Efficiency What does it mean when we say that the rank correlation test has an efficiency rating of 0.91 when compared to the parametric test for linear correlation?

The efficiency value of the rank correlation test is equal to 0.91.

The efficiency rating of a non-parametric test can be used to compare a non-parametric test with its parametric counterpart.

Rank correlation test is the non-parametric counterpart of linear correlation.

The efficiency value indicates a requirement of 100 observations in a sample for the non-parametric rank correlation test. However, the parametric linear correlation only requires 91 observations in a sample to produce similar results.

Since a small sample is required, the efficiency of a parametric test in terms of accuracy, time, and cost is greater than the non-parametric rank correlation test.

Thus, it can be said that if the assumption regarding the distribution of the population is fulfilled, it is better to use the parametric linear correlation.