Spearman’s Correlation
Pearson’s correlation coefficient is, by far, the most commonly used metric for the degree of correlation between two variables. But r has its limitations, the main one being that it assumes that both variables should be normally distributed (which might sound restrictive, but really just means that they behave in a “typical. random fashion”). But sometimes there are independent variables that aren’t distributed in this way, in which case r isn’t an appropriate measurement. Enter Spearman’s correlation coefficient (or Spearman’s rho), denoted by p or re.
a. First. review Spearman’s Rank Correlation e , from the University of Texas, which briefly defines Spearman’s correlation coefficient The first article begins with an in-depth discussion of Pearson’s r, which you may skip but I would recommend that you read, as it supplements our discussions of r from the lessons. About halfway down the page. you will find the section on Spearman Rank Correlation. and you should read from there to the end of the article.
b. You may then review Schober, P. et al. (2018). The authors visually depict some of the main differences between Pearson’s and Spearman’s coefficients. ■ Schober. P., Boer, C., and Schwarte, L (2018). Correlation Coefficients: Appropriate Use and Interpretation . Anesthesia & Analgesia, 126(5), 1763-1768. c. You may also review Hauke and Kossowski (2011): you may ignore the discussions on Kendall’s T. ■ Hauke. J., and Kossowski. T (2011). Comparisons of Values of Pearson’s and Spearman’s Correlation Coefficient on the Same Sets of Data e . Quaestiones Geographicae. 30(2).
Explain what variables are being analyzed and then explain why you think the researchers decided to use p instead of r. If they used both, then explain which one resulted in a more significant coefficient and why you think that happened.