The Goodman jack is widely used to measure the deformation modulus of rock in the field. Although some corrections have been proposed for the jack in the research literature, the accuracy of the test remains unsatisfactory, and the results of the Goodman jack test are lower than the actual values. Based on the corrections proposed in the literature, we reported on two factors which may lead to a lower value of the Goodman jack test: one was the non-full rock/platen contact when the borehole is under- or over-sized (relative to the standard borehole with a diameter of 76.2 mm), and the other was the upper limitation on the applied hydraulic pressure. First, a numerical simulation was carried out to investigate the impact of the borehole diameter on the rock/platen contact and the calculated rock modulus. The results showed that when there is a deviation of 0.6 mm in the borehole diameter, the calculated rock modulus may be 2 to 5 times lower. Based on the results of numerical simulation, a borehole diameter-related correction is proposed for the tests in under- or over-sized boreholes. Second, to ensure that the rock responds elastically during the test, we analyzed the stress around the borehole, and based on the Mohr-Coulomb and tensile failure criteria, deduced the upper limitation for the applied hydraulic pressure. The results of a case study in a coal mine suggest that the pressure limitation is controlled by the initial stress in the field and the rock strength, and influenced significantly by the loading direction of the jack. Finally, based on the observations in the present research, a new procedure is suggested for the Goodman jack test, which may be helpful in improving the accuracy of in situ testing.