Analyses

Somers’ d

Ordered association that picks one variable as the outcome.

Edit on GitHub

When to use it

Use Somers’ d when both variables are ordered and one of them is the outcome. Unlike gamma, d is asymmetric: predicting B from A is not the same number as predicting A from B.

Assumptions

Both columns are ordered categories. Level names are sorted alphabetically.

Running it in Tensr

Analyze → Correlate → Somers D. In chat: “Somers’ d for dose and response.”
Column A and column B are the two ordered variables.

Options

Prop

Type

Reading the output

The same method-by-satisfaction table, as an asymmetric ordinal association.

Somers' D (X = method, Y = satisfaction), n = 96. The primary result is significant (p = < .001). This is large enough, in this sample, to treat the comparison this page is about as a real association rather than noise. Metrics: D(Y|X) = 0.134; D(X|Y) = 0.145; ASE = 0.031; z = 4.271; p-value = < .001.

Reporting (APA 7)

Somers' D (X = method, Y = satisfaction), n = 96. This result is significant (p = < .001). The effect is moderate, so report its size with the p-value.

Coming from SPSS

Analyze → Correlate → Somers’ d. SPSS asks which variable is the dependent.