The bivariate measurement of strengths or association between two variables is called Correlation. It is a statistical technique that determines the status and way of relationship between two variables. Thus, correlation is the relationship of two apparently different aspects with some common connection. The recognition and measurement of that common denominator is what is known as the Correlation. For example, the relationship between poverty and illiteracy can be determined as the impact of one (poverty) over the increase or decrease in the intensity of the other (illiteracy).
There are many techniques of correlation. Pearson/product–moment correlation is the most commonly used technique. Sometimes when there is a requirement of determining the relationship between two variables, after removing the effect of one or more variables, the technique of partial correlation is used, which is an alternative type of Pearson/ product-moment correlation. All statistical techniques work with different types of data, and this is also true in the case of correlation. In other words, it is also applicable to quantifiable data. In quantifiable data, numbers are interpreted from quantities of different sorts through correlation. Correlation does not work on any sort of categorical data, like someone’s favorite color, favorite brand, gender, etc.
The end result of correlation is called the correlation coefficient. It is denoted by “r” and ranges from -1.0 to +1.0. This range of correlation means that the closer r is to +1 or -1, the more intensely the variables are linked. If r is near 0, then the correlation of the two variables tells us that they are not related at all. If r is closer to +1, then the correlation between them states that both of them are directly proportional to each other. In other words, it means that if one variable will expand, the other variable will also expand. In case the correlation between the two variables is -1, the two are inversely proportional to each other. In other words, the correlation means that if one variable expands, then the other will be lessened or become smaller. To make the value of the correlation coefficient easier to understand, the value of the correlation coefficient is squared. That square of the correlation coefficient is equal to the percentage with which the variation of one variable is related to the variation of the other variable. After the correlation coefficient r is squared, the decimal point in it can be ignored. For example, if the value of r is .3, then the square of it will be .9, which means that the correlation coefficient between the two variables is 9%(.3 squared and decimal ignored). If the value of the correlation coefficient is .4, it means that the correlation coefficient between the two variables is 16%(.4 squared and decimal ignored).
While using the correlation technique, we should also keep in mind that it only works on linear relationships and not on curvilinear (where the relationship does not follow a straight line) relationships. For example, in health and financial conditions, the variables are related but not linearly. Therefore, the data will not be quantifiable and therefore cannot be determined by the correlation technique.
