lead author asked children to explain in their own words what a ‘hazard’ was, most of the children were confused about the meaning of ‘risk’ and ‘hazard’. Therefore, at the beginning of disaster education interventions, it would be beneficial to explain the meaning of risk and hazard, as well as the differences between those terms. This can be through (i) interactive questions, that is, asking children to identify hazards that they might face in their daily routine and then guiding them on how these hazards might cause harm; (ii) visual materials, for example, illustrating pictures, posters, maps, satellite images or videos of the potential hazards and associated risks; (iii) local examples, that is, providing examples from the children's local area, such as earthquake-related examples and how earthquake hazards might affect their lives; (iv) games, that is, this interactive approach can be used by children to identify the hazards and risks; (v) group activities: these can help children discuss hazards and risks with their peers, allowing them to develop a deeper understanding, especially if implemented through discussion and problem-based learning methods.

Although the participating children had disaster awareness and preparedness information in their school curriculum via Science, Geography or Social Studies classes, the teaching provided is not systematic and is mostly based on earthquake hazards. It is recommended that disaster awareness and preparedness teaching should include different hazards, with explanations about how reducing the risks associated with those hazards can be beneficial for children and their families.

5 CONCLUSION

This study examined the effects of disaster education on children's risk perception and preparedness using the PRISM survey methodology. The main findings of this research are:
  • Children who received disaster education tended to perceive earthquake and landslide hazards as more of a threat to themselves than children who did not receive the intervention.
  • Disaster education intervention increased children's perception of preparedness importance: children thought it was more important to be prepared for the hazards.
  • Disaster education intervention was effective in encouraging the children to adopt hazard adjustments.

The findings support our hypothesis that disaster education intervention has a positive effect on children's risk perception and preparedness. More importantly, this study showed that children who received disaster education tend to adopt more hazard adjustments for disaster preparedness. The findings also encourage the development of interactive disaster education to help children to cope with hazardous events and better prepare for future disasters.

ACKNOWLEDGEMENTS

The authors wish to thank the headteachers of the participating schools and the children who participated in the data collection. We also would like to thank the anonymous reviewers whose comments strengthened the paper.

    APPENDIX 1: 1 THE DISASTER EDUCATION INTERVENTION

    GOALS AND OBJECTIVES

    1. Raising students' awareness of risks, hazards and disasters in their living environment.
    2. Providing disaster risk reduction (DRR) information to students.
    3. Helping to create a sustainable behavioural culture and essential skills among students.

    EARTHQUAKES

    Goals

    • To raise earthquake awareness.
    • To help students to identify dangerous and safe places for earthquakes (i.e., classrooms, schools, homes).
    • To help students to learn and identify the main rules for how to be prepared for earthquakes.

    Activity 1

    The teacher asks students to define an earthquake, describe an earthquake, describe what they have heard about earthquakes, and explain what causes deaths and traumas during earthquakes in their own words. The teacher helps students to understand that it is not the earthquake itself that kills individuals, but rather the subsequent building devastation and objects that are not properly attached or secured falling to the ground.

    Activity 2

    Students are asked by the teacher about their experiences with earthquakes, including how they felt and what they did both during and after the earthquake event. The teacher asks the students to explain what they may see, feel or hear during an earthquake if they have never experienced an earthquake.

    Activity 3

    The teacher encourages the students to define a ‘hazard’ in their own words. The following is one of the examples of what the teacher writes on the blackboard: ‘Which areas of our classroom are the safest and dangerous?’. The teacher then uses brainstorming and asks the class to come up with names of these places. The list of these locations—namely, the classroom's dangerous and safe areas—is written by the teacher on the blackboard. If more information is needed, the teacher provides it. For example, where heavy bookcases, cabinets, and other similar objects can fall, break or injure someone. The teacher asks students to complete the same activity with family members in their homes, looking for both dangerous and safe locations.

    Activity 4

    The teacher asks the students what the most important life-saving rules are for earthquakes. The teacher then summarises and writes the following on the blackboard:
    1. Make sure all moving items in the room are secured; for example, the chandeliers, bookcases, mirrors.
    2. Identify the safe areas in advance, such as tables that are stable, open areas far away from buildings.
    3. Prepare an emergency bag in advance with the essentials, including a first aid kit, water, food supplies and clothes, and the telephone numbers of family members.
    4. Importance of having a family emergency plan.
    5. Stay away from dangerous areas, such as power transmission lines, staircases, balconies, and any items that might fall down.
    6. Importance of knowing how to protect yourself during an earthquake shake: ‘Drop! Cover! and Hold!’.
    7. Importance of knowing the local evacuation routes before a disaster hits.

    Activity 5

    The teacher runs through the ‘drop, cover and hold’ drill, and after calling out ‘earthquake’, the teacher asks the class to repeat her action. Students are required to get under their desks, cover and hold something.

    Activity 6

    The teacher explains what to do following an earthquake:
    1. Get in touch with the nearest authorities.
    2. Stay calm.
    3. There may be some bad people, so do not trust everyone.
    4. If not told differently, avoid drinking tap water.

    Activity 7

    The teacher shows students some photos and earthquake maps of their neighbourhood and points out earthquake fault lines. The teacher then displays some images and videos of earlier earthquakes to make the students more aware that they live in an earthquake-prone area. The teacher encourages students to think about the hazards and risks that might exist in their living environment, how to protect themselves further and whether or not their current level of preparedness is adequate to deal with earthquake events.

    FLOODS

    Goals

    • To raise flood awareness among students.
    • To help students to identify dangerous and safe places for floods.
    • To help students to learn and identify the main rules for how to be prepared for floods.

    Activity 1

    The teacher asks students to define a flood, describe what they have heard about floods, and describe what causes death and injury during floods in their own words. The teacher advises students that if they follow the necessary preparations, a flood should not be a feared event.

    Activity 2

    The teacher asks the class whether students have ever encountered a flood, what their feelings were, and what they did both during and after the flood. The teacher asks the students to explain what they may see, feel or hear during a flood event if they have never experienced a flood.

    Activity 3

    The teacher shows the students a map of their neighbourhood and asks about where and why a flood event can occur, as well as how we can prevent them. The teacher gives them a short time to reflect on it, and the teacher then assists students in understanding the hazardous areas for floods and how to be safe; that is, settlement locations should be far from the river edges.

    Activity 4

    The teacher discusses the important lifesaving actions during and following the flood events:
    1. You should evacuate the flood zone and get to higher ground immediately. You should never attempt to cross the river since it may suddenly get deeper.
    2. You should avoid using electric sources.
    3. Never enter flood waters in an attempt to escape a flood zone.
    4. You should not enter the building without notification of return from authorities and without damage assessment.
    5. Do not use tap water.
    6. Listen to public emergency information.

    LANDSLIDES

    Goals

    • To raise landslide awareness among students.
    • To help students to identify dangerous and safe places for landslides.
    • To help students to learn and identify the main rules for how to be prepared for landslides.

    Activity 1

    The teacher asks students to define landslides, describe what they have heard about landslides, and describe what causes death and injury during landslides in their own words. The teacher advises students that if they follow the necessary preparations, landslides should not be a feared event.

    Activity 2

    The teacher asks the class whether students have ever encountered a landslide, what their feelings were, and what they did both during and after the landslide. The teacher asks the students to explain what they may see, feel or hear during a landslide event if they have never experienced a landslide.

    Activity 3

    The teacher shows the students a map of their neighbourhood and asks about where and why a landslide event can occur, as well as how we can prevent them. The teacher gives them a short time to reflect on it, and the teacher then assists students in understanding the hazardous areas for landslides and how to protect themselves.

    Activity 4

    The teacher discusses the important lifesaving actions for landslides:
    1. Contact the authorities.
    2. If you don't have enough time to exit the building and move away from the landslip area, stay inside.
    3. Create a life triangle and apply ‘drop, cover and hold’.
    4. When a landslip, mudflow or debris flow is approaching, move as far away as you can to the highest level while warning the nearby residents about the approaching landslide.
    5. Listen to public emergency information.
    6. Avoid entering the damaged buildings to get your belongings.

    STORMS

    Goals

    • To raise storm awareness among students.
    • To help students to identify dangerous and safe places for storms.
    • To help students to learn and identify the main rules for how to be prepared for storms.

    Activity 1

    The teacher asks students to define storms, describe what they have heard about storms, and describe what causes death and injury during storms in their own words. The teacher advises students that if they follow the necessary preparations, storms should not be a feared event.

    Activity 2

    The teacher asks the class whether students have ever encountered a storm, what their feelings were, and what they did both during and after the storm. The teacher asks the students to explain what they may see, feel or hear during a storm event if they have never experienced a storm.

    Activity 3

    The teacher discusses the important lifesaving actions for storms:
    1. Stay inside and keep your doors and windows closed.
    2. If you are outside, look for a nearby a covered secure spot.
    3. Avoid travelling unless it is really necessary.
    4. Avoid touching any fallen or shaking telephone or electrical cables.
    5. Listen to public emergency information.

    WILDFIRES

    Goals

    • To raise wildfire awareness among students.
    • To help students to identify dangerous and safe places for wildfires.
    • To help students to learn and identify the main rules for how to be prepared for wildfires.

    Activity 1

    The teacher asks students to define landslides, describe what they have heard about wildfires, and describe what causes death and injury during wildfires in their own words. The teacher advises students that if they follow the necessary preparations, wildfires should not be a feared event.

    Activity 2

    The teacher asks the class whether students have ever encountered a wildfire, what their feelings were, and what they did both during and after the wildfire. The teacher asks the students to explain what they may see, feel or hear during a wildfire event if they have never experienced a wildfire.

    Activity 3

    The teacher discusses the important issues about wildfires:
    1. Wildfires occur as a result of careless and environmental neglect.
    2. A wildfire can be started by burning unwanted grass or stalks in a forest or nearby field, as well as by leaving glass in a forest.
    3. Biodiversity is severely harmed following a wildfire.
    4. The loss of living cover increases the frequency and intensity of associated risks such soil erosion, flooding, debris flows and air pollution.

    DATA AVAILABILITY STATEMENT

    Research data are not shared.