Extraction of scientific semantic networks from science textbooks and comparison with science teachers’ spoken language by text network analysis

Just as language reflects one’s thoughts, the text of science textbooks reflects the structure of scientific knowledge and thought. Therefore, students’ learning of scientific language leads to their acquisition of the structure of scientific knowledge and thought. The purposes of this study were to extract scientific semantic network from science textbooks as the ideal model of using scientific language, and to examine how well science teachers are implementing scientific semantic networks in science classes by comparing and analysing their spoken language. Through a recursive method, science textbooks were searched for keywords and their relationships, and consequently a scientific semantic network was presented. As a result of the comparison with science teachers’ spoken language, it was found that the teachers were not able to implement the scientific semantic network exactly and efficiently in class, and the focus of the contents of their instruction was also different from the textbook. Science textbook was focused on qualitative aspects of the concept such as meaning, element, characteristic etc, on the other hand science teachers’ language was focused on quantitative aspect of the concept such as calculation, graph etc.