Purpose: To evaluate interobserver agreement for the assessment of calcifications on screening mammography, and to evaluate the difference between routine views and spot compression with magnification views. Materials and Methods: We retrospectively analyzed the mammographic findings of 135 patients who had calcifications on the screening mammography from January to December 2005, and confirmed by biopsy or surgery, or follow up mammography more than 2 years. Two breast-dedicated radiologists, who were blind to the results of calcifications, independently reviewed routine mammography and spot compression with magnification views which were mixed randomly, and assessed calcifications from category 2 to 5 according to the ACR BI-RADS. We evaluated interobserver agreement on routine views and spot compression with magnification views, and compared in malignant and benign calcifications. Kappa value and McNemar test were used in statistical analysis. Results: There were 136 lesions of calcifications in 135 patients. There were 36 malignant calcifications and 6 atypical ductal hyperplasia confirmed by surgery or biopsy, and 94 other benign calcifications confirmed by biopsy or follow up mammography over 2 years. Overall interobserver agreement for the assessment of calcifications on routine views was 0.505, 0.515 in malignant calcifications and 0.462 in benign calcifications. Overall interobserver agreement on the spot compression with magnification view was 0.600, 0.600 in malignant calcifications and 0.566 in benign calcifications. Conclusion: For the assessment of calcifications detected on the screening mammography, assessment on the spot compression with magnification views can increase the interobserver agreement than assessment on the routine views only.