Non-Parametric Alternatives

why would we use a non-parametric test

•If we violate any assumptions for parametric tests, e.g. our data is not normal.

What is the Spearman correlation coefficient, and what are its key features, assumptions, and uses?

The Spearman correlation coefficient, denoted as ρ (rho) or rs, works on ranked data rather than raw values like the Pearson correlation. It can use the same effect size cut-offs as Pearson correlation and assesses the strength of a monotonic relationship between two variables, whether linear or non-linear. At least one variable must be continuous, while the other can be continuous or dichotomous, and the relationship between the variables must be monotonic.

What is the Wilcoxon test, its types, and when is each used?

Like the t-test, the Wilcoxon test comes in two forms: one-sample and two-sample. It can handle any type of data when comparing two groups. The Mann-Whitney U (two-sample) Wilcoxon test is used for between-subjects comparisons, while the one-sample Wilcoxon test is used for within-subjects comparisons.

How do you report Spearman correlation results?

Spearman correlations are reported using the format: rs (n) = correlation coefficient, p = significance. For example, rs (38) = 0.34, p = .009. A general template is: rs (degrees of freedom) = estimate, p = x.

How do you report Wilcoxon test results?

Wilcoxon test results are reported using the format: W = test statistic, p = significance. For example, W = 110, p < .001. A general template is: W = estimate, p = x.