Burland and Burbidge (1984) collected, presented and back-analysed more than 200 hundred case histories of settlement of full scale structures founded on coarse grained soils and proposed an empirical procedure for settlement prediction.

Berardi and Lancellotta (1991) performed further back-analyses on the data base provided by Burland and Burbidge and developed a different approach for the same purpose.

This paper presents some further analyses, performed on the same data base used by Burland and Burbidge and by Berardi and Lancellotta. The results of these analyses are used to assess the reliability of both procedures and provide simple empirical correlations which can be used for settlement predictions for coarse grained soils with a reliability comparable with the reliability of the procedures proposed by Burland and Burbidge and by Berardi and Lancellotta.