Takafumi Kirimura

看護基礎教育における数理・データサイエンス教育のミニマム・エッセンシャルズを検討するなかで、看護学分野においては、現在の大部分の研究アプローチが従来型の仮説ドリブン型(仮説駆動形)アプローチであるのに対し、機械学習(AI)などを用いた数理・データサイエンス手法はデータドリブン型(データ駆動型)アプローチであることを理解しておくことが、看護職として社会に出た際、数理・データサイエンスを用いるときに重要となることを先の論文で言及した。本論文では、仮説ドリブン型アプローチとデータドリブン型アプローチは、互いに独立に存在するのではなく、両者は表裏一体な関係にあり、なおかつこの両アプローチの関係は、数理・データサイエンス教育の哲学的な背景となっていることから、看護学における方法論間の信念対立に陥らないためにも、データドリブン型アプローチがミニマム・エッセンシャルズとして必要不可欠であることを詳述する。|Through this study, we came to the conclusion that the hypothesis-driven approach and the data-driven approach are not independent of each other, but are inextricably linked. This relationship is essential to basic nursing education as it is the philosophical basis of mathematical and data science education. The linkage of these two approaches is also vital to avoid belief conflicts between nursing methodologies.

Recently, we proposed "Eduinformatics,"a new field of education that combines both education and informatics. In addition, we introduced new criteria to utilize student data in Institutional Research (IR). In a previous article, we defined "primary data"as the first standard which is not combined linear data and "secondary data"as the second standard which is a linear combination of primary data. However, in this article we will present new definitions of Primary and Secondary data because our analysis of actual educational data has revealed that Secondary data is not only linear data, but also nonlinear. Moreover, we will present examples in which primary data was used to detect elements that could not be founded through the analysis of secondary data, and were pitfalls of compared analysis performed by IR practitioners.

In recent years, the word "evidence" has come to be commonly used in the public debate over education in Japan. In this paper, in relation to the question of what kind of "evidence" should be required, and in what kind of content should "evidence" be praised in the federal education policy of the United States, the cases of the No Child Left Behind Act of 2001 (NCLB), Comprehensive School Reform (CSR), Investing in Innovation Fund (i3), and Every Student Succeeds Act (ESSA) were examined, and the transition of the policies were overviewed. NCLB gave the definition of "scientifically based research, " and practices and programs called for scientific support by providing evidence on its effectiveness. CSR tried to embody the concept of its definition, where a lower standard called "strong evidence" was prepared separately from "scientifically based research" from a realistic point of view. Still CSR tried to maintain the hierarchical order of evidence with the randomized controlled trial at the top. This feature is also seen in subsequent policies, and can also be seen in the i3 founded in 2009. However, the Every Student Succeeds Act, established in 2015, granted substantial discretion to state and local education authorities in the context of the use of evidence. Moreover, it is said that only the lowest-rated evidence should be presented for grants other than Title I. Therefore, it can be said that the policy of "evidence" in the federal education policy of the United States has undergone a major change.