Spatial Variations in Fertility of South Korea: A Geographically Weighted Regression Approach

Myunggu Jung,Yeohee Choi,Woorim Ko,Youngtae Cho

doi:10.3390/ijgi8060262

Abstract

South Korea has witnessed a remarkable decline in birth rates in the last few decades. Although there has been a large volume of literature exploring the determinants of low fertility in South Korea, studies on spatial variations in fertility are scarce. This study compares the Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) models to investigate the potential role of the spatially heterogeneous response of the total fertility rate (TFR) to sociodemographic factors. The study finds that the relationships between sociodemographic factors and TFRs in South Korea vary across 252 sub-administrative areas in terms of both magnitude and direction. This study therefore demonstrates the value of using spatial analysis for providing evidence-based local-population policy options in pursuit of a fertility rebound in South Korea.

Highlights

In the last few decades, South Korea has witnessed a remarkable decline in birth rates
Four study areas had total fertility rate (TFR) higher than the replacement level (TFR 2.1), and they were mostly found in the southwestern part of South Korea, South Jeolla Province
Various studies of fertility in South Korea have explored geographical variations in fertility based on rural and urban settings [14,15], this study shows that responses of local-level total fertility rates to sociodemographic factors spatially vary beyond the crude rural and urban dichotomy

Summary

Introduction

In the last few decades, South Korea has witnessed a remarkable decline in birth rates. Results of previous studies have varied in terms of the major driving forces and the role of specific factors for fertility declines [6,7,8,9,10]. Instead, it suggests that the fertility-decline process in South Korea is combined with local contexts in which similar sociodemographic factors may result in different fertility outcomes at different places

Methods

Results

Discussion

Conclusion