Methods of establishing a correlation between variables
Correlation refers to the act of measuring how related or associated two variables are. A correlational study can either be positive or negative. A positive correlation is one in which the increase in one variable translates into an increase in another variable (Hancock & Mueller, 2002, p. 59). Alternatively, a variable could also increase in value and the process triggers the reduction in the value of another variable. For instance, if we consider education and income as two types of variables, more education could imply more income and vice versa.
On the other hand, we also have a negative correlation. In this case, an increase in the value of one variable triggers a decrease in the value of the second variable. In the same way, when one variable reduces in value, the other one increases in value (Rosnow & Rosenthal, 2002). This is more of an “inverse’ correlation. An example of a negative correlation is one that associate TV viewing with class grades. Therefore, as the number of hours spent by a student viewing TV increases, this translates into lower grades.
Advantages and disadvantages of positive and negative correlation
Advantages
One of the advantages of using the correlation technique is that it enables us to make predictions regarding two related things. As such, it becomes easier to predict the outcome of one thing based on another (Hancock & Mueller, 2002, p. 60). For instance, there is a positive correlation between, on the one hand, SAT scores and on the other hand, college achievement. Therefore, in case college admission officials wish to predict those students who have a higher probability of succeeding at their institution, all they need to do is to ensure that students who have scored high SAT scores are chosen for admission.
In terms of negative correlation, there is a negative correlation between years of jail time and years of education. Using this kind of negative correlation, prison officials can help to organize remedial education for individuals who have been behind bars for a long time, instead of enrolling them in college classes.
Disadvantages
One of the main disadvantages of a correlation is its inability to measure cause (Hancock & Mueller, 2002). For example, although earlier on we talked of a positive correlation between higher education and higher income, nonetheless, there is no way of knowing if indeed more income did not result in more education, and vice versa. Also, a third variable could be at play.
One of the disadvantages of a negative correlation is that it does not also indicate the cause of the correlation. For example, there is a negative correlation between crying among babies and being held whereby the more frequent a baby is held, the less likely they are to cry.
Circumstances and examples in which a researcher may want to establish a correlation
A researcher may wish to establish a correlation between two variables when he/she already has a hypothesis about how an increase or decrease in one variable may affect a second variable positively or negatively. For example, a researcher may wish to test a correlation between height and self-esteem. Before undertaking the experiment, he/she could have a hypothesis that as one gets taller this affects his/her self-esteem. In this case, the researcher will need to collect the height of the subjects (in inches) and also measure their self-esteem (using a rating scale). Thereafter, he has to determine the correlation between the two variables and then test the hypothesis.
Reference List
Hancock, G., & Mueller, R. (2002).The Reviewer’s Guide to Quantitative Methods in the Social Sciences. New York: Tarlor & Francis. Print. p.59.
Rosnow, R., & Rosenthal, R. (2002). Beginning behavioral research: a conceptual primer. New York: Prentice Hall.