DataTerminology.comNEW
A nonparametric statistic measuring ordinal association between two variables using counts of concordant and discordant pairs.
DESCRIPTION
Kendall’s tau evaluates monotonic relationships by comparing rank orderings. It counts concordant and discordant pairs, adjusting for ties in variants like tau-b. Tau is robust to outliers and appropriate for ordinal data. It complements Spearman’s rho and Pearson’s r for different assumptions and data types. Analysts use tau to assess association strength when distributions are unknown or non-linear. It supports hypothesis testing and interpretation of ordinal relationships.
ALTERNATIVE DEFINITIONS
NIST/SEMATECH e-Handbook
Kendall’s tau is a nonparametric measure of association based on the ranks of the data. It assesses the strength of association between two variables.
Highlights rank-based, nonparametric nature and association strength.
Penn State Statistics
Kendall’s tau is a measure of correlation for ordinal data that compares the number of concordant and discordant pairs.
Emphasizes ordinal correlation and pairwise comparisons.
SciPy
Compute Kendall’s tau, a correlation measure for ordinal data.
Provides computational perspective within scientific computing library.
RELATED TERMS
Pearson Correlation Concordant Pair Discordant Pair Mann–Kendall Test Tie Correction
COMPETING TERMS
Spearman Rank Correlation Pearson Correlation Coefficient
KNOWLEDGE AREAS
TARGET AUDIENCE
Data Scientist Data Analyst BI Specialist Information Analyst
EXTERNAL SOURCES
- Wikipedia - Kendall rank correlation coefficient
- NIST - Tau definition and details
- SciPy - Function for Kendall’s tau
- Penn State - Course notes on ordinal correlation
TOP QUESTIONS
When should Kendall’s Tau be preferred over Pearson’s correlation?
Use tau for ordinal data or non-linear relationships where rank association matters and distribution assumptions are weak.
Category: Requirements
How does Kendall’s Tau handle ties?
Variants like tau-b adjust for ties to improve interpretability, especially with discrete or repeated ranks.
Category: Example
What is a common mistake when interpreting Kendall’s Tau?
Assuming linear relationship strength; tau measures monotonic association, not linearity as Pearson’s r does.
Category: Mistakes
What risks arise from over-reliance on Kendall’s Tau?
Important non-monotonic patterns may be missed; complementary analyses with Spearman’s or visualization can prevent misinterpretation.
Category: Risks
Is Kendall’s Tau robust to outliers?
Yes, being rank-based makes tau less sensitive to extreme values than Pearson’s r.
Category: Opinion
How is Kendall’s Tau computed?
It compares counts of concordant and discordant pairs and adjusts for ties when using tau-b.
Category: Example
What competing measures relate to Kendall’s Tau?
Spearman’s rho and Pearson’s r, which target monotonic and linear relationships respectively.
Category: Alternatives
Where can I learn computations for Kendall’s Tau?
Consult NIST’s e-Handbook and SciPy’s kendalltau function for definitions and practical implementation.
Category: Learning
Why is Kendall’s Tau relevant for data science?
It offers a robust way to assess ordinal associations, useful in nonparametric modeling and exploratory analyses.
Category: Relevancy