Abstract: Focusing on the Yinchuan Plain, we conducted field measurements of soil salinity at sampling points and selected nine factors related to soil salinity. By combining spatial autocorrelation analysis and the multiscale geographically weighted regression (MGWR) model, we investigated the spatial distribution characteristics of soil salinity and the scale and spatial variation of influencing factors. The results showed that the Moran's I index of soil salinity in the Yinchuan Plain was 0.314, indicating significant positive spatial autocorrelation and spatial clustering, with a general pattern of higher salinity in the north and lower salinity in the south. The MGWR model was suitable for studying the influencing factors of soil salinity in the Yinchuan Plain. In terms of scale, groundwater depth, enhanced vegetation index (EVI), and soil moisture bandwidth had the largest scale of influence at 165, followed by elevation, soil pH, slope, surface temperature, and groundwater mineralization, while land use intensity had the smallest scale of influence at 43. In terms of the effects, groundwater mineralization and soil pH had a positive effect on soil salinity, while land use intensity, groundwater depth, EVI, and elevation had a negative effect. Surface temperature had a predominantly positive effect, accounting for 92.7% of the total samples, while soil moisture and slope had insignificant effects. Among the significant influencing factors, the average regression coefficient for groundwater mineralization was 0.384, making it the most significant factor affecting soil salinity. Land use intensity, groundwater depth, EVI, surface temperature, and elevation followed in importance, while the average regression coefficient for soil pH was 0.063, indicating the smallest impact. Each factor exhibited varying degrees of spatial heterogeneity in its influence on soil salinity.