The soil environment imposes a great influence on human health. Soil heavy metal pollution caused by human activities is an important part of environmental problems in urban areas. Due to an inadequate infrastructure, imperfect management, and intensive human activities, the sources of heavy metals in urban fringe areas are often more complicated than those in other areas, such as mining areas and agricultural irrigation areas. To solve this problem, the first step is to locate the source of pollution. However, the traditional methods of source analysis, such as principal component analysis and positive matrix factorization, always require correlations between elements. This study examined the Hg, Cd, Pb, and Cu contents in the Fengdong District of Xi’an, China, and found that these elements are not correlated in this area. Hence, traditional source analysis methods are not applicable in the study area. In response to this problem, this research proposed a new source analysis method based on Pearson’s correlation analysis. The Nemerow index, geoaccumulation index, and ecological risk index were adopted to evaluate soil heavy metal pollution in the study area. Via comparison to the actual situation, it was concluded that the geoaccumulation index is more suitable for source analysis in this area. Through Pearson’s correlation analysis, it was found that the geoaccumulation index is significantly correlated with the various land use types. Among them, transportation land exerted a greater impact on Pb pollution, and industrial land exerted a significant impact on the Hg distribution. The Cu distribution was related to construction land, while the Cd distribution was mainly related to urban land and cultivated land. In addition, the demolition of residential areas and abandoned farmlands imposed significant effects on Pb and Cd pollution, respectively.