The Spearman's rho correlation is used when:

• A. The data are not valid interval-level and/or one variable is highly skewed
• B. The data are perfectly normally distributed interval data
• C. The sample size is extremely large with precise measurements
• D. The variables are categorical

Answer: (A)

Spearman's rho is a nonparametric measure of association that uses the ranks of the data rather than the raw scores. This makes it ideal when the data don’t meet interval-level assumptions or when distributions are skewed, because ranking minimizes the impact of outliers and non-normality. It assesses whether, as one variable tends to increase, the other tends to increase or decrease in a consistent, monotonic way, without assuming a linear relationship or normal distribution.

So, when data are not valid interval-level and/or one variable is highly skewed, Spearman's rho is the best choice. If data were perfectly normally distributed interval data, Pearson correlation would generally be more appropriate. If the variables are categorical, correlation isn’t suitable and other association measures are used. A very large sample with precise measurements doesn’t by itself mandate using Spearman; the measurement scale and distribution are the deciding factors.