Tsunami Understanding Media: Android-Physics Mobile Learning to Improve Problem Solving-Skills and Natural Disaster Preparedness

Received: August 2, 2020 Accepted: October 22, 2020 Published: October 30, 2020 This research aimed to determine the effectiveness of the implementation of tsunami understanding media through Android-based physics mobile learning in improving students' problem-solving skills and natural disaster preparedness. This study employed quasi-experiment research with a nonequivalent control group design. This research subjects were fifty-six students of class XI IPA at SMA N 1 Kretek, Bantul, Yogyakarta. The sampling technique employed in this research was the simple random sampling technique to determine the experimental and control groups. The research data had been obtained through pretest and posttest in the form of essay tests. The data obtained were then analyzed using N-gain. The analysis results were used to determine the effectiveness of the media using the Hotelling's Trace test in MANOVA test analysis. Normally distributed and homogeneous data are required to perform the multivariate analysis. The results showed that the tsunami understanding media through Android-based physics mobile learning improved students' problem-solving skills and natural disaster preparedness compared to the learning using PowerPoint presentation. The research decisions were based on the significant value that was smaller than 0.05 (0.000<0.05), and the value of the Partial Eta Squared was 0.359 in the broad category.


INTRODUCTION
Indonesia is an archipelago located between two continents, Asia and Australia, and located between two oceans, namely, the Pacific Ocean and the Hindian Ocean. As a result, Indonesia is often hit by an earthquake because it is the meeting point of three tectonic plates, namely the Pacific plate, the Indo-Australian plate, and the Eurasian plate (Simkin et al., 2006).
The colliding plates can form volcanoes; thus, Indonesia becomes part of the Pacific Ring of Fire, which is the area with the most active volcanoes and earthquakes in the world (Simkin et al., 2006). Large earthquakes or volcanic eruptions at sea can sometimes trigger a tsunami. Based on the National Geophysical Data Center/World Data Service (NGDC/WDS) data in July 2020, tsunamis in Indonesia are generally triggered by shallow sea earthquakes or volcanic eruptions. A tsunami can cause many damage and casualties, especially if it hits beaches with a dense population. The Aceh tsunami disaster in 2004 was estimated to have caused 220,308 fatalities (Syamsidik et al., 2019).
Bantul, as one of the regencies in Yogyakarta, frequently experiences earthquakes because there is a subduction zone in the south of Java that has the potential of megathrust earthquake magnitude greater than 7.0-8.0 SR (Hartoko et al., 2016;Mulia et al., 2019). On the south coast of Java, there is a seismic gap positioned at the coordinates (109 ° E -110 ° E), potentially causing an earthquake with a large scale or megathrust in the future (Kongko & Hidayat, 2014). The scenario of a tsunami caused by the megathrust earthquake on the south coast of Java, such as Bantul, Gunungkidul, and Pacitan, has been simulated to map and determine the risk of the disaster (Kongko & Hidayat, 2014;Mardiatno et al., 2015;Mulia et al., 2019). Based on the results of computer simulations, an earthquake with a magnitude of 8.2 can cause a tsunami with a height of 9 meters with an average distance of 2 km. It can damage the Bantul Regency, Yogyakarta, covering an area of 20 km 2 (Kongko & Hidayat, 2014). Disaster preparedness is one of the main focuses on the framework developed by the United Nations. The Sendai Framework for Disaster Risk Reduction 2015-2030 aims to reduce the damage caused by natural phenomena such as earthquakes, tsunamis, floods, droughts, and storms through preventive ethics (UNISDR, 2015). Several steps need to be carried out to reduce damage and casualties due to tsunami, such as socialization, simulation, construction of facilities and infrastructure, and research on geographical conditions (Bakar Sambah & Miura, 2019;Pahleviannur, 2019;Zuhdi et al., 2019). Understanding of disasters and their mitigation needs to be informed to citizens on an ongoing basis through communities and institutions such as schools (Parwanto & Oyama, 2014;Setiawati et al., 2013). Disaster Risk Reduction Program (DRR) through the implementation of disaster mitigation programs is a necessity that must be done through formal and non-formal education with curriculum design from Badan Standarisasi Nasional Pendidikan (BSNP) by integrating the subjects that are strongly related to disasters (Septikasari & Ayriza, 2018).
Critical thinking skills, problem-solving, integrating environment, collaborating, and using digital tools as learning tools are required by the student in 21 st -century (Alismail & Mcguire, 2015;Scott, 2015). Students with low problem-solving skills can influence physics mastery in physics learning, such as cannot apply equations, mathematical calculations, and errors in using units (Jua et al., 2018). The poor problem-solving skills can be caused by teaching styles, learning methods, and inappropriate media (Azizah et al., 2015;Mustofa & Rusdiana, 2016).
Smartphones with Android OS can help educators develop learning applications that contain information, simulations, explanations, and evaluation that can be accessed anytime and anywhere (Astra et al., 2015;González et al., 2014;Jengathe & Rojatkar, 2015). Smartphones can be utilized as a medium to spread dan share information related to natural disasters and their mitigation (Nadian et al., 2015;Winarni et al., 2018). Learning by using a smartphone application as learning media can also improve critical thinking, creative thinking, and problem-solving skills (Astuti et al., 2018;Ismail et al., 2016;Mulhayatiah et al., 2019).
According to the background, disaster education is rarely carried out at disasterprone schools because there are no specific subjects regarding disaster preparedness. Teachers can use a tsunami understanding media in physics learning to understand tsunami and physics's subject together. The learning media used the Android OS to be installed on students' smartphones and can be accessed offline anywhere and anytime. The learning media is equipped with materials and evaluations on natural disaster preparedness and physics problems. This research aimed to determine the effectiveness of the implementation of tsunami understanding media through android-based physics mobile learning to improve problem-solving skills and natural disaster preparedness.

METHOD
This research employed the quasiexperimental with non-equivalent control group design. The design can be seen in Table 1. O1 is the pretests for the experimental and the control classes, O2 is the posttests for the experimental and control classes, and X is the applied learning media.
This research was conducted at SMA N 1 Kretek in February-March 2020. The population of this research was fifty-six students of class XI IPA. The entire population was used as research subjects consisting of 28 students of Class XI IPA 1 and 28 students of Class XI IPA 2. The research subjects were determined by random sampling.
This study aimed to determine the effectiveness of the tsunami understanding media through Android-based physics mobile learning to improve problem-solving skills and natural disaster preparedness. The problem-solving skills can be measured using five essay questions of traveling wave topic. Natural disaster preparedness can be measured using eleven essay questions. The research instruments can be seen in Figure 1 and Figure 2. In this study, the natural disaster preparedness indicators covered the knowledge and attitudes towards the risk of natural disasters from the tsunami before, during, and after the disaster. It was focused on the socialization and training of natural tsunami disasters. The risks of natural tsunami disasters focused on the south coast of Java, especially the Bantul district's beaches. Problem-solving skills indicators referred to several indicators shown in Table  2. Experts, teachers, and peers had validated the tsunami understanding media through Android-based physics mobile learning and the research instruments used in this research. The validation results showed that the media and research instruments were valid and suitable for learning physics. The display of the media is shown in Figure 3 and Figure 4.  This research was conducted by giving a pretest. The tsunami understanding media through Android-based physics mobile learning was used in the experimental class, while PPT was used in the control class. Both groups used the same learning model, namely problem-based learning and learning cycle (5E). At the last meeting, a tsunami disaster simulation was conducted. Students should open the media to access the material in the experimental class, while in the control class, students must open a paperbased media. Worksheets were used to collect data and answer questions. Each group presented the results in front of the class. After all learning processes had been completed, the students were given a posttest. The research process is presented in Figure 5. The pretest and posttest were administered to collect the data.
The following formula can determine the increase in each dependent variable using Ngain:

The Improvement of the Problem-Solving Skills and Natural Disaster Preparedness
The improvement of students' problemsolving skills and natural disaster preparedness was obtained from the N-gain value of pretest and posttest from each assessment instrument. The category of improvement can be seen in Table 3 (Sundayana, 2015). It was found that the students' frequency in improving their problem-solving skills and natural disaster preparedness based on the N-gain category is shown in Table 4 and  Table 5.  The highest frequency of improvement can be found in the experimental class with 23 students (high category). Five students were in the intermediate category.
Meanwhile, the highest frequency of natural disaster preparedness can be found in the experimental class with a frequency of 20 students (high category). Eight students were in the intermediate category. From these results, it can be concluded that the students' problem-solving abilities and natural disaster preparedness improved well.
This research was in line with other studies that integrate media disaster education, such as earthquake and landslide, in physics learning (Labibah et al., 2019;Rany et al., 2020). The physic learning integrated with natural disaster education can also improve natural disaster preparedness (Rahmawati et al., 2020). Meanwhile, this research focused on tsunami disaster. In the evaluation, the students were asked to determine the appropriate actions in the event of a tsunami under certain conditions. Providing disaster information from pre, while, and post-disaster can increase awareness (Pahleviannur, 2019). Knowledge of tsunami disaster preparedness can be improved through the learning media. Information on risks and actions if a disaster does occur is essential in natural disaster preparedness (Rafliana, 2017;Ranke, 2016). Also, some disaster materials in learning media had been integrated with physics learning, such as the relationship between ocean waves and traveling waves, the theory of density when an object or person floats in the sea, and the hydrostatic pressure on a car door when it drifts.
This research was also in line with other studies that use physics mobile learning by integrating certain local wisdom to improve problem-solving skills (Gebze et al., 2020;Shabrina & Kuswanto, 2018). The use of digital media as a learning media can make it easier to train problem-solving skills because students can directly see the material presented in the form of moving images or videos (Mulhayatiah, 2019).
Hesse (2015) states that problem-solving is a decision to process problems through reasoning skills. This decision is included in the evaluation of choosing the right decision when a tsunami occurs, from how to identify scenarios, decide and apply learned preparedness, and re-evaluate actions based on procedures. Besides, in solving physics problems, steps are given from what is known and asked for answers and apply the equation by referring to the material available in the learning media. These steps are linked to indicators in training problemsolving skills. So, through the problems displayed in the media, problem-solving skills can be improved.

The Effectiveness of Problem-Solving Skills and Natural Disaster Preparedness
The results of effectiveness research can be analyzed if the prerequisite assumptions are met.

a. Normality Test
A multivariate normality test was done by looking at the relationship between Mahalanobis distance and chi-square values. If the scatter-plot for the graph between the Mahalanobis distance and chi-square tends to form a straight line with more than 50% values, it can be concluded that the population is normally distributed. The data used in the normality test were the N-gain data of each group. The results of the analysis are shown in Table 6. In the control class and the experimental class, the Mahalanobis distance value was smaller than the chi-square value (more than 50%). The Pearson correlation was used to determine the degree of correlation between the Mahalanobis Distance and the Chi-Square. The correlation levels are shown in Table 7.  Table 7 shows that sig. value was less than 0.05. thus, the scatter plot tends to form a straight line. It can be concluded that the data population of the experimental and the control classes were usually distributed.

b. Homogeneity Test
The homogeneity tests of variancecovariance matrices were analyzed using the Box-M homogeneity test to find out whether the experimental and control classes had homogeneous covariance matrices or not. The results of the Box-M homogeneity test using the SPSS 23 program are shown in Table 8. Based on the homogeneity test, the sig. value was equal to 0.149. Thus, the value was higher than the significance level of 0.05. It can be seen that the experimental and control classes had homogeneous or equal covariance matrices.

Hotelling's Trace Test
To determine the difference in the improvement of problem-solving skills and natural disaster preparedness, the Hotelling's Trace test was used in the MANOVA test analysis. The hypotheses of the research were; H0: There was no significant difference between the improvement of problemsolving skills and natural disaster preparedness in the experimental class and the control class H 1 : There was a significant difference between the improvement of problemsolving skills and natural disaster preparedness in the experimental class and the control class The decision was determined by looking at the sig value. If it is less than <0.05, then H0 is rejected. On the other hand, if it is more significant than> 0.05, then H0 is accepted. The results of the analysis are shown in table 9. The results of the Hotelling Trace test analysis showed that the sig. The value was 0.000, less than 0.05. Based on the hypothesis, H0 was rejected. The result showed a significant difference in problemsolving skills and natural disaster preparedness in the experimental class and the control class. The effectiveness of the media can be seen in the partial eta squared value (0.359). According to Cohen (1988), if the partial eta squared value is more than 0.14, it has an effect size value within a broad category.
Even though both groups used the same learning model, the experimental class treatment was more effective because the students can access the learning material anytime and anywhere. Students; mobility does not depend on time, location, and space (Zhang, 2015). The benefits and advantages of a smartphone in learning are in line with McQuiggan (2015), who states that it can be effectively used outside of school because it can be accessed easily. Even though smartphones have several advantages, using smartphones as learning media in class requires extra supervision to avoid activities outside of learning, such as accessing social media and playing games.

CONCLUSION AND SUGGESTION
Learning by linking physics and natural disaster preparedness can improve problemsolving skills and knowledge about natural disaster preparedness. Tsunami understanding media through Android-based mobile learning physics is more effective in learning than learning through PPT media. For further research, teachers are expected to monitor smartphone usage in learning better. The development of similar learning media using different media types, disasters, and dependent variables can be done in future research.