Soil chemical compositions are multivariate in nature and provide datasets suitable for the application of multivariate factor analytical techniques. One of the analytical techniques, the positive matrix factorization (PMF), uses weighted least squares by fitting the data matrix to determine the weights based on the error estimates of each data point. In this research PMF was employed to apportion the sources of a pollutant (Pb) in 104 soil samples taken within a 2-km radius of a contaminated site around a lead battery plant in Changxing, Zhejinag Province, China. PMF successfully partitioned the variances into sources related to soil background, the lead battery plant, agricultural activities, and other mixed contamination sources such as road dust emission and vehicle exhaust. It was estimated that the lead battery plant and agricultural activities contributed 50.71% and 29.28%, respectively, to the total Pb pollution. Combining the results from PMF with a geostatistical approach, the main sources (the lead battery plant and agricultural activities) of the soil Pb contamination near the contaminated site was also successfully demarcated. This research indicates that PMF combining with geostatistics is a useful tool for source identification and apportionment.