The purpose of this study was to examine the potential impact of the testing procedure, the shape of the test sample, loading method and sample size on the KIc value of polymer concrete (PC) materials. The research involved experimental investigations using five different testing techniques and specimen types, namely the single edge notched beam (SENB), short bend beam (SBB), semi-circular bend (SCB), edge notch disc bend (ENDB), and center cracked Brazilian disc (CCBD). A typical PC mixture made of mineral silicious aggregate, ML506 epoxy resin, chopped E-glass, and foundry sand filler. Despite the difference in the shape and loading type of the tested samples, the KIc data obtained from all groups of specimens are in good agreement with together and with the SENB proposed by RILEM. Depending on the test type, the KIc value varied from 1.43 to 1.74 MPa.m0.5 and the discrepancy between the data was mainly attributed to the type of loading (compression or bending) and the crack type (center crack or edge crack). The T-stress affects the fracture toughness for different testing samples and configurations. The lowest fracture toughness corresponds to the CCBD specimen (the center cracked disc loaded diametrically). The other test samples with edge cracks and loaded by a three-point bend setup showed KIc = 1.7 - 1.74 MPa.m0.5. Moreover, the fracture toughness data for PC mixtures can be achieved by utilizing sub-sized samples like SBB (for smaller amounts of PC material) instead of larger beam samples (i.e., SENB).