October 3, 2013
Correlations measure the relationship between two variables. Establishing correlations allows researchers to make predictions that increase the knowledge base. Different methods that establish correlations are used in different situations. Each method has advantages and disadvantages that provide researchers information that is used to understand, rank, and visually illustrate how variables are related. The Pearson’s, Spearman, Kendall Rank, and positive and negative correlation are methods used to establish a correlation between variables. The Pearson method is a simple linear correlation used or illustrate how strong of a relationship two variables have. The Spearman method ranks data by order or name and is often used because the equation is simpler than Pearson’s. The Kendall Rank method measures the strength of dependence between two sets of random variables. Depending on the use of the information will determine the best method for the research project. Each method has advantages and disadvantages. The research project and individual researcher weigh out the positive and negatives in order to determine the best method. It is agreed that the Pearson’s method is easy to understand and illustrates the strength of the correlation. A prominent disadvantage is there can be confusion because it may be assumed that correlation establishes causation. The Spearman advantage is it can rank order or name data in various ways, depending on the data collected. The disadvantage discussed is there may be a focus on ranking and not on the information that creates the rank. The positive and negative correlation method has the advantage that a lot of variables and situations can be used. Variables can be studied using this method that experiments cannot be conducted on. The clear disadvantage discussed is that no cause and effect relationship can be assumed. Another discussed disadvantage of...
References: Ling, R. F. (1982). Reviewed work (s): Correlation and Causation. by David A. Kenny Source: Journal of the American Statistical Association, Vol. 77, No. 378,(Jun., 1982), pp. 489-491 Published by: American Statistical Association. Journal of the American Statistical Association, 77(378), 489.
Wright, S. (1921). Correlation and causation. Journal of agricultural research, 20(7), 557-585.
Please join StudyMode to read the full document