Unveiling the spatiotemporal dynamics and determinants of public complaints: a multiscale geographically weighted negative binomial analysis

Amidst China’s vigorous economic growth and rapid urbanization, the public’s aspiration for an enhanced quality of life has become increasingly evident. This has led to elevated standards for urban services, simultaneously giving rise to a series of urban issues and a surge in public complaints. Using the 12345 hotline complaint data from Fujian Province in 2022, this research applies Multiscale Geographically Weighted Negative Binomial Regression (MGWNB) to explore the spatiotemporal dynamics and determinants of public complaints. The statistical analysis reveals that noise pollution and labor rights issues are the primary sources of public dissatisfaction. The spatiotemporal analysis demonstrates that the volume of complaints follows a seasonal pattern, with higher numbers in summer and autumn, and the southeastern coastal areas of Fujian consistently exhibit a high number of complaints and complaint rates. The MGWNB model indicates that the annual average temperature, PM 2.5 levels, illiteracy rate, and per capita GDP are positively correlated with complaints. Conversely, the number of schools and hospitals per capita, the proportion of the secondary industry, impervious surface area, CO₂ emissions, the proportion of the elderly population, and per capita public fiscal expenditure generally show a negative correlation. Moreover, the MGWNB model further uncovers significant multiscale spatial heterogeneity in the impacts of variables such as the intercept, the number of schools and hospitals per capita, temperature, proportion of the secondary industry, and PM 2.5 concentration. Finally, this paper proposes a framework for interpreting public complaints.