Earticle

본 연구는 ODA 정책을 수립하고 결정하는데 있어서 원조를 받는 수원국들의 특성을 분류하여 그 특성에 맞는 중장기 전략을 수립하는 것이 필요하다는 전제하에 국가분류를 체계적으로 하기 위한 방법론을 개발하는 것을 1차적 목적으로 한다. 더 나아가 본 연구는 분류된 개발도상국가들에 대하여 국가의 특성을 고려한 ODA 정책방향을 제시하는 것을 2차적 목적으로 한다. 이 연구를 위하여 선진 국과 개발도상국을 포함한 166개 국가들에 대하여 SDGs 점수와 1인당 GDP를 이용하여 머신러닝 분 석방법에 해당하는 신경망 분석과 의사결정나무 분석을 실시하였다. 그 결과 166개 국가들은 모두 12 개의 그룹으로 분류되었다. 개발도상국에 해당되는 노드는 모두 7개로서 이들 노드에 포함되는 국가 그룹에 대하여 개별 국가 그룹별로 ODA 전략을 제시하였다.

This study primarily aims to develop a systematic methodology for classifying countries, based on the premise that it is necessary to establish mid- to long-term strategies tailored to the characteristics of recipient countries when formulating and deciding ODA policies. Furthermore, the secondary objective of this research is to propose ODA policy directions that consider the specific characteristics of the classified developing countries. For this study, neural network analysis and decision tree analysis, which are machine learning analytical methods, were conducted using SDG scores and GDP per capita for 166 countries, including both developed and developing nations. As a result, the 166 countries were classified into 12 groups. Seven nodes corresponded to developing countries, and ODA strategies were proposed for each individual country group included in these nodes.

(Purpose) The purpose of this study is to enhance the effectiveness of Official Development Assistance (ODA) policies by systematically classifying recipient countries and establishing tailored ODA policies based on these classifications. Specifically, the study aims to identify an appropriate methodology for categorizing developing countries using specific criteria, and to apply it to actual data, providing valuable insights for donor countries in formulating future ODA policies. (Design/methodology/approach) The data used in this study are the basic statistics on the Sustainable Development Goals (SDGs) published annually in the SDGs Report. The analytical method employed is decision tree analysis. (Findings) The results indicate that the 167 countries analyzed were classified into 10 distinct nodes. The study further limited the scope to the five nodes representing the most disadvantaged developing countries and suggested future directions for aid policies for each of these nodes. (Research implications and Originality) This study emphasizes the need for systematic classification of recipient countries and the establishment of tailored ODA policies based on these classifications, in a context where the scale and importance of ODA continue to grow. The research's significance lies in its presentation of a practical methodology that can assist donor countries in formulating ODA policies by classifying developing countries through decision tree analysis and proposing specific policy directions for the most vulnerable countries.