PRINCIPAL COMPONENT ANALYSIS ON DATA OF GEOGRAPHICAL AND NON GEOGRAPHIC FACTORS THAT AFFECT THE ACCESS TO HEALTH SERVICES IN INDONESIA

• Main Author: Anggarawan, Muhung
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/46372
• Institution: Institut Teknologi Bandung

There are many phenomena that have many variables today, so we need a method to process data from these phenomena. One statistical method that can analyze many variables is Principal Component Analysis (PCA). PCA can be used to reduce the number of variables without reducing meaningful information from data. The new variables are called the principal component. In general, the principal components obtained are expected to be two or three so it is easy to visualize. There are many application of PCA, the example is in the health sector. PCA can analyze the closeness between provinces in Indonesia based on the access to health services. Health is a human right and an element of community welfare. Thus, improving the quality of access to health services is very important. There are twenty factors that affect access to health services. Nine geographical factors and eleven non-geographical factors were analyzed separately. Geographical data has the best result for the three principal components with 79.4% absorption of information. While non-geographical data has the best result for the two principal components with 68% absorption of information. In addition, provinces in Indonesia can be grouped according to their closeness according to the PCA results for geographic data and non-geographical data using a cluster method called k-means. For geographical data, the result is five groups and for non-geographical data the result is three groups with low, middle and high levels. Provinces with low levels are Papua, West Papua, Maluku, North Maluku, West Kalimantan, Central Kalimantan and North Kalimantan which the access to health services should be improved immediately.