High-resolution electrical resistivity data of the subsurface convey significant information about permafrost constituents and characteristics. The constituents of frozen ground comprise four phases and exhibit complex inherent correlations. In the Arctic Coastal Plain, the presence of salinity, clay minerals, and organics further complicates the analysis of permafrost constituents. The objective of this study is to develop correlations between electrical resistivity (ER) and permafrost attributes and to improve the understanding of permafrost heterogeneity. There are two challenges in the ER data interpretation: the interpretation of permafrost attributes by solely relying on ER data can lead to inaccurate conclusions; mismatch between coarse resolution of borehole data and high-resolution ER profiles limits the ability to accurately inverse the permafrost attributes. In this study, an uncertainty-based approach was presented to analyze the attributes and spatial patterns of shallow permafrost. This study developed a physical model that considers the effects of unfrozen water content, porosity, and temperature on the ER of soils and captures the impact of salt concentration and particle surface interactions (between molecules and electric charges on soil particles) on the freeze-thaw behavior of soils. Hierarchy Bayesian modeling was used to integrate prior laboratory observations and explore plausible parameter sets for target sites. Monte Carlo simulations were employed to map the spatial distributions of permafrost attributes based on electrical resistivity tomography and temperature data collected from a site in Utqiaġvik, Alaska. The analytical results were validated based on the shear wave velocity profile and borehole data at the same site.