Point Biserial Correlation Coefficient on Climate Variables and Dengue Cases Using R Programming

• Published in: Springer Proceedings in Mathematics and Statistics
• Main Author: Zukarnain Z.A.; Muhamad Krishnan N.F.; Jamaludin M.; Rahman N.A.; Ahmad A.
• Format: Conference paper
• Language: English
• Published: Springer 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85209586878&doi=10.1007%2f978-981-97-3450-4_16&partnerID=40&md5=a54f9561fc724c66956973f1b3706a91

The number of people infected with dengue fever is on the increase across the world. Dengue fever is present in urban and semi-urban settings, and rural areas are also affected in certain nations. Dengue fever is affected by rain, relative humidity, temperature, and unplanned fast urbanization. This study focuses on Kota Bharu, aiming to identify whether the climate characteristics, including average temperature, mean relative humidity, and total rainfall, affect dengue cases. This study employed a point biserial correlation coefficient to see if the features correspond to the output. R programming was applied to check whether there was a correlation between dengue cases (yes/no) and climate parameters (average temperature, mean relative humidity, and rainfall). Point biserial correlation was used as the target variable for dichotomous variables. The methodology involved several steps, including data pre-processing, cleaning, and analysis. According to the findings, only mean relative humidity correlates with dengue cases in Kota Bharu. Since there is a negative correlation, dengue fever rises with low humidity. However, different regions might give different results of the correlation. Understanding the factors that lead to a rise in dengue cases and education initiatives can assist in enhancing a region’s early warning system. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.