Classifıcation of OECD Countries According to Health Data With Clustering Analysis

Authors : Mustafa Filiz, Olkan Budak

Pages : 1-15

View : 46 | Download : 48

Publication Date : 2024-01-14

Article Type : Research Paper

Abstract :This study was carried out to determine how OECD countries are clustered according to the determined health data, which ones are similar, and which countries are better. 36 countries were included in the study and 10 variables, which are among the health indicators of the countries, were used. Centroid tree graph and k-means clustering analysis, one of the non-hierarchical clustering analysis methods, were used to analyze the data. With the ANOVA test, the differences in the variables according to the clusters were determined. It was observed that seven clusters were formed in the centroid method. As a result of the K-mean clustering analysis, it was seen that the distance from the selected countries was the USA the least and Turkey the most. It has been seen that among the variables selected in the clustering of OECD countries under seven clusters, variables such as life expectancy at birth, infant mortality rate, per capita health expenditure, Gini coefficient, crude death rate, the share of health in GDP, and the number of nurses/midwives play an important role. It was concluded that the countries in the 1st cluster had the best values in terms of health indicators of 36 countries, and the countries in the 5th cluster had the worst values. In addition, as a result of the ANOVA test, it was decided that other health indicators other than maternal mortality rate, number of patient beds, and number of physicians play an important role in clustering OECD countries under seven clusters.
Keywords : Health, Health indicators, K mean, OECD, Tree graph