Rammed-earth (RE) is a construction material manufactured from the soil. The soil is compacted at its optimum water content, inside a formwork to build a monolithic wall. RE material is attracting renewed interest throughout the world thanks to its sustainable characteristics: low embodied energy, substantial thermal inertia, and natural regulator of moisture; on the other hand, the existing historic RE buildings is still numerous. This is why several research studies have been carried out recently to study different aspects of this material. However, few investigations have been carried out to explore the possibility of applying the nondestructive techniques on RE walls. This paper presents an assessment of the well-known rebound hammer test on RE walls. The calibration curves of the rebound hammer test have been established for conventional concrete where the rebound number is more than 20. For RE material with lower compressive strengths, a new calibration curve must be established. In the present study, two soils were used and different homogenized specimens with different dry densities were manufactured and tested, to plot a general calibration curve. Then, this calibration curve was applied to RE specimens; different results at different positions in an earthen layer were observed, due to the inhomogeneity of the material. The final results showed an acceptable accuracy of the calibration curve in the prediction of the compressive strength of RE material.