Which test is used for measuring correlation with ordinal data?

Spearman's rho is the appropriate test for measuring correlation with ordinal data because it assesses the strength and direction of the association between two ranked variables. This method is particularly useful when the data does not meet the assumptions necessary for parametric tests, such as normality, which are required for measurements like Pearson's correlation.

Ordinal data, which consists of rankings (e.g., rating scales) rather than continuous measures, can be effectively analyzed using Spearman's rho since it operates on the relative positions of data points rather than their actual values. This makes it robust for testing correlations even when the data is not normally distributed.

The other options are typically not suitable for ordinal data. Pearson's correlation is designed for interval or ratio data and assumes a linear relationship and normal distribution of the variables involved. The Mann Whitney U test and Wilcoxon test are non-parametric tests, but they are used for different purposes—specifically, comparing differences between two groups rather than measuring the correlation between two sets of rankings. Thus, Spearman's rho is the clear choice for correlation analysis with ordinal data.