2. Kendallâs Tau is a nonparametric analogue to the
Pearson Product Moment Correlation.
3. Similar to Spearmanâs Rho, Kendallâs Tau
operates on rank-ordered (ordinal) data but is
particularly useful when there are tied ranks.
4. Letâs consider an investigation that would lend
itself to being analyzed by Kendallâs Tau:
5. An iron man competition consists of three
consecutive events:
6. An iron man competition consists of three
consecutive events: Biking 110 miles,
7. An iron man competition consists of three
consecutive events: Biking 110 miles, Swimming
2.5 miles
8. An iron man competition consists of three
consecutive events: Biking 110 miles, Swimming
2.5 miles and Running 26.2 miles
9. An iron man competition consists of three
consecutive events: Biking 110 miles, Swimming
2.5 miles and Running 26.2 miles. Researchers
are interested in the degree to which the rank
ordered results from the biking and the running
events are independent of one another.
10. An iron man competition consists of three
consecutive events: Biking 110 miles, Swimming
2.5 miles and Running 26.2 miles. Researchers
are interested in the degree to which the rank
ordered results from the biking and the running
events are independent of one another.
11. An iron man competition consists of three
consecutive events: Biking 110 miles, Swimming
2.5 miles and Running 26.2 miles. Researchers
are interested in the degree to which the rank
ordered results from the biking and the running
events are independent of one another. Here is
the data for 6 individuals who competed:
12. Individuals Rank order for
Biking Event
Rank order for
Running Event
Bob
Conrad
Dallen
Ernie
Fen
Gaston
13. Individuals Rank order for
Biking Event
Rank order for
Running Event
Bob 1st
Conrad 2nd
Dallen 2nd
Ernie 3rd
Fen 4th
Gaston 5th
14. Individuals Rank order for
Biking Event
Rank order for
Running Event
Bob 1st 1st
Conrad 2nd 1st
Dallen 2nd 2nd
Ernie 3rd 3rd
Fen 4th 4th
Gaston 5th 4th
15. Because both variables are expressed as rank
ordered data, we will use either a Kendallâs Tau
or a Spearmanâs Rho.
16. Because both variables are expressed as rank
ordered data, we will use either a Kendallâs Tau
or a Spearmanâs Rho.
Note â even if only one variable were ordinal
and the other were scaled or nominal, you
would still use Kendallâs Tau or a Spearmanâs
Rho by virtue of having one ordinal variable.
17. Because there are ties in the data, we will use
Kendallâs Tau instead of the Spearmanâs Rho.
Individuals Rank order for
Biking Event
Rank order for
Running Event
Bob 1st 1st
Conrad 2nd 1st
Dallen 2nd 2nd
Ernie 3rd 3rd
Fen 4th 4th
Gaston 5th 4th
18. Kendallâs Tau renders a result that is identical to
Spearmanâs Rho and the Pearson Correlation
âą Therefore it shares the same properties as these
other methods:
â It ranges from -1 to +1.
â Itâs direction is determined by the sign (- +)
â The closer the value is to -1 or +1, the stronger the
relationship
â The closer the value is to 0, the weaker the
relationship.
19. Kendallâs Tau renders a result that is identical to
Spearmanâs Rho and the Pearson Correlation
-1 0 +1
âą Therefore it shares the same properties as these
other methods:
â It ranges from -1 to +1.
â Itâs direction is determined by the sign (- +)
â The closer the value is to -1 or +1, the stronger the
relationship
â The closer the value is to 0, the weaker the
relationship.
20. Kendallâs Tau renders a result that is identical to
Spearmanâs Rho and the Pearson Correlation
-1 0 +1
âą Therefore it shares the same properties as these
other methods:
â It ranges from -1 to +1.
â Itâs direction is determined by the sign (- +)
â The closer the value is to -1 or +1, the stronger the
relationship
â The closer the value is to 0, the weaker the
relationship.
21. Kendallâs Tau renders a result that is identical to
Spearmanâs Rho and the Pearson Correlation
-1 0 +1
âą Therefore it shares the same properties as these
other methods:
â It ranges from -1 to +1.
â Itâs direction is determined by the sign (- +)
â The closer the value is to -1 or +1, the stronger the
relationship
â The closer the value is to 0, the weaker the
relationship.
22. Kendallâs Tau renders a result that is identical to
Spearmanâs Rho and the Pearson Correlation
-1 0 +1
âą Therefore it shares the same properties as these
other methods:
â It ranges from -1 to +1.
â Itâs direction is determined by the sign (- +)
â The closer the value is to -1 or +1, the stronger the
relationship
â The closer the value is to 0, the weaker the
relationship.
23. Kendallâs Tau renders a result that is identical to
Spearmanâs Rho and the Pearson Correlation
-1 0 +1
âą Therefore it shares the same properties as these
other methods:
â It ranges from -1 to +1.
â Itâs direction is determined by the sign (- +)
â The closer the value is to -1 or +1, the stronger the
relationship
â The closer the value is to 0, the weaker the
relationship.
24. Kendallâs Tau renders a result that is identical to
Spearmanâs Rho and the Pearson Correlation
-1 0 +1
âą Therefore it shares the same properties as these
other methods:
â It ranges from -1 to +1.
â Itâs direction is determined by the sign (- +)
â The closer the value is to -1 or +1, the stronger the
relationship
â The closer the value is to 0, the weaker the
relationship or evidence of INDEPENDENCE.