Indonesia has a very high flood disaster risk index (Marfai, Sekaranom & Ward 2015; Yamamoto, Sayama & Apip 2021). The data from the National Disaster Management Authority (BNPB) of the Republic of Indonesia states that from 2014 to 2023, flood disasters occupied the highest percentage, with a total of 8334 events. In the future, the probability of flood disasters is expected to increase. This is due to global climate change, which causes an increase in extreme rainfall in most parts of Indonesia (Yamamoto et al. 2021).

Therefore, efforts are needed to reduce the impact of flood disaster risk. One of them is to encourage the right behaviour before, during and after a flood disaster (Toyoda & Tanwattana 2023). This can be done with flood disaster simulations that are carried out regularly. Furthermore, disaster education has been recognised as playing an important role in increasing the knowledge and capacity to reduce the impact of disasters (Gampell et al. 2021; Novak, Lozos & Spear 2019; Oktari et al. 2015). Scientific evidence also shows that cognitive activities during education will have long-term effects on human neurological function (Eslinger et al. 2009). This will change the way educated individuals think, reason and solve problems (Baker et al. 2011). Disaster education is believed to motivate people, especially students, to take disaster preparedness actions (Hoffmann & Muttarak 2017).

However, disaster education generally follows the cognitive learning model (Tatebe & Mutch 2015; Yasuda, Muramoto & Nouchi 2018). Teachers only transfer basic conceptual and theoretical knowledge of disaster prevention. The knowledge includes how disasters occur and the impacts of disasters. Unfortunately, it does not effectively increase students’ motivation to take preparedness actions (Muzenda-Mudavanhu, Manyena & Collins 2016; Shaw, Kobayashi & Kobayashi 2004; Yildiz, et al, 2021). In addition, conceptual learning is also not relevant to the objectives of disaster education as described in the Sendai Framework for Disaster Risk Reduction conceptual framework, which emphasises knowledge, skills and attitude (Kelman & Glantz 2015).

Tsai et al. (2015b) argued that disaster education has been using a conventional module-based approach, which is based on one-way information transfer process. This learning approach often has low emotional involvement, and it does not have an impact on behavioural change (Feng et al. 2020). In addition, disaster education has also not combined knowledge with action (Aksa et al. 2025). Previous researchers have found that the traditional learning approach has been recognised as limited and ineffective (Mossoux et al. 2016). It was argued that lecture and recitation learning methods generally do not have a significant impact on student learning outcomes (Kankanamge et al. 2020).

Therefore, a new pedagogical approach is needed to make disaster learning fun for students. Game-based learning is considered a good method to enable students to use the skills and thought processes they have learned to take appropriate actions during a disaster (Galbusera et al. 2022; Tsai et al. 2019). Serious Disaster Games (SDG) contribute positively to the learning process because they are heuristic (Mitsuhara et al. 2013; Mossoux et al. 2016; Tsai et al. 2019). Players can experience complex situations that are virtually illustrated (Mossoux et al. 2016). Game-based learning has a significant influence on education (Tsai et al. 2015a). Many researchers recognise game-based learning as having great potential when compared to traditional teaching and learning methods (Gampell et al. 2021; Kankanamge et al. 2020; Tanwattana & Toyoda 2018). Researchers assert that game-based learning can increase student engagement in learning as well as increase disaster awareness (Tsai et al. 2019). However, so far, there is very limited research when it comes to developing and implementing game-based disaster learning, especially in Indonesia. This study aims to develop and test SDG: Flood Alert! in improving evacuation behaviour.

The results showed that Serious Disaster Game: Flood Alert effectively increases student knowledge (T-test, p < 0.005). There was an increase in knowledge score after learning using SDG Flood Alert! From the three aspects of knowledge assessed, scores significantly increased in the component of what to do during a flood (Pre-test M = 1.96 on a scale of 4 and Post-test M = 3.91 on a scale of 4). Most students were able to answer well on the actions that should be taken during a flood disaster (Table 3). This is because SDG Flood Alert product is designed using Kolb’s experiential learning cycle theory, which consists of several stages: Concrete experience, reflective observation, abstract conceptualisation (building working models and hypotheses) and active experimentation (new learning tried out in practice). For example, at level 1 (before the flood), players can take actions to reduce the impact of the flood, such as identifying evacuation routes, storing important documents and preparing a disaster preparedness kit. In the next level, players can also practice evacuation by not crossing undercurrents, waterways and puddles. The findings of this research corroborate with previous research which shows that game-based learning effectively improves knowledge and capacity in disaster risk reduction (e.g. D’Amico et al. 2023; Schueller et al. 2020; Tsai et al. 2019).

The advantage of the developed SDG Flood Alert product is that it provides a fun learning experience for students. They not only learn based on theory and conceptual knowledge, but are also involved in a flood disaster simulation. During the game, students must search for answers and determine what actions to take during a flood. This provides long-term retention for them when facing future disasters. Game-based learning strongly influences child development and enhances practical experience (Tsai et al. 2015). Turning knowledge into action is important in disaster risk reduction (Aksa et al. 2020a, 2020b; Aksa 2022). The SDG: Flood Alert is therefore a disaster learning media that combines knowledge and action by engaging directly in the game.

In addition, the use of SDG: Flood Alert also significantly affected students’ self-efficacy (T, p < 0.05).

There was an increase in scores after learning using SDG: Flood Alert (Table 3). Most students were confident that they could effectively deal with a flooding emergency (pre-test M = 2.31 and post-test M = 4.31). Students are also confident they will be able to deal with a flooding emergency even though the water level is critical, and the water speed does not allow it to move easily (Table 4).

The post-test results also show that students know how to handle flood emergencies (M = 4.40 on a scale of 5) (Table 4). This increase in self-efficacy is because SDG: Flood Alert is designed to present various challenges during a flood. Players are directed to evacuate by not passing through undercurrents or puddles and not touching electrically charged objects. The game scenario effectively increases students’ confidence and readiness to face floods.

The findings of this research corroborate previous researches, which found that using serious games as interactive learning media significantly increases self-efficacy and disaster preparedness (Chittaro & Sioni 2015; Feng, Duives & Hoogendoorn 2021a; Feng et al. 2021b). Students who have high self-efficacy tend to take protective actions in the face of disasters (Chittaro & Buttussi 2015). Those with high self-efficacy believe they can take the right action or decision when a disaster occurs. This is because self-efficacy is an important indicator of individual decision-making during emergencies (Liu et al. 2023; Newnham et al. 2017; Yip et al. 2013).

The SDG: Flood Alert presents an engaging learning scenario for students to face danger that can depict actual flood conditions. This can train them to take appropriate and effective actions to manage better flood disaster events. The results show that students are motivated and confident to face flood emergencies.

Meanwhile, related to intrinsic motivation, most students enjoy learning using SDG: Flood Alert (M = 4.40 on a scale of 5). Most students also mentioned that SDG: Flood Alert is very interesting (M = 4.06 on a scale of 5).

Only a small number mentioned that SDG: Flood Alert is boring (M = 1.80 on a scale of 5) (Table 5). The high intrinsic motivation among students after using SDG: Flood Alert is because the SDG gameplay is designed interactively; a player can walk in four directions and can interact, such as picking up or storing objects. Gameplay SDG: Flood Alert allows players to explore the surrounding environment and complete mission commands.

Based on these findings, SDG: Flood Alert can be used as an interactive learning media for students. The SDG: Flood Alert can be an alternative solution to overcome the limitations of disaster learning using a conventional module-based approach.

In the context of disasters, experience has a significant effect on preparedness. However, not all people have had direct disaster experiences. This study concluded that SDG: Flood Alert plays a crucial role in increasing students’ behaviour during flood evacuations. It can be further concluded that SDG: Flood Alert is effective in increasing students’ knowledge, self-efficacy and intrinsic motivation. The SDG: Flood Alert provides an enjoyable learning experience for students, which prepares and empowers them to respond effectively during flood events. Students not only learn theory and conceptual knowledge, but are also involved in disaster simulations, which train students’ decision-making skills. SDG: Flood Alert learning can provide long-term retention for students. This study further concluded that its findings corroborate with previous researches, which found that game-based learning is effective in increasing students’ capacity for disaster risk reduction. This research highlights that turning knowledge into action is an important component of disaster risk reduction.