Objectives Facial asymmetry in orthodontic treatments can be evaluated by both posteroanterior (PA) cephalograms and conebeam computed tomography (CBCT) images. This study aimed to assess the agreement between CBCT and two-dimensional PA images in terms of cephalometric measurements. Methods In this descriptive analytical study, CBCT and PA radiographs were taken from nine human dry skulls. Two observers marked the bilateral landmarks including CO (orbit center), J (jugale), 6C, 6A, 1A, 1C, GO (gonial angle), AG (antegonion), AR (articular) and Ma (mastoidal). The distance between two identical points on both sides was measured on both the PA and CBCT images. The differences were calculated and the agreement between the two modalities was checked by using intraclass correlation coefficient (ICC). Results The mean differences between CBCT and PA measurements were as following: for CO=1. 48, J=11. 64, U6C=0. 75, U6A=1. 8, L6C=13. 33, L6A=3. 0, U1C=0. 96, U1A=0. 62, L1C=0. 22, L1A=0. 45, GO=0. 87, AG=6. 67 and AR=0. 71 mm. The agreement was the highest for GO (ICC=0. 931) and CO (ICC=0. 902), and the lowest for U6A (ICC=-0. 041) and J (ICC=0. 038) landmarks. Conclusion Given the negligible differences between the two modalities, conventional PA cephalograms can be as competent as CBCT in detecting maxillofacial asymmetry with lower patient radiation dose.