Statistical shape modeling techniques were used to quantify the 3 dimensional anatomic variations of the sacrum and sacroiliac joint surface in the general population. 25 pelvises were segmented from a from database 223 patients who had CT studies performed for evaluation of abdominal pain at the University of Colorado Hospital. Surface STL meshes were created using Simpleware ScanIP. The statistical shape models was created using the Coherent Point drift algorithm for registration and correspondence. SSM analysis and ordination was performed utilizing procrustes analysis and principal component analysis(PCA). PCA provides dimensionality reduction of the data for visualization and interpretation of shape features. We found variation in the sacroiliac joint morphology could be quantitatively described using our CPD-SSM technique. We also found the first, second and third PCA modes describe changes in the sacroiliac joint shape.