Kendall τ (aprox)

This calculator estimates the Kendall τ correlation coefficient, a non-parametric measure of correlation for scalar variables. It is useful in research studies when it is not possible to assume a normal distribution for the variables. The calculation is made using the formula: (C - D) / (n(n-1)/2), where C is the number of ordered pairs that are in the same order in both variables, D is the number of ordered pairs that are in the reverse order in both variables and n is the number of data.

The Kendall τ is a robust correlation measure that is less affected by extreme values or outliers. It is also less influenced by missing data. Additionally, the Kendall τ is a non-direct correlation measure, that is, it does not assume that the variables are related in a linear or non-linear way.

This calculator is useful in a variety of situations, including health, economic, environmental and social data analyses, where it is necessary to evaluate the correlation between non-normal variables. Additionally, it is useful in cases where it is necessary to perform a non-parametric hypothesis test, such as the hypothesis of equality of correlations between different populations.