This paper presents the application of Exploratory Spatial Data Analysis (ESDA) and Kriging from GIS (ArcGIS8.3) in disease mapping through the analysis of hepatitis B in China. The research shows that geostatistical analysis techniques such as Kriging and ESDA have a good effect in disease mapping. Kriging methods can express properly the spatial correlation. Furthermore, unlike model-based methods, which largely depend on assumption for disease data, the Kriging method is more robust for the data. So it can be used more widely and is more operational. What’s more, the Kriging method may be adapted to interpolate nonstationary spatial structure. This can expand its application more largely. At last, the Kriging method can estimate the uncertainty of prediction while many deterministic methods cannot do so. In conclusion, it is an effective operational procedure to gain a deep insight into the disease data through ESDA before mapping disease using the Kriging method.