Development and Evaluation of a Computer Detective Game for Microbial Food Safety Education

Abstract

Past studies have suggested that young people lack knowledge regarding food safety, and that food safety education is appropriate for integration into science education since it often involves science knowledge. Thus, this study combined the methods of inquiry-based and game-based learning to develop a computer detective game, called the Poison Riddle, that makes students actively explore food safety knowledge through playing the role of a science detective to solve a task related to food poisoning at home in a virtual world. To evaluate the learning effectiveness of this game, 109 high school students were selected as the research participants. The research findings indicated that this game could help students improve their microbial food safety knowledge. Most students had positive participation perception and gaming behaviors related to the game. In addition, the analysis indicated that the students who successfully solved the game task gained more positive gaming behaviors, food safety knowledge, and significant sequential behaviors compared to those who were unsuccessful in solving the game task.

Keywords

educational game food safety education microbial food safety and youth inquiry-based learning

Foodborne illness, also known as food poisoning, has always been a global health issue. A research suggests that one in every six U.S. citizens has been ill due to food poisoning, and that the disease causes 3,000 people to die annually (Scallan et al., 2011). It is clear that food poisoning is highly prevalent, and devoting more efforts to the prevention of food poisoning worldwide is necessary. Although food poisoning can be prevented through education, past studies have suggested that young people lack knowledge regarding food safety (Ferk et al., 2016; Lange et al., 2016; Majowicz et al., 2017), and that little time is allocated for specialized food safety courses provided in formal school curricula (Carraway-Stage et al., 2015; Fischer & Frewer, 2008; Stage et al., 2018). Therefore, providing food safety knowledge to the general public is crucial to prevent food poisoning.

In previous studies, methods such as traditional teaching, posters, films, and case studies have been used to design food safety educational activities (Shearer et al., 2013). Following the advances in technology, computers, online courses, and even social media are also considered to be effective methods to improve students’ food safety knowledge (Fenton et al., 2006; Mayer & Harrison, 2012; Pintauro et al., 2005; Wantland et al., 2004). Food safety education often involves microbiology knowledge because microorganisms such as bacteria are a common cause of food poisoning. In addition, some researchers have suggested that students have limited science knowledge regarding the pathogens when it comes to foodborne illness (Ferk et al., 2016). Therefore, some scholars have recommended that food safety education can be integrated into schools’ science education curricula (Beffa-Negrini et al., 2007; Shearer et al., 2014).

Besides the traditional lectures in science classrooms, helping students learn science and understand scientific knowledge through scientific inquiry activities has gradually become a crucial trend in science education (Furtak et al., 2012) under the advocacy of science education organizations and relevant scholars (National Research Council, 2013). Accordingly, an increasing number of researchers have recently advocated the use of computer simulation combined with gamification mechanisms to conduct scientific inquiry activities that help students acquire scientific knowledge. Examples of such mechanisms can be found in projects such as Quest Atlantis: Taiga Park (Barab et al., 2010), River City (Ketelhut & Nelson, 2018), and Crystal Island (Rowe & Lester, 2015) as well as in virtual performance assessment (Baker et al., 2016). These projects all used computers to simulate virtual environments, such as forests, swamps, or rivers to place students in a virtual game environment that enables them to engage in scientific inquiry activities. In addition, these activities mainly required students to proactively solve science-related problems and complete tasks to acquire scientific knowledge and inquiry skills. For example, the Taiga Park mission in Quest Atlantis requires students to act as field investigators to solve a scientific problem caused by the sudden decline in the number of fishes in a river. In this virtual world, students engaged in activities such as scientific evidence collection, experiments, and analyses that are similar to real-world experiences (Barab et al., 2007a). Moreover, these virtual experiences have been found to improve students’ science-learning effectiveness (Barab et al., 2007b; Hickey et al., 2009).

Therefore, if the attempt is made to integrate food safety education into schools’ science education curricula, providing activities to improve students’ food safety knowledge through these computer-stimulated scientific inquiry games may be a feasible method. Based on this idea, a computer game named the Poison Riddle was developed in this study. Previous studies have revealed that food poisoning cases at home are more common than large-scale food poisoning (Pouillot et al., 2010; Young et al., 2015). This indicates that food preparation and handling at home is the main cause of food poisoning cases. Therefore, this study focused on food safety education at home. The proposed game enables students to play the role of a detective to solve a scientific detective task related to food poisoning at home in a virtual world. In addition, past studies have also suggested that high school is the appropriate time for students to start learning about food safety (Diplock et al., 2017, 2018), as students at this age could cook on their own and are old enough to learn the scientific facts behind food poisoning (Haapala & Probart, 2004; Lynch et al., 2008). Hence, the computer detective game developed in this study was mainly designed for high school students and its learning effectiveness was evaluated by selecting high school students as the research participants. The main purposes of this study are listed as follows:

• Develop a computer detective game that could enhance students’ knowledge of microbial food safety.

• Evaluate the changes of students’ knowledge on the microbial food safety following their participation in the proposed detective game.

• Examine students’ perceptions after their participation in the proposed detective game.

• Study behavioral differences among students while playing the proposed detective game.

The Poison Riddle Game

The Poison Riddle is a first-person view’s single-player game that can be played individually on any web browser. In the game, each participant plays the role of a member of a scientific detective team, and their goal is to identify the cause of food poisoning in a household after lunch. Players can see different virtual household spaces simulated by this game, such as the living room, dining room, kitchen, bedroom, study room, and balcony. They can also interact with different nonplayer characters (NPCs) such as the household’s father, mother, and siblings in different game scenes to help them obtain clues about the reason for food poisoning in the household. Figure 1 represents the interface of the game and the game scene of the virtual bedroom. When the player clicks on the virtual character in the bedroom, a preset information menu appears, and the player can select clues they wishes to obtain from the character. For example, when the player clicks one of the information icons depicted on the left side of Figure 1 (i.e., what the character ate for lunch), a message about the meal the character had before food poisoning will appear, as illustrated on the right side of Figure 1. Players may also use the notepad function to record the crucial messages.

Figure 1.

Game Interface and Clue Collection Method of the Poison Riddle.

Moreover, to identify the cause of food poisoning, the player can collect different food samples for bacterial culture and observation. As illustrated in the upper left image of Figure 2, the player can collect sausage leftovers from lunch in the dining room. Next, as illustrated in the upper right part of Figure 2, a bacterial incubator can be used for bacterial culture in the virtual study room. When operating the bacterial incubator, the player can choose three different temperatures (low, high, and moderate) to culture the bacteria. Once the bacteria are cultured, a microscope can be used for bacteria observation. As indicated in the lower right part of Figure 2, if bacteria are observed, the name of the bacteria will appear on the simulated screen of the microscope. The player can use functions such as zoom out, zoom in, and moving on the microscope to observe the bacteria. Meanwhile, the player can collect observable bacteria data into the simulated computer in the study room; subsequently, they can see the collected bacteria data in a sheet on the simulated computer, as illustrated in the lower left part of Figure 2. Thus, the player can use the virtual computer to conduct a comparative analysis of the experimental data.

Figure 2.

Bacterial Culture and Observation in the Poison Riddle Game.

Thus, each player can identify the cause of food poisoning in this virtual household by collecting information about the diet and the symptoms of poisoning from the household members, and by examining the collected food sample and bacteria culture. However, the premise is that the player must have a basic understanding of bacterial food poisoning, such as the pathogens and symptoms of various types of bacterial food poisoning. Therefore, in the virtual study room, microbiology-related learning materials for bacterial food poisoning are provided for the players (as illustrated in the left part of Figure 3). Moreover, the function of the online test has also been designed for players to assess their understanding of bacterial food poisoning and to provide an opportunity to acquire game points from the tests.

Figure 3.

Online Learning Materials and Online Tests in the Poison Riddle Game.

To provide assistance for each player, an NPC, called the captain of the scientific detective team, was designed in this game. This NPC introduces the purposes of the game to the players at the beginning of the game, and asks players to complete three simple training tasks so that they could quickly understand each function of the game before starting the formal game task. The players must report their results to the captain following the completion of each task. They must report three main investigation results: the source of the poisoned food, the name of the bacteria implicated, and the main cause of food poisoning, to the captain by using the format of multiple-choice questions after completing the formal task. Moreover, this game uses virtual game points to encourage game participation. Players can earn game points by engaging in activities, such as collecting new information, food samples or performing online tests. They can also use the game points to exchange up to eight hint messages about the game from the captain in every 1000 points. According to players’ game point scores, the top 20 players can become members of the leaderboard in this game. Meanwhile, players must earn a certain number of points before reporting their final investigation results to the captain. All players’ click behaviors and scores in game are stored securely in MySQL database. This game was pilot tested by a 12th-grade class of 30 students before the formal experiment, and the participants held a positive attitude toward the game (Tsai, 2018).

Research Method

To evaluate the effectiveness of the self-developed computer game, this study conducted one group quasi-experiment research. The participants, procedures, and instruments of this research are presented in the following sections.

Research Participants

This study randomly selected four classes of twelfth-grade students from a high school with eight twelfth-grade classes in Kaohsiung City of Taiwan as the participants. After removing 16 students who were unable to participate fully in the experiment due to absence, this study finally had 109 participants (71 males and 38 females). All the participants had taken a basic biology course in their freshmen year of high school.

Research Procedures

The experiment was conducted over 3 weeks, encompassing three class periods (lasting 50 minutes each) in the participants’ science and living technology course. The experimental procedure began by asking students to perform a self-developed pretest about microbial food safety knowledge. Following the pretest, students were asked to operate a computer by themselves to play the Poison Riddle game. Finally, the participants were asked to complete a posttest on microbial food safety knowledge (with the same test contents as the pretest) and a participation perception scale.

Research Instruments

Knowledge Test on Microbial Food Safety

In this study, a knowledge test on microbial food safety was created; the goal was to understand the effect of the self-designed detective game on students’ learning of food safety knowledge. The questions in the test were based on knowledge related to microbial food safety emphasized in the detective game. The main knowledge concepts of the test included classification of microorganisms, bacteria and bacterial food poisoning, and the prevention of bacterial food poisoning. The test contained a total of 20 multiple-choice questions such as “Which type of bacteria is Vibrio cholerae?”, “What type of bacterial food poisoning can be caused by Staphylococcus aureus?”, and “Which of the following substances may contain Bacillus cereus?” The maximum score of the test was 100 points, and the appropriateness of the questions in the test was reviewed by a qualified high school teacher. The test’s Kuder–Richardson reliability was .72.

Participation Perception Scale

This study created a 14-item participation perception scale rated on a 5-point Likert-type scale to understand students’ perceptions toward the detective game. Among the 14 items of the scale, 4 were related to the player’s gaming feelings, such as “I feel that this detective game is enjoyable” and “In this detective game, I applied myself to earn points.” Six of the items were related to their learning perception, such as “I gained knowledge related to food poisoning through this detective game” and “I gained knowledge about bacteria observation methods through this detective game.” The remaining four items were related to cognitive load perception, such as “I feel that the detective game contains an excessive amount of relevant information, and thus the game is too difficult for me” and “I was frequently distracted during the detective game.” Furthermore, an open-ended question was included in the end, so that students could freely write and express their thoughts, feelings, and suggestions about the game. The Cronbach’s α values for the overall scale, game perception scale, learning perception scale, and cognitive perception scale were .94, .92, .97, and .91, respectively.

Results

The Analysis of Students’ Knowledge Acquisition

To understand whether students’ knowledge about food safety improved following their participation in the Poison Riddle game, this study asked students to complete the knowledge test of microbial food safety before and after playing the game. This study found that students’ mean score in the knowledge pretest was 44.11 (N = 109, SD = 10.94), and their mean score for the knowledge posttest was 72.25 (N = 109, SD = 15.40). A paired sample t test indicated that students’ posttest score was significantly higher than their pretest score—t(108) = 14.77, p<.05. In other words, students showed significant improvement in the results of the food safety knowledge test following their participation in the Poison Riddle game. Therefore, the computer detective game developed in this research may contribute to students’ learning of microbial food safety knowledge.

The Analysis of Students’ Participation Perceptions

To understand students’ perception of the Poison Riddle, this study asked the research participants to fill out a participation perception scale following their participation. Students’ mean score for the game perception scale was 3.64, suggesting that most students have a positive attitude toward game participation. A detailed analysis also indicated that students’ mean score on the item “I feel that this detective game is enjoyable” was relatively low (M = 3.14), which could affect students’ mean score on the game perception scale. The result implied that the Poison Riddle game can still be improved in the future to provide more enjoyable gaming experience for players. However, the nature of this game is not merely entertainment; players are required to constantly engage in learning and problem solving during the gaming process, which could affect the participants’ feelings regarding the enjoyment of this game.

Students’ mean score on the learning perception scale was 3.76, which indicates that most students held a positive view toward the benefit of science knowledge learning. Moreover, a detailed analysis indicated that students’ mean scores on questions relevant to learning perception were all above 3.6. For example, students maintained a positive attitude toward items such as “I gained knowledge related to food poisoning through this detective game” (M = 3.91) and “I feel that this detective game is helpful to learn scientific knowledge” (M = 3.72).

Students’ mean score on the cognitive load perception scale was 2.83, which suggests that most students held a relatively negative view on whether this game increased their cognitive load. For example, students gave a mean score of below 3 for items such as “I feel that solving the tasks of this detective game was overly difficult” (M = 2.73) and “I feel that the detective game contains an excessive amount of relevant information, and thus the game is too difficult for me” (M = 2.95). This suggests that most participants did not think that the gaming process increased their cognitive load.

Finally, after analyzing students’ responses to the open-ended question at the end of the scale, this study found that most students had a relatively positive attitude toward the game. Responses included for example: “I think the game was extremely enjoyable and fascinating. If I were asked to read books on this topic, I would have fallen asleep immediately. Through this game, I gained a considerable amount of everyday life knowledge”; “I think that this game is entertaining and informative. That is, through a relaxing game style and the provision of detailed information, this game enabled us to gain knowledge and interest in food poisoning and microbiology”; “I gained knowledge about microbiology”; “I gained a substantial amount of relevant knowledge”; and “The game was enjoyable”. However, a small number of students expressed that the game was relatively difficult. For example, these were responses such as “Interesting information was provided, but the load of information was beyond my ability to absorb,” “The game was helpful for learning, but the information regarding bacteria was suddenly overwhelming and thus difficult to remember,” “I found the game difficult,” and “More hints should be provided to enable us to complete the game more easily”. These negative responses could be the reason why students did not provide a high rating on the enjoyment of the game. This also implied that this game can be improved in terms of the level of difficulty or by providing more support.

The Analysis of Students’ Gaming Behaviors

To understand students’ behavior and performance in the Poison Riddle game, this study used computer game records left by the students to conduct a research analysis. First, as presented in Table 1, on average, students collected information from the characters of family members 70 times, read online learning materials 55 times, and performed online tests 166 times with a mean accuracy of .72. In addition, the students, on average, cultured bacteria, used the microscope, and collected crucial experimental evidence 21 times, 25 times, and 16 times, respectively. Therefore, the students’ frequent gaming behaviors within a short time indicated that most students participated proactively in this game. However, the game’s record indicates that only 85 people successfully found the final answer regarding the cause of food poisoning in the Poison Riddle game. That is, 24 people or 22% of the students could not find the correct answer, and only 78% of the students could successfully finish the tasks in the game.

Table 1.

Group Comparison of Various Gaming Behaviors and Participant Perceptions.

Variables		Group		t test
Variables	Overall (n = 109)	Successful group (n = 85)	Unsuccessful group (n = 24)	t test
Gaming perception	3.64	3.68	3.53	.69
Learning perception	3.76	3.80	3.61	.92
Cognitive load perception	2.83	2.76	3.08	−1.21
Number of times collecting clues	70	73	59	2.05*
Number of times reading learning materials	55	55	57	−.24
Number of times taking online tests	166	185	98	1.41
Accuracy of online test results	.72	.74	.69	56.34*
Number of times culturing bacteria	21	22	16	2.13*
Number of times using the virtual microscope	25	27	20	1.87
Number of times collecting evidence	16	18	10	2.78*

*p < .05.

To further investigate the differences in students’ gaming behavior, this study separated the participants into two groups: the successful and unsuccessful group, depending on whether they finished the tasks in the game, and further analyzed the differences in gaming behavior and participation perceptions between the two groups. The results are presented in Table 1. In terms of gaming behaviors, compared with the unsuccessful group, the successful group engaged in a significantly higher number of information collections from household members, bacteria culture, and experimental evidence, and exhibited significantly higher online test result accuracy. In terms of participation perceptions, the successful group did not have a significantly higher score in gaming perception and learning perception compared with the unsuccessful group. Although the successful group also had a lower cognitive load score than the unsuccessful group did, the difference was not statistically significant.

The analysis above indicated that the students in the successful group exhibited more proactive inquiry learning behaviors compared with their counterparts in the unsuccessful group. Therefore, the difference between the two groups of students’ performances in the knowledge test on microbial food safety was further investigated. This study adopted the student group (successful/unsuccessful) as an independent variable, the pretest score of food safety knowledge test as the covariate variable, and the posttest score of this test as a dependent variable to perform a one-way analysis of covariance (ANCOVA). Before the ANCOVA, the assumption of regression homogeneity was tested, F(1, 105) = .26, p > .05; it was not violated. In addition, the results on the Levene’s test of homogeneity of variance were nonsignificant, F(1, 107) = .003, p > .05, which did not violate the homogeneity of variance. The results of ANCOVA indicated that F(1, 106) = 6.69, p < .05, η² = .06, which suggests that whether or not students successfully solved the game tasks had a significant effect on their effectiveness in acquisition of microbial food safety knowledge after excluding the effects of the covariate. The least significant difference test indicated that the successful student group’s performance (M = 74.24) on the knowledge posttest of microbial food safety was significantly higher than that of the unsuccessful group (M = 65.18). Therefore, students who could successfully solve the game tasks gained more scientific knowledge compared with those who failed to solve the game tasks.

Moreover, this study applied data mining techniques to conduct a sequential pattern mining on students’ scientific inquiry behaviors in the game. Before conducting the sequential pattern mining, this study translated the raw action log data into inquiry-related actions. This included naming the action of investigating suspicious food in the game as “investigate food,” naming the action of various interactions with NPCs as “investigate clue,” naming the action of collecting suspicious food samples as “collect sample,” naming the action of various bacteria culturing, bacteria observation, and evidence collection in the laboratory as “data analysis,” naming the action of reading learning contents as “read content,” naming the record of using the notepad function in the game as “take note,” naming the record of engaging in online tests as “take test,” and naming the record of browsing hint messages provided in the game as “read hint.” Subsequently, this study deleted some raw action log data that were not related to scientific inquiry. For example, actions such as changing the password, checking the ranking list, entering different virtual rooms, and saving information in the game were filtered out from the raw data. Finally, each student’s inquiry sequences were retrieved from the revised action log data.

This study applied the prefixspan algorithm (Pei et al., 2004) in SPMF, an open source data mining tool (Fournier-Viger et al., 2014), to mine frequent sequential patterns of consecutive two-actions (length = 2) with the threshold of support rate = 0.3. A total of 55 frequent sequential patterns were identified from all students’ inquiry sequences. Subsequently, Jaccard similarity coefficient (Bazaldua et al., 2014) was used to extract the most interesting sequential patterns from the 55 sequences (as shown in Table 2). The 10 interesting patterns revealed that the students exhibited frequent inquiry behaviors such as investigating suspicious foods or clues first and then taking notes or tests in the game. Behaviors such as reading online learning materials or taking notes first and then taking online tests were also found.

Table 2.

Group Comparison of Interesting Sequential Patterns.

Sequential pattern	Successful group	Unsuccessful group	χ ²
Investigate food → take note	.68	.50	2.71
Investigate food → take test	.99	.88	6.79*
Investigate clue → take note	.69	.42	6.20*
Collect sample → read content	.79	.79	.001
Collect sample → take note	.67	.42	5.10*
Collect sample → take test	.97	.88	2.90
Read content → take test	.77	.83	.51
Take note → take test	.68	.54	1.63
Read hint → read content	.64	.67	.08
Read hint → take test	.74	.71	.10

*p < .05.

This study sought to understand the difference in the occurrence of interesting patterns between students who solved the game and those who did not. To achieve this goal, this study conducted a statistical analysis of chi-square test to make comparisons. As presented in Table 2, with the exception of the sequence for “read content → take test,” other sequences’ rates of occurrence in the successful group were higher. In addition, the rates of occurrence for three sequences: “investigate food → take test,” “investigate clue → take note,” and “collect sample → take note” in the successful group were significantly higher than those of the unsuccessful group.

Discussion

Every country should devote efforts to prevent food poisoning because it not only endangers the health of a country’s citizens but also hinders the country’s economic development. However, past research indicates that the general public normally lacks food safety knowledge, and specialized food safety courses are rarely provided in schools. Because food poisoning often involves microbial knowledge, it is appropriate to be integrated into science education curricula. Therefore, this study combined the methods of inquiry-based and game-based learning, which are proposed in current science education, to develop a computer detective game that makes students proactively learn food safety knowledge and to evaluate the effectiveness of this game.

This study found that most participants had very limited knowledge about microbial food safety before taking part in this game, which is consistent with the results of previous studies (Ferk et al., 2016; Lange et al., 2016; Majowicz et al., 2017). Nevertheless, after short-term participation in the game developed in this study, the students’ microbial food safety knowledge increased significantly, indicating that this game could help the students improve their food safety knowledge to a certain degree. This finding supports earlier researches on the use of computer networks to enhance students’ food safety knowledge (Lynch et al., 2008; Mayer & Harrison, 2012) and also supports previous studies conducted on the use of computer simulation games to enhance students’ scientific knowledge (Hickey et al., 2009; Lee et al., 2010; Rutten et al., 2012).

This study also found that most of the students exhibited a positive attitude toward this game. Specifically, most students agreed that this game helped them gain knowledge related to food poisoning, suggesting that the proposed game can encourage students to engage in learning about food safety. This is consistent with previous researches’ proposition that games can encourage learning (Connolly et al., 2012). Although this study found that most students did not give as high a score for enjoyment perception as they did for the learning perception, Iten and Petko (2016) found that students’ enjoyment perception in an educational game was not a major factor influencing their engagement with the game. Therefore, students’ enjoyment perception toward this game is acceptable, while providing more enjoyable gaming experience for players can be the goal for the continuous improvement of the game in the future.

Moreover, the analysis of the students’ gaming behavior indicates that most students performed online tests more than 100 times and repeatedly engaged in the investigation of detective tasks and reading learning contents within a short time, suggesting that this game could attract students’ active engagement and encourage them to learn food safety knowledge. Such results imply that this game can stimulate students’ learning motivation, which supports previous findings indicating that scientific inquiry games can stimulate student’s learning motivation (Barab et al., 2007a; Ketelhut et al., 2008; Rowe et al., 2009).

In addition, a comparison of those who successfully solved the game task revealed that the successful group of students exhibited more proactive gaming behaviors compared to their unsuccessful counterparts. The analysis of group comparison also indicated that students in the successful group gained more microbial food safety knowledge than those in the unsuccessful group did. Based on these findings, it may be inferred that if students participate proactively in the game, their chance of completing the game tasks will be higher, which also leads to better knowledge acquisition of microbial food safety. In other words, this game could be beneficial to students’ learning about food safety knowledge if they can proactively participate in the game.

Finally, this study conducted a sequential pattern mining through all students’ inquiry actions in the game. First, 55 frequent inquiry patterns were acquired, and then 10 interesting patterns were extracted from those 55 frequent patterns. After comparing the difference in the occurrence of 10 interesting patterns between students who solved the game and those who did not, the finding revealed that the occurrence of 10 interesting patterns in the successful group was higher and three patterns appeared significantly more often in the successful group.

Among these three behaviors, the notetaking behavior, following investigation or information collection, was particularly noteworthy. This implies that the students in successful group were good at using the notetaking function provided in the game to record clues. This also indicated the possible reasons why the unsuccessful group students were unable to solve the game. Thus, the notetaking function in the game can be enhanced in the future to facilitate students’ use of this function.

Conclusion

In sum, the computer detective game developed in this study enabled students to actively engage in self-directed learning of food safety knowledge and to have positive participation perceptions toward the game. Accordingly, this game should help high school students improve their microbial food safety knowledge. Therefore, the game could be a feasible alternative if schools are unable to allocate time to conduct food safety education in formal curricula, and this study provides a practical game mode that can combine inquiry-based and game-based learning in food safety education.

However, the main limitation of this study is the use of one group quasi-experimental design. Future studies should enhance the rigorousness of the experiments by using the experimental design with a control group, to verify the findings of this study. Moreover, the findings of this study also indicated that some students were unable to complete the game tasks independently and their learning effectiveness regarding the knowledge acquisition of microbial food safety was unsatisfactory. Although this could have been caused by their negative participation behaviors or different sequential inquiry patterns according to the analysis of gaming behaviors, this game should be improved in the future. For example, the game could be made more interesting and less difficult by changing its gaming mechanisms. Alternatively, more guidance could be provided, such as enhancing the notepad function to facilitate students’ use of the tool to conduct scientific inquiry. In addition, further research can be conducted on data mining of scientific inquiry behaviors. For example, using other algorithms to mine longer sequential patterns would yield more practical information from students’ inquiry behaviors.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was financially supported by the “Institute for Research Excellence in Learning Sciences” and “Chinese Language and Technology Center” of National Taiwan Normal University from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan, and sponsored by the Ministry of Science and Technology, Taiwan, under Grant No. MOST 106-2511-S-003-019-MY3, 107-2511-H-003-046-MY3, 106-2511-S-415-005, and 107-2511-H-415-008-MY2.

Author Biographies

Hsien-Sheng Hsiao is currently a professor in the Department of Technology Application and Human Resource Development, National Taiwan Normal University, Taiwan. His research interests include STEM education and technology education.

Fu-Hsing Tsai is currently a professor in the Teacher Education Center, National Chiayi University, Taiwan. His research interests include digital game-based learning, STEM education, and technology education.

I-Ying Hsu is currently a teacher and the section chief of Information & Media in the Affiliated Senior High School of National Kaohsiung Normal University, Taiwan. He received the PhD degree from the Department of Technology Application and Human Resource Development, National Taiwan Normal University, Taiwan. His research interests include e-learning, STEM education, and technology education.

References

Baker

R. S.

Clarke‐Midura

Ocumpaugh

(2016). Towards general models of effective science inquiry in virtual performance assessments. Journal of Computer Assisted Learning, 32(3), 267–280.

Barab

S. A.

Gresalfi

Ingram-Goble

(2010). Transformational play: Using games to position person, content, and context. Educational Researcher, 39(7), 525–536.

Barab

S. A.

Sadler

T. D.

Heiselt

Hickey

Zuiker

(2007a). Relating narrative, inquiry, and inscriptions: Supporting consequential play. Journal of Science Education and Technology, 16(1), 59–82.

Barab

S. A.

Zuiker

Warren

Hickey

Ingram-Goble

Kwon

E.-J.

Kouper

Herring

S.-C.

(2007b). Situationally embodied curriculum: Relating formalisms and contexts. Science Education, 91(5), 750–782.

Beffa-Negrini

P. A.

Cohen

N. L.

Laus

M. J.

McLandsborough

L. A.

(2007). Development and evaluation of an online, inquiry‐based food safety education program for secondary teachers and their students. Journal of Food Science Education, 6(4), 66–71.

Bazaldua, D. A. L., Baker, R. S., & San Pedro, M. O. Z. (2014). Combining expert and metric-based assessments of association rule interestingness. In Proceedings of the 7th International Conference on Educational Data Mining (pp. 44–51).

Carraway-Stage

Hovland

Showers

Díaz

Duffrin

M. W.

(2015). Food‐based science curriculum yields gains in nutrition knowledge. Journal of School Health, 85(4), 231–240.

Connolly

T. M.

Boyle

E. A.

MacArthur

Hainey

Boyle

J. M.

(2012). A systematic literature review of empirical evidence on computer games and serious games. Computers & Education, 59, 661–686.

Diplock

K. J.

Dubin

J. A.

Leatherdale

S. T.

Hammond

Jones-Bitton

Majowicz

S. E.

(2018). Observation of high school students’ food handling behaviors: Do they improve following a food safety education intervention? Journal of Food Protection, 81(6), 917–925.

10.

Diplock

K. J.

Jones-Bitton

Leatherdale

S. T.

Rebellato

Dubin

J. A.

Majowicz

S. E.

(2017). Over-confident and under-competent: Exploring the importance of food safety education specific to high school students. Environmental Health Review, 60(3), 65–72.

11.

Fenton

G. D.

LaBorde

L. F.

Radhakrishna

R. B.

Brown

J. L.

Cutter

C. N.

(2006). Comparison of knowledge and attitudes using computer‐based and face‐to‐face personal hygiene training methods in food processing facilities. Journal of Food Science Education, 5(3), 45–50.

12.

Ferk

C. C.

Calder

B. L.

Camire

M. E.

(2016). Assessing the food safety knowledge of university of Maine students. Journal of Food Science Education, 15(1), 14–22.

13.

Fischer

A. R.

Frewer

L. J.

(2008). Food‐safety practices in the domestic kitchen: Demographic, personality, and experiential determinants. Journal of Applied Social Psychology, 38(11), 2859–2884.

14.

Fournier-Viger

Gomariz

Gueniche

Soltani

C. W.

Tseng

V. S.

(2014). SPMF: A Java open-source pattern mining library. The Journal of Machine Learning Research, 15(1), 3389–3393.

15.

Furtak

E. M.

Seidel

Iverson

Briggs

D. C.

(2012). Experimental and quasi-experimental studies of inquiry-based science teaching a meta-analysis. Review of Educational Research, 82(3), 300–329.

16.

Haapala

Probart

(2004). Food safety knowledge, perceptions, and behaviors among middle school students. Journal of Nutrition Education and Behavior, 36(2), 71–76.

17.

Hickey

D. T.

Ingram-Goble

A. A.

Jameson

E. M.

(2009). Designing assessments and assessing designs in virtual educational environments. Journal of Science Education and Technology, 18(2), 187–208.

18.

Iten

Petko

(2016). Learning with serious games: Is fun playing the game a predictor of learning success? British Journal of Educational Technology, 47(1), 151–163.

19.

Ketelhut

D. J.

Dede

Clarke

Nelson

Bowman

(2008). Studying situated learning in a multi-user virtual environment. In E. Baker, J. Dickieson, W. Wulfeck, & H. O’Neil (Eds.), Assessment of problem solving using simulations (pp. 37–58). Erlbaum.

20.

Ketelhut

D. J.

Nelson

(2018, October 4–5). The role of immersive virtual environments in raising science self-efficacy. In M. Ciussi (Ed.), Proceedings of the 12th European Conference on Games Based Learning [Paper presentation]. SKEMA Business School, Sophia Antipolis, France (pp. 281–287). Conferences and Publishing International Limited.

21.

Lange

Göranzon

Marklinder

(2016). Self-reported food safety knowledge and behaviour among Home and Consumer Studies students. Food Control, 67, 265–272.

22.

Lee

S. Y.

Mott

B. W.

Lester

J. C.

(2010, June). Optimizing story-based learning: An investigation of student narrative profiles. In International conference on intelligent tutoring systems (pp. 155–165). Springer.

23.

Lynch

R. A.

Dale Steen

Pritchard

T. J.

Buzzell

P. R.

Pintauro

S. J.

(2008). Delivering food safety education to middle school students using a web‐based, interactive, multimedia, computer program. Journal of Food Science Education, 7(2), 35–42.

24.

Majowicz

S. E.

Hammond

Dubin

J. A.

Diplock

K. J.

Jones-Bitton

Rebellato

Leatherdale

S. T.

(2017). A longitudinal evaluation of food safety knowledge and attitudes among Ontario high school students following a food handler training program. Food Control, 76, 108–116.

25.

Mayer

A. B.

Harrison

J. A.

(2012). Safe eats: An evaluation of the use of social media for food safety education. Journal of Food Protection, 75(8), 1453–1463.

26.

National Research Council. (2013). Next generation science standards: For states, by states. The National Academies Press.

27.

Pei

Han

Mortazavi-Asl

Wang

Pinto

Chen

Hsu

M. C.

(2004). Mining sequential patterns by pattern-growth: The prefixspan approach. IEEE Transactions on Knowledge and Data Engineering, 16(11), 1424–1440.

28.

Pintauro

S. J.

Krahl

A. G.

Buzzell

P. R.

Chamberlain

V. M.

(2005). Food safety and regulation: Evaluation of an online multimedia course. Journal of Food Science Education, 4(4), 66–69.

29.

Pouillot

Lubran

M. B.

Cates

S. C.

Dennis

(2010). Estimating parametric distributions of storage time and temperature of ready-to-eat foods for US households. Journal of Food Protection, 73(2), 312–321.

30.

Rowe

J. P.

Lester

J. C.

(2015) Improving student problem solving in narrative-centered learning environments: A modular reinforcement learning framework. In Conati

Heffernan

Mitrovic

Verdejo

(Eds.), Artificial intelligence in education. AIED 2015. Lecture notes in computer science ( Vol. 9112). Springer.

31.

Rowe, J., Mott, B., McQuiggan, S., Robison, J., Lee, S., & Lester, J. (2009). Crystal island: A narrative-centered learning environment for eighth grade microbiology. In workshop on intelligent educational games at the 14th international conference on artificial intelligence in education (pp. 11–20). Brighton: UK.

32.

Rutten

Van Joolingen

W. R.

Van der Veen

J. T.

(2012). The learning effects of computer simulations in science education. Computers & Education, 58(1), 136–153.

33.

Scallan

Hoekstra

R. M.

Angulo

F. J.

Tauxe

R. V.

Widdowson

M. A.

Roy

S. L.

Jones

J. L.

Griffin

P. M.

(2011). Foodborne illness acquired in the United States—Major pathogens. Emerging Infectious Diseases, 17(1), 16–22.

34.

Shearer

A. E.

Snider

O. S.

Kniel

K. E.

(2013). Development, dissemination, and preimplementation evaluation of food safety educational materials for secondary education. Journal of Food Science Education, 12(2), 28–37.

35.

Shearer

A. E.

Snider

O. S.

Kniel

K. E.

(2014). Implementation and assessment of food safety educational materials for secondary and postsecondary education. Journal of Food Science Education, 13(1), 4–11.

36.

Stage

V. C.

Kolasa

K. M.

Diaz

S. R.

Duffrin

M. W.

(2018). Exploring the associations among nutrition, science, and mathematics knowledge for an integrative, food‐based curriculum. Journal of School Health, 88(1), 15–22.

37.

Tsai

F.-H.

(2018). Designing a computer detective game for science education. Proceedings of 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC), 1, 797–798.

38.

Wantland

D. J.

Portillo

C. J.

Holzemer

W. L.

Slaughter

McGhee

E. M.

(2004). The effectiveness of Web-based vs. non-Web-based interventions: A meta-analysis of behavioral change outcomes. Journal of Medical Internet Research, 6(4), e40.

39.

Young

Waddell

Harding

Greig

Mascarenhas

Sivaramalingam

Papadopoulos

(2015). A systematic review and meta-analysis of the effectiveness of food safety education interventions for consumers in developed countries. BMC Public Health, 15(1), 822.