Correlation is defined as a numerical coefficient that indicates the extent to which two variables are related.

The coefficient will always be between +1.0 and -1.0. A perfect positive correlation (+1.0) would exist if every subject varied an equal distance from the mean in the same direction (measured by a Z score) on two different variables. For ex, if every subject who was 1A score above the mean on variable X was also 1A score above the mean on variable Y, and every other subject showed a similar relationship between deviation score on X and on Y, the correlation would be +1.0. Similarly, if all subjects who were above/ below the mean on variable X were an equal distance in the opposite direction from the mean on variable Y, the correlation would be -1.0. A value of 0 means no relationship between the variables. Positive correlations results when subjects who receive high score in one variable also receive high scores on another variable. Negative correlations result when scores are high on one variable and low on the other variable (Inverse relationship). Although a correlation coefficient indicates the amount of relationship between two variables, it does not indicate the cause of that relationship. Just because one variable is related to another that does not mean that changes in one will cause changes in the other. Other variables may be acting on one or both of the related variables and affecting them in the same direction. Best fit line- linear estimate of the relationship between two variables. Points are closer to the line if the coefficient is closer to +1.0 or -1.0. Points are on the line if it s +1.0 or -1.0. Points drift away (until there is no line) if the coefficient approaches 0. A 0.0 correlation is also called an orthogonal relationship- one that is not statistically significant. Furthermore, the line has a positive slope for positive correlation and a negative slope for negative correlation. It is the absolute value, not direction (positive/ negative), that determines the strength of the relationship. Pearson s correlation coefficient is a measure of linear relationships. It will not work on curvilinear data. An absolute r value of 0.5 to 0.7 is considered low, 0.7 to 0.8 is moderate, and 0.9 or higher is good for predicting Y from X. Values lower than 0.5 are not statistically significant. If two variables are correlated at r=0.8, then they have 64% common variance (r2). This means that 64% of the variability in the Y variable can be explained by variance in the X variable. (ie. 64% related due to common factors, not chance). The remaining 36% of the variance in Y is unexplained. (ie. Not accounted for by the correlation coefficient). This unexplained variance is responsible for error when predicting Y from X. A significant correlation does not prove causation; it only shows that a nonchance relationship exists. The fact that two variables are related does not mean that a change in one will necessarily produce a corresponding change in the other. For instance, a student s shoe size and vocabulary are highly correlated. In other words, the larger the shoe size, the larger the vocabulary the student has. It is easy to see that shoe size and vocabulary have nothing to do with each other, but they are highly correlated.

The reason is that there is a confounding factor, age. The older the student is, the larger the shoe size and the vocabulary. You cannot compare models by comparing the r values. If N is small, r must be large to be significant. When N is large, small r values may be significant. When the correlation between two variables is sufficiently high, we can predict how an individual will score on variable Y if we know his or her score on variable X. The best fit line is created in such a way that it balances the data points. That is, the total vertical distance from each point below the line up to the line is balanced by the total vertical distance from each point above the line down to the line. The average distance of the points above the line is the same as the average distance of the points below the line. The vertical distance from any point to the line is called a residual (can be positive/ negative).