The correlations are the association between the variables. Each variable is assigned with a number to represent the value of a variable. However, sex, origin, and political preference can be treated as a variable but we don’t know how to assign a number to represent each category. In general, a variable can take more than one value. The linear correlation is a simplest form of association. There are some problem arises during the evolution of the two different data using the correlation, e.g., in nuclear power plant, there is a correlation between the radiation and cancer because people can get cancer due to the leak out of nuclear radiation. The analysis cannot able to get a result from the correlation. Now, the analysis should be made between the two data; the smoking and the heart pulse rate. Does the smoking causes the pulse rate to increase? The conclusion cannot be reached simply because the there is a correlation between the smoking and the heart pulse rate. The tobacco companies argued that there were some other factors causes the heart rate to increase, e.g., drinkers, using the toxic elements, and so on. There are logically correct. The causal relationship makes no sense. The analysis cannot able to get a result from the correlation, which means, when we try to calculate a correlation between smoking and heart pulse rate, it is difficult to control all the relevant factors. We can correlate any two variables, but it does not represent the causality. The main disadvantage of the correlation is when we report relationship between the two data sets; we cannot represent the causality. The researchers cannot able to find the data set responsible for the increase or decrease in correlation.
“Correlations are hard to intercept.” U texas. Web. 8 Feb 2014. http://www.utexas.edu/courses/bio301d/Topics/Correlation/Text.html