It is a statistical method to determine the relationship between two or more variables. Its values can lie between +1 and -1. The direction of the relation is determined by sign. If increase in one variable leads to increase in dependent variable, then relation is positive. On the other hand, if increase in one variable leads to decrease in dependent variable, then value will be negative. In case no relation exists between two variables, correlation will be zero.

(i) Uses of Correlation:

1. Decision maker will come to know the nature of relationship between variables and the degree of the relation.

2. Getting a quantitative figure for correlation makes the decision making process, objective.

ADVERTISEMENTS:

3. It helps in forecasting and planning because, changes in variables and its impact can be estimated beforehand.

4. It helps the researcher in identifying such factors which can stabilize the economy.

(ii) Precautions in applying correlation:

1. It is difficult to distinguish between dependent variables and independent variables.

ADVERTISEMENTS:

2. Both the correlated variables are affected by a third variable, which has not been taken into consideration by the researcher.

3. The correlation may be due to chance.

4. Very high degree of correlation between two variables does not necessarily; indicate a cause and effect of relationship between them.

(iii) Types of correlation:

ADVERTISEMENTS:

1. Single and Multiple Correlations:

Only two variables are considered in single correlation, i.e., one independent and another dependent variable. In case of multiple correlations, the relation between more than two variables is judged.

2. Partial and Total correlation:

In the case of partial correlation, relation of two or more variables is considered assuming other variables to be constant. Total correlation is based on all the variables without assuming any variable to be constant.

ADVERTISEMENTS:

3. Linear and Non-linear correlation:

When variation in the values of two variables has a constant ratio, there will be linear correlation between them. In non-linear correlation, the amount of change in one variable does not bear a constant ratio to the amount of change in the other related variable.

(iv) Methods of determining correlation:

(a) Scatter diagram.

ADVERTISEMENTS:

(b) Karl Pearson’s coefficient of correlation

(c) Spearman’s Rank coefficient of correlation.