Matric suction in the laboratory testing of an unsaturated specimen is usually controlled by the axis-translation technique or the osmotic technique. To investigate possible differences in the axis-translation and osmotic techniques in shear testing of unsaturated soils, three series of unsaturated triaxial shear tests were conducted on a compacted expansive clay with an initial degree of saturation of 82%. Two of these series were performed using the axis-translation and osmotic techniques at HKUST. Another series was performed using the osmotic technique at ENPC. In the series using the axis-translation technique, the single-stage testing procedure was compared with the multi-stage testing procedure and these two testing methods were found to provide consistent shear strength information. The consistency of the results from the osmotic technique at HKUST and ENPC demonstrated the reliability of the osmotic technique. Based on the extended Mohr-Coulomb shear strength formulation, comparison of the results using the axis-translation technique and the osmotic technique showed that there was no difference in determining the internal friction angle ø'. However the values of ø° in the test series using the axis-translation technique were slightly larger than those in the test series using the osmotic technique by 3º - 4º.