Abstract: Six groups of full-scale field tests on installation damage of two kinds of PET uniaxial geogrid placed in gravel are performed. Geogrids are installed in three different manners, i.e., directly in gravel, an additional sand layer above grids, additional sand layers both above and under grids, to simulate different installation conditions. Residual tensile strengths of geogrids exhumed after installation are tested, and installation damage factors are decided for grids under different working conditions. The effectiveness of sand layer for decreasing installation damage is assessed. The tests indicate that it is effective to place a 10 cm-thick sand layer over grids to decrease installation damage and increase the allowable tensile strength of grids. Reduction factors under two conditions with an additional sand layer are quite close, which reveals that the reduction of strength is mainly ascribed to the installation procedure of fill materials above grids. The level of installation damage depends mainly on the grain size of fill materials and parameters of compaction, and the difference of specifications of geogrids has rather small influence.