Presented datasets consist of separate .csv files, that contain PCA scores and information about sex and population affinity. Each sample (Old Kingdom, contemporary Egypt, pooled sample etc.) has two files with PCA scores, one for shape and for form (shape + size together) as well as separate .csv file with information about sex/population. PCA scores were obtained after principal component analysis was applied to each dataset, that consisted of 3D models of skulls. PCA scores together with information about sex/population were used for training SVM for classification of sex and population. Details about samples bellow. The contemporary Egypt sample consists 96 CT scans, 49 males and 47 females, 3D models were obtained from DICOM data using Avizo. The Old Kingdom period sample consists of 54 3D models of skulls of individuals from Abusir and Giza, specifically 32 males and 22 females. 3D data from both samples are stored at the Laboratory of 3D Imaging and Analytical Methods, Department of Anthropology and Human Genetics, Faculty of Science, Charles university. Presented datasets were assembled and used as a part of study focused on testing the reliability of sex estimation method developed by Musilová et al.(2016) on non-European population. The datasets are made publicly available to enable reproducibility of the aforementioned work and to ensure data sharing and engagment for other studies focused on sex estimation using skull.