Utilize imagery and crowdsourced data on spatial employment modelling

Abstract

Background: Spatial employment modeling investigates employment distribution, patterns, influencing factors, neighboring area impact, and regional policy efficacy. Conventional studies often rely on traditional data sources, which may overlook critical employment-related phenomena. In 2022, Java recorded the lowest labor absorption rate in Indonesia, necessitating a new approach. Aim: This study combines imagery, crowdsourced data, and official statistics to identify factors influencing labor absorption in Java Island. Method: Geographically Weighted Regression (GWR) was employed to account for spatial effects in the data. Results: The model reveals that nighttime light intensity in urban and agricultural areas, along with environmental quality, significantly enhances labor absorption across Java. Internet facilities, universities, and the number of micro and small industries also positively influence most districts/cities. Conclusion: Incorporating new data sources offers valuable insights for understanding employment patterns and can enrich employment research frameworks.

Keywords

• spatial
• geographically weighted regression (gwr)
• crowdsourced data
• imagery data
• employment