This study was conducted to investigate the effect of core sampling on Warner–Bratzler shear force evaluations of beef and pork loins (Longissimus thoracis et lumborum muscles) and to determine the relationship between them. Steaks of 2.54 cm from beef and pork loins were cooked and five round cross-section cores and five square cross-section cores of each steak were taken for shear force evaluation. Core sampling influenced both beef and pork shear force values with higher (P < 0.05) average values and standard deviations for square cross-section cores. There was a strong and linear relationship (P < 0.01) between round and square cross-section cores for beef (R^2 = 0.78), pork (R^2 = 0.70) and for beef + pork (R^2 = 0.82) samples. These results indicate that it is feasible to use square cross-section cores in Warner–Bratzler shear force protocol as an alternative and potential method to standardize sampling for shear force measurements.