13.13
Kendall's Tau Test
Kendall's tau test, also known as the Kendall rank coefficient test, is a nonparametric method for assessing association between two variables. This t…
Consider an example where 35 pieces of vintage porcelain teacups are ranked independently by an artisan and an ordinary buyer.
Kendall's tau test can be used to find if any association between these two rankings exists.
Here, the null hypothesis states that there is no correlation between the artisan's rankings and the buyer's. The alternative hypothesis is that a correlation exists between these two rankings.
First, arrange the data in a specific order, for instance, per the artisan's ranking.
These paired ranks are required to be converted into counts.
In the conventional method of calculating counts, first, locate a rank and count the total number of ranks higher than that in the same column. Repeat this process for all the ranks in the data.
The test statistic tau can be calculated using the following equation.
The significance of this quantity can be obtained using a suitable computer-based tool.
Kendall's tau is particularly useful for finding linear or non-linear monotonic associations between the variables when there are no ties in the data.
View the full transcript and gain access to JoVE Core videos
Q1: What is Kendall's tau test used for in statistics?
Kendall's tau test is a nonparametric method for assessing association between two variables when their distributions are unknown. It identifies whether a correlation exists between paired rankings without requiring strict distributional assumptions. The tau coefficient ranges from -1 to +1, where +1 indicates perfect positive correlation and -1 indicates perfect negative correlation, making it valuable for analyzing ordinal data.
Q2: How do you calculate the test statistic in Kendall's tau?
To calculate Kendall's tau, first arrange data by one variable's ranking. Then count how many ranks in the second column are higher than each rank position. Repeat this for all ranks to obtain counts. The test statistic tau is calculated using these counts in a specific equation. The significance of the resulting tau value can be determined using computer-based statistical tools.
Q3: What does a tau value of +1 or -1 indicate?
A tau value of +1 indicates that the ranks of two variables are perfectly similar, suggesting a strong positive correlation between them. Conversely, a tau value of -1 indicates that the ranks are perfectly dissimilar, suggesting a strong negative correlation. Positive tau values indicate positive relationships, while negative values signify negative relationships between the variables.
Q4: How does Kendall's tau compare to Spearman's rank correlation test?
Kendall's tau and Spearman's rank correlation test are equivalent and equally precise nonparametric methods for assessing rank correlation. Both tests are valid for analyzing associations between variables without strict distributional assumptions. However, Kendall's tau calculation is more straightforward when there are no ties in the data ranks and is more widely used for such data.
Q5: What are the null and alternative hypotheses in Kendall's tau test?
The null hypothesis states that no correlation exists between the two sets of rankings being compared. The alternative hypothesis is that a correlation does exist between the rankings. These hypotheses are tested by calculating the tau statistic and determining its significance, which indicates whether sufficient evidence exists to reject the null hypothesis.
Q6: When is Kendall's tau test most appropriate to use?
Kendall's tau test is particularly useful for finding linear or non-linear monotonic associations between variables when there are no ties in the data. It is valuable for analyzing ordinal data and exploring relationships without relying on strict assumptions about underlying distributions. The test is especially appropriate when sample and population distributions are unknown.
Q7: What is the rank intersection method in Kendall's tau calculation?
The rank intersection method involves arranging data into two columns with rankings from each variable, then drawing lines connecting matching ranks between columns. The total number of intersections formed by these lines is counted as X. This intersection count is then used in an equation to calculate N, which contributes to determining the tau coefficient and correlation strength.