Analyses

Jonckheere–Terpstra test

Test whether scores rise across an ordered set of groups.

Edit on GitHub

When to use it

Use Jonckheere–Terpstra instead of Kruskal–Wallis when the groups have an order and you expect the scores to rise or fall along that order.

Assumptions

Groups are independent. Pass group_order when the label sort is not the order you mean. Otherwise numeric labels sort numerically and other labels sort alphabetically.

Running it in Tensr

Analyze → Nonparametric Tests → Jonckheere–Terpstra. In chat: “Jonckheere–Terpstra test of score by method.”
Set group order to lecture, online, workshop when those names should be the sequence.

Options

Prop

Type

Reading the output

The teaching methods ordered Workshop, then Lecture, then Online. That is not the order of the exam means, so a monotone trend in this order is not expected.

Jonckheere–Terpstra trend test for score across ordered groups (Workshop, Lecture, Online). The primary result is not significant (p = .805). This is not large enough to treat the comparison this page is about as a reliable association. The result is non-significant: the data are still compatible with no effect. Metrics: J statistic = 1409; Z = -0.861; p-value = .805.

Reporting (APA 7)

Jonckheere–Terpstra trend test for score across ordered groups (Workshop, Lecture, Online). This result is not significant (p = .805). Report the estimate with that p, and do not describe the pattern as a reliable effect.

Coming from SPSS

Analyze → Nonparametric Tests → Jonckheere–Terpstra.