With the continual growth in global trade and the emergence of new food products, food hygiene and safety issues are becoming increasingly complex, with food safety receiving enormous international attention (Ministry of Health and Welfare, 2016). The World Health Organization (WHO) reported that eating unsafe substances can cause more than 200 diseases, including diarrhea and serious illnesses, such as cancer. Therefore, acquiring the necessary food safety literacy to distinguish between food safety and nutrition is a crucial competency to improve personal health. The Literacy Information and Communication System conducted a health literacy workshop in 2011, involving approximately 1,600 national and international public health and medical experts; it recommended establishing a hypothetical relationship between knowledge, attitudes, and behavior to create a new set of measurement methods for food safety literacy (Pleasant et al., 2011). Thus, integrating knowledge, attitude, and behavior is crucial in order to assess "food security literacy" in the 21st century (Frewer et al., 2013; Jevsnik et al., 2008; Rollin et al., 2011; Yeung & Morris, 2001). In recent years, there has been an international emphasis on literacy assessment, such as the Programme International Student Assessment, which was supported by the Organisation for Economic Co-operation and Development in 2013, using contextual-based tests to assess whether students can apply knowledge and analyze and solve problems in everyday life. First, this study aimed to develop effective food safety literacy tests, food security attitude scales, and food safety behavior scales. Second, we explored the effect of Evidence-Informed Practice (EIP) in teaching interventions on junior high school students' food safety literacy. During our pilot study, we randomly selected 467 ninth-graders from southern Taiwan to respond to research instruments regarding food safety literacy, food security attitudes, and food safety behaviors. We used the content validity index (Lawshe, 1975), item analysis, and KR 20 to examine the tests' validity, discrimination, reliability, respectively, and applied item response theory to determine a suitable model of the relationship between the tests and participants. We found that all the tests met the index of validity and reliability. Then, we conducted a quasi-experimental study design and selected 104 ninth-graders from a public junior high school in Kaohsiung City through convenience sampling; the school is known for its moral education, encouraging teachers to develop innovative teaching. This study lasted six weeks and included two teachers with similar teaching experiences who implemented different teaching styles. Both teachers had majored in health education and had 17 years of teaching experience. Teacher A was the first author of this article, graduated from the Department of Science and Education of Southern Normal University, taught two classes of students in the experimental group. The teacher B, who graduated from the Department of Health Promotion and Health Education of the Northern Normal University, was familiar with health education theory, and taught two classes of students in the comparison group. To ensure a high response rate, the first investigator informed the students of this study's purpose and research tools, mentioning that the contents of the questionnaire would be properly preserved and not be used for individual comparison. Students agreed to participate in the study and responded to two scales and a food safety literacy test that took approximately 15-20 minutes, after which they were given a small gift. Four classes participated in this study, two of which comprised the comparison group that received traditional narrative teaching; the teaching content and time were the same for this group. The other two classes in the experimental group received an EIP teaching intervention. The experimental group students ( n = 53) participated in a 6-week one period of EIP teaching model; with the same teaching intervention time, the comparison group students ( n = 51) received a teacher-centered intervention concerning new trends in dietary consumption; the intervention convened 45 minutes per week for 6 consecutive weeks. The EIP intervention involved students asking questions, thinking, gathering evidence, and making judgments. The teacher in this group questioned students to promote their thinking, enabling them to search for empirical information to support the discussion and interpretation of food security issues, such as false advertising, supplementing them with new scientific findings. Meanwhile, the comparison group was taught about food security issues in a traditional manner, with the teacher presenting and students listening. We conducted descriptive statistics, independent t-tests, and paired sample t-tests to compare the students' similarities and differences on their food safety literacy tests, food security attitude scales, and food safety behavior scales after the interventions. In addition, we used a one-way analysis of covariance to examine the effects of the EIP intervention. The results revealed that junior high school students' beliefs concerning maintaining healthy behaviors involved medium and high scores. They believed that a balanced diet, regular exercise, and good hygiene could increase the body's resistance to disease. More than 80% of the students consumed healthy food, and nearly 40% of these students consumed more than two types of healthy food. Students with a higher belief in the value of maintaining health had a significantly higher score in food safety literacy. However, those students with a frequent intake of healthy food tended to have lower scores on their food safety literacy. This study developed and validated well-structured food safety literacy scales and an effective EIP teaching model that enhanced the experimental group students' food safety literacy. This empirical evidence provides fresh insight into how to promote junior high school students' food safety literacy. Our findings have implications for implementing the EIP teaching model and administering effective food safety literacy scales with context-based items in more educational settings.
