As researchers, it is necessary to know differing methods of establishing correlation between variables. Further, researchers must be aware of the advantages and disadvantages of each of these methods, as well as where each method should be applied. Finally, researchers need to know particular circumstances in which correlation needs to be established.
Correlation is important because it shows relationships between variables. It can be further compared to a summary of the variable relationship and is known as the correlation coefficient. It can be determined in a variety of ways. For instance, the correlation coefficient can be calculated utilizing the Spearman’s rho or Pearson’s r (Weiss, 2012, Ch. 14). For example, the Spearman’s rho is used for ordinal scales. Pearson’s r is commonly used for ratio level scales. Pearson’s r provides a summary that is designed to describe the statistical relationship between interval variables. This is crucial because it shows the degree of relationship. For instance, a Pearson’s r correlation coefficeint of 0 suggests there is minimal relationship between the variable. In contrast, a Pearson’s correlation coefficient that moves away, either positively or negatively, from 0 suggests a more expansive relationship between the variables (Weiss, 2012, Ch. 14). Thus, a Pearson’s r correlation is designed to show the degree of the relationship between two ratio or interval variables. However, it has the disadvantage of Spearman’s rho is used primarily in conjunction with the ordering of the values. For example, it orders the data and utilizes the difference in order for its compution (Weiss, 2012, Ch. 14). However it has the disadvantage of being non-parametric. This is especially true if there are many points. Despite this disadvantage, the Spearman’s rho has the advantage of considering underlying ranked scales. This shows that not all variables and relationships are created equally and for the most beneficial interpretations, needs to be taken into consideration.
- Weiss, N. A. (2012). Introductory Statistics. Pearson.