A total of 110 students (4^th, 5^th, and 6^th grades) in three elementary schools in Jeollanam-do Province were recruited and randomly distributed into two groups: lecture- and experience-based nutritional education groups. There was no significant difference between the two groups in distribution of gender or grade (Table 1).

Lecture- and experience-based educational programs shared identical contents with respect to dietary knowledge, eating behaviors, and dietary habits but with differential approaches (lecture- and experience-based). These methods were reviewed by three experts in nutritional education. To examine the impacts of these two educational methods, students' dietary knowledge, dietary behaviors, and dietary habits were analyzed using a pre/post-test. The pretest was taken on June 20, 2012, which was 1 week before the programs started. To evaluate the effects of the educational program, a follow-up survey was conducted 12 weeks after the programs finished.

Both lecture- and experience-based educational programs consisted of five sessions: 'Learn my eating behavior', 'Eat breakfast', 'Choose healthy snacks', 'How to wash hands properly', and 'How to prevent obesity'. Each lecture-based session took 20 minutes. The experience-based sessions were administered to five groups of between 4-6 students. Each experience-based session took 20 minutes, and each topic was rotated from one group to another in every session. Experience-based educational program included action-learning activities such as role-playing and games without teacher-centered contents. The educational contents of each session were as follows.

In the 'Learn my eating behavior' session, body composition assessment and obesity test were performed for each participating student 1 week before the program started. The results of tests were distributed to the students before the programs started. The 24-hour recall method was used to design one meal sample for every student, and the meals were analyzed using the Computer Aided Nutritional Analysis Program (CAN-PRO ver. 4.0, Seoul, 2011). This analysis and the two previous test results were used for one-to-one counseling with every student. In the counseling session, the sample meal of each student was compared to the dietary recommended intake for upper grade elementary students in Korea in order to help the student revise his or her own meal.

In the 'Eat breakfast' session, students were taught the pros of eating breakfast and cons of skipping breakfast. Each student designed his or her own breakfast with the Food Balance Wheels for one meal. Students were encouraged to discuss and help each other edit and revise excessive or insufficient nutrients in their meals.

The 'Choose healthy snacks' session covered sub-topics such the definition of fast food, pros and cons of snacks, and how to choose healthy snacks. After the session, students practiced what they had learned through a fishing game by separating given images of foods into healthy and unhealthy groups. Healthy ones were marked with green and unhealthy with red. Further, students learned about healthy traditional Korean foods that can be substituted for unhealthy foods they used to consume.

In the 'How to wash hands properly' session, students watched an instructional video provided by the Ministry of Food and Drug Safety focused on the need to wash hands, when to wash hands, and how to wash hands properly. After the session, students practiced what they learned in a fluorescence dye experiment. Specifically, students placed their hands in a fluorescent dye and then checked their hands in a fluorescence view box. After washing their hands with soap, they checked their hands again in the view box. The instructor corrected the hand-washing habits of students whose hands retained fluorescence residue.

The 'How to prevent obesity' session included topics such as the cause of obesity, preventive methods against obesity, and diseases and illnesses caused by obesity. The students participated in the Fat vest experience, which simulates the 'look' and 'feel' of excess body fat. Each student wore a Fat vest and engaged in activities such as hula-hooping, jumping rope, foot bag juggling, and body flexibility tests in order to experience the physical discomfort associated with obesity. Further, students learned the importance of proper diet and cardiovascular activities in preventing obesity.

Educational contents and experienced items in five sessions are summarized in Table 2.

To assess the effect of each nutritional education method, a survey was administered as a measurement technique. The survey questionnaires were designed in collaboration with the Jeolla-Namdo Department of Health Promotion and Administration based on previous research by Lee [5] and Jin [4]. The questionnaires covered topics such as dietary knowledge, dietary behaviors, and dietary habits. The questionnaire consisted of 10 categories and 51 questions.

Dietary knowledge: This section consisted of 10 questions testing students' knowledge of balanced meals, relationship between breakfast and body composition, definition of snacks, preventive dietary behaviors for obesity, essential foods for body composition and immunity function, empty-calorie foods, calcium-rich foods, traditional seasonal foods, and traditional fermented foods. A correct answer was given 1 point; a wrong one, 0.

Dietary behaviors: This section consisted of 11 questions examining students' compliance for dietary guidelines using questions. The subjects were asked to give their frequency of consuming healthy foods such as vegetables, daily products, meat, fish, egg, soybeans, tofu, and fruits as well as that of indulgence foods, snacks, soft drinks, and fast food, in addition to general eating behaviors such as salt contents, eating rate, regular daily meal schedules, and overeating.

The points were scaled based on students' frequencies: 5 points for 6 or 7 times a week, 3 points for between 3-5 times, and 1 point for 0-2 times. Cronbach's α of the questionnaires for dietary behavior was .72.

This section consisted of 13 questions examining the dietary habits of subjects in order to determine the daily routine of dietary behaviors. The questions focused on the following subjects: eating breakfast regularly, eating balanced meals, consuming snacks, exercising, considering losing weight, checking nutrition facts before purchasing food, shaping proper dietary behaviors, eating disliked but healthy foods, and eating foods with gratitude and pleasure. Again, points were scaled based on students' frequencies of performing proper dietary habits: 5 points for 6 or 7 times a week and 1 point for none. For the 'meals taste better than snacks' question, points were given based on students' answers: 5 points for 'Always' and 1 point for 'Never'. For 'Wash hands after using the bathroom' question, 5 points were given to 'more than 10 seconds' and 1 point for 'do not wash hands'. For using chopsticks, 5 points were given for proper usage and 1 point for improper usage. The questions on negative dietary habits, including 'Complain of no favorite food in meal', 'Eat snacks more often', and 'Prefer snacks to meals', were inversely scored. Cronbach's α of the questionnaires for dietary behavior was .68.

Statistical analysis was performed using IBM SPSS statistics 21 (IBM corporation, Somers, NY, USA). Categorical data are presented as percentages (numbers) and continuous data are presented as mean ± SD. Proportions were compared by Fisher's exact test. The differences between the values before and after each nutritional education method were tested by a paired t-test. P < 0.05 was considered statistically significant.

Changes in dietary knowledge in response to educational methods are presented in Table 3. In the lecture-based group, average scores for dietary knowledge did not significantly differ before/after education. However, knowledge regarding 'What is a balanced meal?' (P < 0.01), 'Which of three meals provide the most energy for the brain and body?' (P < 0.01), and 'What is a traditional food with high contents of helpful lactobacillus and dietary fiber that prevents constipation?' (P < 0.05) significantly changed after education using the lecture-based method. The experience-based group did not show significant changes in total scores for knowledge (from 7.55 to 7.51). However, knowledge relating to 'Which of the choices are empty-calorie junk food?' significantly decreased (P < 0.05). These findings suggest that the lecture-based method was more effective in improving dietary knowledge compared to the experience method.

Table 4 presents changes in dietary behaviors caused by the two educational methods. In the experience-based group, total score for dietary behaviors significantly improved from 3.29 to 3.59 (P < 0.05). The score for 'Eat soybeans or tofu everyday' increased after education (P < 0.05). In contrast, the score for dietary behaviors did not change in the lecture-based group, although the score increased for 'Eat or drink less snacks, soft drinks, and fast food' (P < 0.05). The result indicates that the experienced-based method may effectively improve children's dietary behaviors.

Table 5 presents changes in dietary habits by the two educational methods. The lecture-based group showed a significantly increased total score for dietary habits from 3.50 to 3.74 (P < 0.05). The scores for questions such as 'Complain of no favorite food in meal' and 'Eat disliked but healthy foods for health' significantly increased after lecture-based education (P < 0.05). The total scores of the experience-based group did not change. Scores for 'Check nutrition facts before purchasing food' and 'Eat disliked but healthy foods for health' significantly increased in the experienced-based group (P < 0.05 and 0.01). Overall, the lecture-based method had obvious effects on students' dietary habits.