As with any survey, other descriptive methods are often used in education is the study of correlation. This study examined the relationship of two or more variables, namely the extent to which variation in one variable associated with variations in other variables. The degree of relationship variables declared in a single index called the correlation coefficient. The correlation coefficient can be used to test hypotheses about the relationship between variables or to declare a large-small relationship between two variables.

Correlation study aimed to test the hypothesis, carried out by measuring the number of variables and calculate the correlation coefficient between these variables, in order to determine which variables are correlated. For example, researchers wanted to know which variables are associated with professional competence if the school principal. All variables that had something to do (eg educational background, academic supervision, etc.) are measured, then calculated the correlation coefficient to determine which variable has the strongest correlation with the principal managerial skills.

Strength of the relationship between variables is indicated by a score correlation coefficient varies between -1 to +1. The correlation coefficient is a quantity that is obtained through a statistical calculation based on the collection of measurement data from each variable. Positive correlation coefficient indicates a directly proportional relationship or alignment, a negative correlation coefficient indicates that berbading inverse relationship or non-sejajaran. Score 0 for the correlation coefficient showed no relationship between variables. The bigger the correlation coefficient either positive or negative direction, the greater the strength of the relationship between variables.

For example, there is a positive correlation between IQ variables and academic achievement; implies a high IQ will be followed by a high learning achievement; in other words there are parallels between IQ and academic achievement. Conversely, a negative correlation indicates that a high value on one variable will be followed by a low value on other variables. For example, there is a negative correlation between absenteeism (absenteeism) and academic achievement; implies that high attendance will be followed by a low learning achievement, in other words there are inequalities between attendance and academic achievement.

In a correlational study, there are at least two variables must be measured so that it can be seen to do. In addition it may also analyzed the relationship between the three variables or more.

The meaning of a correlation which is denoted in the letter r (minor) could contain three things. First, the strength of the relationship between variables, second, the statistical significance of relations between the two variables, and the three-way correlation. Strength of the relationship can be seen and the size of the correlation index. Value near zero means the weak link and vice versa value close to one indicate strong relationships.

That enough factors affect the size of the coefficient of correlation is the reliability of instruments used in measurement. Achievement test that is too easy for smart kids and dumb too difficult for children will result in small correlation coefficients. Therefore, instruments that do not have a high reliability will not be able to reveal the degree of meaningful or significant relationship.