Assessing the effectiveness of a water safety program for junior high school students in Japan1

Abstract Water safety programs to protect children from water-related accidents and increase their awareness of the risk factor crucial to maintaining safety. This study examined the behavior of 48 junior high school students after an intervention program on water safety in a river using the Rohrmann’s Risk Communication Model and the Analysis, Design, Development, and Evaluation (ADDIE) model. A self-administered questionnaire was designed to collect responses about the children’s perception of danger at three-time points: pre-post-post (5 months) after the program’s implementation. We hypothesized that water safety education can increase children’s awareness of appropriate behavior around water and raise their awareness of danger. The evaluation indicators consisted of an increase in the awareness of the water area risk perception, general risk acceptance, and risk-taking behavior. The results suggest that practical water safety program may have a positive effect on the students’ awareness of water safety and help prevent water accidents. This study is unique in that it was conducted in a natural body of water.


Introduction
Globally, hundreds of water-related accidents involving adults and children are recorded each year during the summer months. The number of children affected by water accidents is six times higher than the actual number of deaths (A. E. Peden et al., 2018;Suominen et al., 2002). Indeed, drowning is one of the leading causes of death among children under 15 years old (M. M. Peden & McGee, 2003;Morrongiello et al., 2013;Taneja et al., 2008). As per the data from the Community Safety Bureau of the National Police Agency in Japan (2020) approximately 1,298 of the water accidents reported in 2019 involved 1,538 people, of which 695 drowned or went missing. In the last fiscal year in Japan, 45 people under the age of 18 reportedly died in water accidents, of which 38 (84.4%) were elementary school students or older students. For children younger than the junior high school age, 118 of the water accidents reported involved 190 children, of which 30 drowned or went missing. While fatal drowning accidents often involve infants and toddlers, more than 80% of the children who drowned in water accidents were of elementary school age or older, and the risk of drowning in open freshwater areas was deemed higher among adolescents (Davey et al., 2019). While only approximately 15.8% of children were involved in water accidents, and the number of deaths seems relatively low, it is still important that children and adults understand the risks and are educated about water safety.

Literature review
Typically, preventive efforts to guard against water accidents have either focused on the environmental improvements or behavioral changes (A. E. Morrongiello et al., 2013; R. C. Franklin & Peden, 2017). The environmental improvements include the construction of shallow water areas where children can play safely, the posting of hazard signs advising of any danger zones, and installing fences. The behavioral changes involve education about drowning prevention and first -aid measures if an accident occurred (Wang et al., 2020). While environmental improvements are important and have been moderately successful, they are not foolproof measures against accidents during unguarded moments or reckless acts. Therefore, to reduce unwanted water accidents, behavioral changes may be more effective. Previous studies have highlighted the importance of parental supervision (Moran, 2009(Moran, , 2010Petrass et al., 2011), as many accidents occur when responsible adults are absent (Kemp & Sibert, 1992) or lack of close supervision during children's water activities (Blum & Shield, 2000;Cody et al., 2004;Moran, 2009Moran, , 2010. Therefore, educating parents about the drowning risks and the need for supervision around water could improve parental supervision (Morrongiello et al., 2013;Wang et al., 2020). Because parents sometimes underestimate the need for child safety supervision near water (Moran, 2009;Morrongiello et al., 2013;Petrass et al., 2011), implementing parent-focused swimming programs could aid in improving parental supervisory behavior (Morrongiello et al., 2014). While parental supervision could be effective in preventing water accidents for children under 10, most children over that age are often accompanied by people other than their parents, which means that children's own water safety and swimming skill knowledge is important (Petrass & Blitvich, 2014). Furthermore, if water accidents are the result of disasters triggered by natural hazards, children need to rely on their disaster prevention knowledge (Sandseter & Kennair, 2011). Therefore, children over 10 years must have sufficient water safety knowledge and skills to know how to react in a crisis.
Unfortunately, in Japan, water safety education is considered the responsibility of the parents. Schools only provide physical education programs that are focused on improving swimming skills with the belief that good swimming abilities are effective in preventing children from drowning (Brenner et al., 2009;Rubio et al., 2015;Stallman et al., 2017;Wang et al., 2020). However, recent studies have found that even though older children have better swimming abilities than younger children (Irwin et al., 2009;Moran et al., 2012), they are more likely to be involved in water accidents (Chan et al., 2020). For example, in a Hong Kong study on the relationship between the children's swimming abilities and water accidents, Chan et al. (2020) determined that swimming distance and basic swimming skills were not significant indicators of non-fatal aquatic events. Swimming techniques and treading water were the only basic aquatic survival skills reported to be significantly associated with water accidents (Asher et al., 1995;Petrass & Blitvich, 2014). Therefore, these skills may be more important in water safety education than normal swimming abilities as (Moran, 2019). Linnan et al. (2011) confirmed in a Bangladesh cohort study, where the acquisition of the safety skills and knowledge led to reduction in infant drownings. Wang et al. (2020) uncovered similar trends in China. The majority of caregivers (83.1%) had no knowledge or skills to help a drowning child, and most of the drownings occurred in ponds, canals, rivers, and wells. They were unpreventable if safety measures were not employed. However, Turgut et al. (2016) reported the success of the water education program for 476 school children aged 10 to 14 in Turkey. Using a pre-and post-test approach, the authors concluded that knowledge and training on how to behave during drowning incidents were key lifesaving techniques.
As most children in Japan experience disasters triggered by natural hazards, such as large-scale earthquakes and typhoons, it is proposed that disaster safety programs, such as the aquatic survival skills, should be taught in school to highlight the importance of the water danger awareness. However, these types of programs are not provided in Japanese schools or in other countries, primarily because of the risks associated with the implementation of the disaster prevention programs, including practical water safety education.
Knowledge-based approaches to water accident prevention do not always lead to changes in the children's behavior (Lund & Aarø, 2004). Indeed, the relationship between the accident prevention attitude change and behavior remains weak. Attitude-changing interventions do not necessarily have a significant effect on the behavioral changes (Aarø & Rise, 1996;Verplanken & Orbell, 2022;Wicker, 1969). The attitude change alone does not lead to behavioral change because of the influence of the interrelated processes (Lund & Aarø, 2004;Verplanken & Orbell, 2022); therefore, when conducting disaster prevention education in schools, an experiential approach based on real situations is needed to affect the student's behavioral change.
In Japan, swimming education is often conducted in a clothed manner during physical and health education swimming classes (Ministry of Education, Culture, Sports, Science, and Technology, 2008). Clothed swimming is a learning method popularized by Araki and Sano (1993). They theorized that the method could better save lives in times of a water disaster as students are taught how to stay afloat with their clothes on until the rescue (Inagaki & Kishi, 2012;Inagaki, 2015;Matsui et al., 2016). Several clothed swimming studies, which used physiological indices have been conducted in Japan to examine the effects of clothed swimming on human bodies (Choi et al., 1994;Ebisu et al., 2001;Matsui et al., 2016;Nomura, 1991;Tsubakimoto et al., 1992). Studies have examined the integration of clothed swimming into school physical education programs and as part of the sessions at swimming pools (Inagaki & Kishi, 2014;Kishi & Inagaki, 2018;Nozawa, 2009;Obayashi et al., 1993). Although these studies represent accumulated knowledge on the effects of clothed swimming and teaching practices, most programs focused on the classes conducted in swimming pools rather than in natural bodies of water, such as rivers or lakes.
Water accident prevention research on children tends to focus on infants and toddlers. Even when the research targeted elementary school students and older, the studies were mostly focused on preventive exercises in swimming pools (Hamilton et al., 2019;Kishi & Inagaki, 2018;Taylor et al., 2020). Furthermore, some researchers tend not to consider the influence of warm weather on the drownings. However, the probability of drowning is six times higher when the temperature is 30°C than when it is 15°C (Chauvin et al., 2020). Furthermore, warm weather may cause behavioral changes resulting in recklessness because of the increased water-related activities, and decreased use of personal flotation devices (Fralick et al., 2013). Hence, most water accidents involving elementary school students and older tend to occur during leisure activities in the hot summer months, to investigate these hypotheses, this study examined the efficacy of a water accident prevention program for elementary and junior high school students, conducted in a natural body of water; in this case, a river. In addition to the water accident prevention programmers, children are offered a range of disaster prevention/safety education programs focused on the traffic accidents, earthquakes, and fires (Midtbust et al., 2018;Suarmika et al., 2022), and epidemics such as COVID-19 (Shaw et al., 2021). However, it is difficult to provide specific disaster prevention education programs in schools because of the limited time available within the general safety education.

Research objectives
Faced with these difficulties, this research had two main aims: 1) to formulate and implement a water safety program using a natural body of water that could be practiced at school sites, and 2) to verify the effectiveness of the program by determining its impact on the students' awareness of water accident prevention and general safety. The program featured first-year junior high school students swimming clothed in a river, after which their awareness of water accident prevention, general risk acceptance, and behaviors were assessed using the Rohrmann's Risk Communication Model. In this study, the ADDIE model proposed by Gagnè et al. (2004) was adopted in the development of a water accident prevention program. The ADDIE model is characterized by five stages in its design: Analysis, Design, Development, Implementation, and Evaluation, and based on a problem-solving model. The steps of this study were as follows. Based on the current situation of the water-related accidents and the needs of the society (analysis phase), a strategy to prevent water-related accidents more realistically was implemented in natural waters (design phase), and a safe and effective practical lesson plan was developed in collaboration with specialized researchers, school teachers, and fire department personnel (development phase). In the implementation phase, an educational practice program was implemented for the first-year junior high school students, in which the children themselves got a direct experience in a natural body of water, and in the evaluation phase, the program was evaluated based on the results of a self-administered questionnaire survey focusing on the cognitive changes in the students.

Materials and methods
The present study Rohrmann's (2000) Risk Communication model, proposed a "risk recognitioninformation search-measure implementation" process. The model has been proven to be helpful for practical safety education, in which risk recognition is identified as an indispensable premise for taking measures against danger. We theorized that the model was useful for water safety education as it highlights the specific risks in the water areas. After noticing the difficulties in separating behavior modification from attitude modification, Lund and Aarø (2004) concluded that behavior modification programs that incorporated two-way communication practiced in small groups were the most effective. Therefore, a practical program that incorporated a mimicry of drowning events in a river was instructive.

Overview of the water safety program
The main characteristic of this water safety education program is that it was conducted in a river, a natural body of water, which is considered unusual for the water safety education. To ensure the students' safety, the course was conducted in collaboration with several experts: three junior high school teachers (two classroom teachers and a physical education teacher), seven fire department officers, and three university researchers of different specialties (a physical education specialist, an educational technology specialist, and a developmental psychologist). All the groups were involved in the planning and implementation of the program, which was conducted from 13:00 to 15:10 on a day in July 2022 and included the development of a self-administered questionnaire, which was distributed before and after the experiment.
The target class had 48 first-year junior high school students from the Chubu region of Japan. A special timetable was established for the clothed swimming practice, which was conducted on River A, located a 10-minute walk away from the target school. The river depth was approximately 50-100 cm, and its width was approximately 60 m. Before the commencement of the practical class, the class practitioners, school staff, and fire department officers held two meetings to confirm the program timetable and develop a response system to any possible emergencies. A university teacher with 20 years of experience teaching clothed swimming played a central role as the T1, and the physical education teacher played a major role as the T2 in teaching and assisting the students. The seven fire department officers were stationed in the dangerous areas on the riverbanks to ensure the students safety. The weather on the day of the program implementation was fine; however, as the river was swollen because of the rain the previous day, the river was higher than expected. Since the maximum consideration was the students' safety, the fire department officers were consulted to confirm whether the event should continue. All the participating students were required to wear school-issued jerseys, socks, and waterproof shoes over their swimsuits.

Practical program content
This was a practical program aimed at increasing the students' awareness of the ways to avoid water accidents and prevent their serious consequences, including death or serious injury. Therefore, the program content was developed based on these two considerations. To avoid water accidents, it was important that the children understood and acquired aquatic competence (Leavy et al., 2016;R. Franklin & Scarr, 2014;Taylor et al., 2020). By practicing in real-life river environment, the research team was able to teach the four river danger factors, i.e., water depth, water temperature, riverbed, and flow velocity. The fire department officers taught the practical content, i.e., coping and rescue methods. Therefore, the program had three main foci: (1) "moving and walking on a sandbar," (2) "floating position," and "posture in flowing water" as the countermeasures against water accidents, and (3) "throwback" and "human chain" as water accident rescue methods. The practical program is outlined in Table 1 (also see Figure 1 for illustration of the program execution). As the practice session involved some risks, care was taken to avoid confusing the students in any way. During the first orientation, the following procedure was followed: The program's practitioner (T1) gave the children some safety tips for swimming in clothes, the fire department officers outlined the dangers and safety concerns in rivers, and the physical education teacher (T2) described the program, the evacuation sites, and the emergency procedures. The practice activities in the river were conducted in groups with the class teachers and researchers overseeing the students and encouraging those who were unable or reluctant to enter the river.

Data collection instrument -questionnaire
In addition to the collection of the demographic data, a self-administered questionnaire was redesigned from a survey instrument used for earthquake disaster prevention education by Rajib et al. (2004). It included three scales: Risk awareness for water areas; the scale of risk acceptance; and risk-taking behavior.
(1) Risk awareness of water areas. The category of the risk perceptions in the "whole of Japan," was excluded. Six categories from Inagaki and Kishi (2014) were used to examine the changes in the students' awareness (see Table 2). The breakdown of the six categories was as follows: the possibility of water accidents, the perception of water depth, the perception of water temperature, the perception of the riverbed, the perception of current flow velocity, and the perception of entering the water while clothed. In response to the question, "Do you think you will have an accident?," the respondents were asked to answer on a 5-point Likert scale: "Yes, I think I will," "Yes, I think I probably will," "I cannot say either way," "No, I think I probably won't," and "No, I do not think I will." (2) Scale of Risk Acceptance. The Scale of Risk Acceptance (SRA) developed by Yoshino and Kinoshita (1996) was used in this study. The SRA assessed five main factors, i.e., prudence, recklessness, challenging, safety first, and fate i.e., each of which had three categories (15 items in total). The responses were created using a 4-point Likert scale, ranging from "applicable" (3 points) to "not applicable" (0 points), with the sum of the scores for each factor being the factor's score.
(3) Risk-taking behavior. The Risk Propensity Questionnaire (RPQ) developed by Moriizumi et al. (2010) was also used. The RPQ had four factors and 17 items, i.e., gambling orientation (5 category), situational behavior (6 items), confident behavior (3 items), and safety consideration (3 items), which were used to measure the behavioral choice tendencies toward risk. The responses were create using a 5-point Likert scale ranging from "very applicable" (5 points) to "not applicable at all" (1 point), and the sum of the scores for each factor was used as the factor's score.
The questionnaire was first distributed to the students three days before the practical program. It was then re-issued immediately after the program (post1), and five months after the program was conducted (post2). The Intraclass Correlation Coefficient (ICC) based on the data at the three-time points was calculated as the r test-retest reliability for each of the SRA and RPQ subfactors (see Table 3). As a the analysis, a repeated measures analysis of variance was conducted based on the children's scores collected at the above three-time points to clarify the internal changes in children due to the practice program. The statistics were analyzed using IBM SPSS v25.

Ethical considerations
The ethical approval was granted by the Ethics Review Committee of Gifu Seitoku Gakuen University (No. 202101). The principal of the target junior high school was informed about the research outline, and permission was obtained to conduct it exercise. The principal also served as an observer in the exercise planning stages. The target students and their parents were provided with an oral and written explanation of the outline of the exercise practice as well as educational effects and the voluntary nature of the participation. Regarding the questionnaire surveys, the researcher explained at each distribution stage that the responses were voluntary, and that the participation would not affect grades, and that there was no disadvantage for non-response. The Students were also informed that the results of the survey would be published in a form that would not identify them personally. The students who agreed to the conditions were allowed to participate in the project. All gave informed consent. Three students refused to participate due of health problems.

Water accident prevention awareness
To examine how the practical river program raised the students' awareness of water dangers when they submerged in their clothed state, a repeated measures analysis of the variance was conducted. That consisted of six risk perception items for water areas as the dependent   variable and the survey time as the independent variable (see Table 2). All six items were found to have significant awareness changes over time. The Tukey method was then applied to these results (see Figure 2), which revealed that when compared with the pre-survey (post1), the awareness of the water risk had increased for all items. Except for the awareness of the risk of entering the water when clothed, the awareness of the other five risk items increased immediately after the practical training and was maintained over the 5 months after. There were several possible reasons for that result: (1) the students wore life jackets during the practice; (2) the river was swollen and dangerous due to the rainfall the day before; and (3) the students were desperate to complete the practical training. This was the first time most of the students had worn life jackets, and many were surprised at being able to float on the river. Some students mentioned the life jackets in their reflections after the exercise and said: "The river current was fast and scary, but it felt good to float with the life jacket on." In addition, since this exercise was conducted in a fast-flowing river, the children were desperate to avoid drowning, so we speculated that they may not have been aware of the fact that they were in the river with their clothes on. In their reflections after the practice, some of the students commented, "The flow of the river was so fast that I was surprised, and all I could do was just to move around in the river." Because the practice was conducted in such an unusual environment, even if the students thought about getting clothed and going into the river immediately after the exercise, the conditions in the river, such as the depth and coldness of the water, were so intense that they probably forgot the sensation of being clothed after 5 months. The results suggested that the awareness of the risks related to the exercise that made such a strong impression at first persisted within 5 months.

Impact on general safety awareness
A repeated measures analysis of variance was conducted, with the five SRA factors and the four RPQ factors as the dependent variables and the survey time as the independent variable (Table 3). The SRA "prudence" factor showed a significant change over time; thus, a subsequent Tukey test was conducted (see Figure 3), which found that the "prudence" factor was significantly lower 5 months later than immediately after the exercise. Meanwhile, the RPQ "situational behavior" factor also showed a significant change over time, with the subsequent Tukey test (see Figure 3) showing that the "situational behavior" was significantly higher 5 months later that immediately after the exercise. The "situational behavior" factor included categories such as "riding a bicycle without lights on at night" and "crossing at a red light when walking if no cars are coming"; that was a factor associated with the choice of risky behavior depending on the situation. Both the SRA "prudence" factor and the RPQ "situational behavior" factor indicated that there was a raised risk awareness immediately after the practical training (tendency to choose more cautious behavior and avoid risky behaviors); however, this awareness was noted to have disappeared 5 months after. Furthermore, it was determined that the exercise did not affect the other four SRA factors or the other three RPQ factors. Because the exercise was conducted when the river was swollen due to heavy rain, the fire department officers had instructed the students before the exercise to be careful when entering the river. As the natural conditions changed suddenly, the fire department officers also perceived those conditions as serious and repeated their warning so that students could protect themselves. Therefore, it could be inferred that immediately after the practice, only the "cautiousness" factor and the "situational aggressiveness" factor, which were directly related to the lecture content of the fire department were affected. However, as those decreased significantly within 5 months, careful caution is required to judge the effects of this practical training on those two factors.
Therefore, it can be concluded that practical water safety education in natural bodies of water was effective in raising the students' awareness of the water-related risks. However, this practice had little effect on their general risk awareness and risk-taking tendencies.

Summary
Drowning has been identified as one of the leading causes of death in children (M. M. Peden & McGee, 2003). Water incidents in natural bodies of water have contributed to death or injuries among adolescents, but this area of study is under-researched. In this study, a practical intervention in water was designed and implemented among 48 junior high school students in Japan. The experiment took place in a river, and instructions how to prevent water accidents and increase the students' awareness of water accident prevention were given before the full-clothed immersion into the river. The self-administered questionnaires administered before and after the -experiment further verify the effects of that educational procedure. This study was based on two hypotheses: (1) water accident prevention education increases students' awareness of water area risks, and (2) water accident prevention education affects students' general awareness of danger. As per the results of our analysis, it was determined that the water area risk awareness increased immediately after the exercise in all six categories. In addition, for five items, the water area risk awareness was present even 5 months after the survey. Hypothesis 1 was confirmed as it was found that the riverside training raised the students' awareness of the water area risks. However, only two of the nine factors showed any change over time for general risk acceptance and risk-taking behavior, and it did not appear that there was any effect of the water accident prevention education on the general risk perception, and therefore, Hypothesis 2 was not confirmed.
According to Green and Hart (1998), the children's accident risk aversion and reporting of personal experiences (e.g., trapped in a railroad door) have significant impact on their future behavior. On the other hand, Drupsteen et al.'s (2013), examination of the process from incident to accident prevention, found that risk is not immediately learned from accidents. Furthermore, Lindberg et al. (2010) noted that effective learning from incidents required a follow-up that leads to effective interventions. This practice was designed to simulate an incident for the learners to learn about general risk as well as to change their attitude toward water hazards. Therefore, as claimed by Green and Hart (1998), Hypothesis 1 was confirmed by the change in attitudes toward natural water bodies, but, as pointed out by Lindberg et al. (2010), follow-up steps and actions leading to effective intervention after the practice were inferred. Therefore, the failure to do so was the reason why Hypothesis 2 was not confirmed.

Impact of practical education on high school children
Experimental training was considered effective in enhancing student knowledge of water accident prevention. The training location on a river appeared to have had a lasting effect on the student's awareness of the dangers in natural water areas. This awareness was still present at 5 months later. However, when that exercise was meant as a part of water safety education, it became clear that it could not be generalized to other populations. There were two possible reasons that this exercise did not affect the general risk perceptions of the students. First, the program made a very strong impression on the participants, which led to increase in their awareness of the water area risks; however, this awareness did not transfer to a general awareness of risks and coping behaviors in other areas. The second reason was the disproportion in the time spent on cautioning students about the hazards associated with the natural bodies of water. More time was spent on instructing the students before the program exercise than after the program. Therefore, more time is needed to find out how the course content could be generalized to other similar situations The accident prevention model "Knowledge → Attitude → Practice {behavior} change (KAP model)" was effective (Lund & Aarø, 2004). This is why accident prevention education programs in general tend to focus on knowledge-based educational activities and face-to-face knowledge teaching. However, existing research is often ignored or not fully used when required (Ajzen, 1988;Zimbardo & Leippe, 1991), and knowledge-based attitudes are not associated with behavior (Wicker, 1969). Therefore, even if the knowledge of the danger was known, in many cases, the body does not react (cannot take action) when faced with a dangerous situation. Conversely, once a person experienced a critical situation, even if ignorant, there was often a learning effect; therefore, when faced with a similar situation in the future, their response may be different. Classical psychological theorists posit that experience (behavioral choice depending on the situation) leads to a change in attitude (Bandura, 1969;Festinger, 1957). Therefore, experiencing an actual dangerous situation can enhance the educational effect of accident prevention and safety education, as in this program. Experiential practice can lead to behavioral changes and further enable students to learn how to cope with actual water accidents. The clothed swimming practice was also expected to help students during potential water accidents. For example, by learning to fear unknown natural water areas and the dangers of being submerged while clothed, students could prepare countermeasures, such as not entering rivers when water currents are rapid. It is also important to verify whether school education practices, as implemented, contribute to the prevention of water accidents.

Future issues
This study sought to the students' awareness of the risks associated with water accidents using a practical education program. Various other factors related to water accidents should be considered, such as the parental income and education level and their association with children's swimming abilities (Chan et al., 2020;Irwin et al., 2009). Furthermore, low family income may be negatively related to incidents of child drowning (Saluja et al., 2006) as special gear may be used for water safety such as life vests and floating aids. Therefore, future studies can incorporate such indicators such as the parents' socioeconomic status into the evaluation of a practical water accident prevention programs in natural bodies of water. Reason (2000, p. #) pointed out that "culture transcends the psychology of one person" and that safety behaviors depended on cultural factors and vary systematically across cultures. Ethnic norms have also been explored in order to understand water accidents (Saluja et al., 2006), which suggests the importance of conducting cultural comparisons (Chan et al., 2017). This case studied only one ethnic group in a Japanese junior high school; therefore, it would be interesting to see if the results would be the same in groups from different cultural backgrounds.

Conclusion
This experiential study explored two perspectives to assess high students' performance while swimming with clothes on in a natural body of water. The first perspective was the increase in the students' awareness of water accident prevention; the effects of the practical training t remained after 5 months, which indicated that awareness had been sufficiently increased as a result of the practical training. The other perspective was whether this practical training education exercise increased students' general awareness of safety given the diversity of threats and dangers. However, while we confirmed that students were more cognizant of water accident prevention, the exercise failed to raise the students' awareness of general safety issues.
There is a need for school safety education so that children are equipped with skills that would enable them to respond appropriately when faced with unknown threats and dangers. Just as the COVID-19 pandemic required epidemic prevention education, students must be given safety instructions that increase their general awareness of all types of threats and dangers as there may be time constraints when a new threat emerges. Therefore, the effect of practical training on one threat can give students an awareness of other threats and dangers in general. Since the practical water safety training in this study did not successfully raise students' general safety awareness, it may be necessary to provide additional instructions to ensure that the effects of the practice are generalized in future courses.

Data availability statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Acknowledgments
We would like to express our sincerest gratitude to all the people involved in the junior high school experiment, as well as to all the children and students who participate in this research. In addition, this research would not have been possible without the cooperation of the fire departments that have jurisdiction over the junior high school. Again, we would like to express our most sincere gratitude to them.

Citation information
Cite this article as: Assessing the effectiveness of a water safety program for junior high school students in Japan 1 , Toshiyuki Kishi, Masafumi Ohnishi & Ryousuke Inagaki, Cogent Education (2023), 10: 2205810.

Statements of ethics approval
This study was conducted with the approval of the Ethics Review Committee of Gifu Seitoku Gakuen University (No. 202101).

Disclosure statement
No potential conflict of interest was reported by the authors.