Somers’ D

Somers’ d (Somers’ delta), is a nonparametric measure of the strength and direction of association that might exist between an ordinal dependent variable and an ordinal independent variable. It is first proposed in 1962 by Robert H. Somers. It is an alternative to the Goodman and Kruskal’s gamma, except that the Somers’ d distinguishes between a dependent and independent variable and the Goodman and Kruskal’s gamma does not.

It has values between − 1 (all pairs of the variables disagree) and 1 (all pairs of the variables agree).

Assumptions Underpinning Somers’ d:

  • there is a dependent variable and an independent variable that are both measured on an ordinal scale
  • there needs to be a linear relationship between the variables

Calculating Somers’ D in SPSS:

Example of a Study Using Somers’ d:

  • This study on the “Validation of a system of foot ulcer classification in diabetes mellitus” used Somers’ d to look at the relationship between the independent variables (area, depth, sepsis, ischaemia and neuropathy categorization of the diabetic ulcers) and the dependent variables (healed and unhealed).

Comments are closed.