Analyses

Sign test

Test whether paired scores change direction, ignoring how large the change is.

Edit on GitHub

When to use it

Use the sign test when you only care whether each pair went up or down. It is the simpler alternative to the Wilcoxon signed-rank test and the paired t-test.

Assumptions

Each row is one pair. Ties (no change) are counted and then set aside.

Running it in Tensr

Analyze → Nonparametric Tests → Related Samples. In chat: “Sign test of before and after.”
Before column and after column are the two measurements.

Options

Prop

Type

Reading the output

A second set of 48 pairs whose true change is about zero. The sign test only counts who went up and who went down.

Sign test comparing before_ns and after_ns, n = 48 pairs. The primary result is not significant (p = .471). 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: Positive diffs = 27; Negative diffs = 21; p-value = .471.

Reporting (APA 7)

Sign test comparing before_ns and after_ns, n = 48 pairs. This result is not significant (p = .471). Report the estimate with that p, and do not describe the pattern as a reliable effect.

Coming from SPSS

Analyze → Nonparametric Tests → Related Samples.