Abstract
As the saga continues, the team now more than ever must be able to navigate through the new developments with regard to Greenleaf’s opponent. The team is tasked with addressing two more research questions that are both answered by using one of two Kendall rank-order tests. The two research questions that are addressed in this chapter are as follows: (1) Is sentence associated with the General Aggression Score? (2) Is sentence associated with the General Aggression Score when testosterone level is fixed? Kendall’s Rank-Order Correlation Coefficient addresses the former, and Kendall’s Partial Rank-Order Correlation Coefficient addresses the latter.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The reasons why Pearson’s Correlation Coefficient is inappropriate are discussed on page 89 in Chap. 5 under the discussion regarding Somer’s Index.
- 2.
- 3.
The powers of a positive integer are expressed by factorial design. For example, 4! or 44.
- 4.
Page 243 in Chap. 10 contains a detailed discussion of statistical significance for z-scores.
- 5.
When dealing with sample size for Kendall’s Rank Order, two other options exist besides the N > 30 option. Those two options are N ≤ 10 and N > 10. When N ≤ 10, the exact probability, or p-value, can be found utilizing an Upper-tail probabilities for T table. When N > 10, T can be assumed to be normally distributed a Mean = μT = 0 and Variance = σT 2 = 2(2N + 5)/9N(N−1). z is then found using the same formula as for N>30 as discussed here; however, the value for T can be found in a Critical Values table for T.
- 6.
Kendall’s Partial Rank-Order Correlation Coefficient and Kendall’s Rank-Order Correlation Coefficient share many of the same steps and tasks. Refer to the discussion regarding the Rank-Order Correlation Coefficient on page 122 of this chapter to assist in understanding some of the steps for the Partial Rank-Order Correlation Coefficient.
- 7.
The formula to be used when there are no ties can be found on page 139.
- 8.
The calculations here are identical to the calculations discussed previously in this chapter in dealing with Kendall’s Rank-Order Correlation Coefficient and Kendall’s Partial Rank-Order Correlation Coefficient.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Science+Business Media New York
About this chapter
Cite this chapter
Linebach, J.A., Tesch, B.P., Kovacsiss, L.M. (2014). Agreeing to Disagree. In: Nonparametric Statistics for Applied Research. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-9041-8_6
Download citation
DOI: https://doi.org/10.1007/978-1-4614-9041-8_6
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-9040-1
Online ISBN: 978-1-4614-9041-8
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)