With the advent of the 4th industrial revolution era, the explosive increase in knowledge and information demands a higher level of competency in solving complex problems and applying what has been learned in the context of life rather than acquiring ...
With the advent of the 4th industrial revolution era, the explosive increase in knowledge and information demands a higher level of competency in solving complex problems and applying what has been learned in the context of life rather than acquiring fragmentary knowledge. In accordance with these changes, in the past, the ability to memorize a lot of knowledge and solve a given problem was valued, but now, data-driven problem solving capabilities that can make rational decisions based on data are required.
Data-driven problem solving ability can be approached in two aspects: 'problem solving ability' and 'data-driven'. And ‘mathematics’ and ‘data science’ are highly valued as central fields for cultivating data-driven problem solving skills. The importance of mathematics and data science can also be seen in the 2022 revised curriculum. However, mathematics and data science learning in current school education has limitations in serving as a central subject for cultivating data-driven problem solving skills. The reason for this is, firstly, that although the possibility of using real-life data in education has increased, there are almost no cases in which data in real contexts are used in mathematics problem solving learning. Second, most data science education in school education tends to be conducted as a separate education, separated from general subjects, and is mainly focused on fostering computational thinking skills. Data-driven mathematics education can be considered as a way to overcome the limitations of mathematics and data science learning in school education and effectively develop data-driven problem solving skills.
Research on data-driven education is continuously being conducted. However, most examples focus on data science itself. And rather than utilizing data within the curriculum context, the collection and analysis of data itself tends to be regarded as the purpose of instruction. In addition, looking at the cases so far, most of the studies on data-driven learning have been conducted in the context of secondary or higher education, and there is a lack of research on cases of data-driven learning and their effectiveness in the context of elementary education. Moreover, despite the emerging importance of mathematics and data science, there are few examples of mathematics education based on data science. Therefore, there is a need for research on data-driven mathematics education that can collect, analyze, process, and evaluate data in a practical context linked to mathematics subjects in the context of elementary education and make rational decisions based on data.
Meanwhile, for effective data-driven convergence teaching and learning, teachers' systematic and prescriptive instructional design efforts are required. However, studies including systematic instructional design for data-driven learning are lacking. In addition, studies on data-driven instructional models and teaching strategies that instructors should take to conduct data-driven teaching and learning are insufficient. In particular, in the context of elementary education, it is necessary to conduct research on data-driven instructional models and teaching strategies that consider the characteristics of subjects.
Therefore, this study developed a instructional model including teaching strategies that can be used to effectively plan and conduct data-driven problem solving classes in elementary school mathematics, and confirmed its internal and external validity. The research questions of this study are as follows. First, what are the data-driven problem solving instructional model in elementary school mathematics? Second, is the data-driven problem solving instructional model in elementary school mathematics internally valid? Third, is the data-driven problem solving instructional model in elementary school mathematics externally valid?
To answer these research questions, design and development research methods were used. According to this method, model development, internal validation of the model, and external validation of the model were conducted in the order. First, in the model development stage, an initial instructional model and teaching strategy for data-driven problem solving classes in elementary school mathematics were derived through prior literature review and expert interviews. Secondly, internal validation, opinions were obtained from experts through two expert reviews. Based on the expert validation opinion, the 3rd instructional model and instructional strategy were derived. Finally, in the external validation stage, the external validity was verified by applying the modified and supplemented instructional model and teaching strategy to the field. External validation was conducted through pre/post-tests of the teacher's and learner's response to the class to which the class model was applied, the learner's data science-based problem solving skill, attitude toward learning mathematics, and data literacy. Based on the results, The final instructional model and teaching strategy were derived.
The finally developed data-driven problem solving instructional model in elementary school mathematics is largely divided into 1)before class, 2)during class, 3)after class, and 4)overall class based on the execution time of class stage. The model consists of eight steps; ‘Preparation', ‘Guidance', ‘Support for understanding problems and data', ‘Support for data-driven problem solving', ‘Support for deriving results & Sharing', ‘Organization & Expansion of thinking', ‘Suggestion of additional data', ‘Support for reflection & Feedback'. The final teaching strategy consists of 27 teaching strategies and 62 detailed strategies that support each step of the instructional model.
The conclusions of this study are as follows. First, the instructional model developed in this study has significance in terms of systematic design and utilization of data-driven problem solving classes in elementary school mathematics. Second, it has significance as a case of data-driven teaching and learning in elementary school mathematics. Third, the instructional model developed in this study can have a significant impact on the improvement of learners' data-driven problem solving ability. Fourth, the instructional model developed in this study can have a significant effect on the improvement of learners' attitudes toward learning mathematics. Fifth, the instructional model developed in this study can have a significant effect on the improvement of learners' data literacy. Sixth, the design and implementation of mathematics classes for data-driven problem solving learning requires environmental conditions.
Based on the limitations of this study, suggestions for follow-up studies are as follows. First, it is necessary to conduct additional research targeting various grade groups in elementary schools or expand the research to middle and high schools. Through this, it is possible to seek refinement and generalization of instructional models and teaching strategies. Second, the instructional model of this study can be developed into a cross-curricular convergence instructional model that includes multiple subjects, not limited to one subject. Third, it is necessary to apply the model of this study from a long-term perspective and confirm its effect. Fourth, it is necessary to conduct a rigorous experimental design to verify the effectiveness of the instructional model of this study. Fifth, it is necessary to identify the cause of the affective area where no significant results were found. Sixth, it is necessary to prepare various data that can be used for education, and to develop and spread data-driven materials.