Analyses

Goodman–Kruskal gamma

Association between two ordered categorical variables.

Edit on GitHub

When to use it

Use gamma when both variables are ordered categories and you want to know whether higher scores on one tend to come with higher scores on the other. Gamma runs from −1 to 1. It ignores ties.

Assumptions

Both columns are ordered. Tensr sorts level names alphabetically, so the names need to sort in the true order.

Running it in Tensr

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

Options

Prop

Type

Reading the output

Teaching method against the 1–4 satisfaction rating. Method and satisfaction share a mild ordinal association through the exam.

Goodman–Kruskal γ for method and 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: Gamma = 0.198; ASE = 0.001; z = 157.702; p-value = < .001; Concordant = 1348; Discordant = 902.

Reporting (APA 7)

Goodman–Kruskal γ for method and 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 → Gamma.