Please enable JavaScript.
Coggle requires JavaScript to display documents.
Statistics: Chapter Fifteen - Coggle Diagram
Statistics: Chapter Fifteen
Correlation: Technique used to measure relationship between two variables.
Usually the two variables are simply observed in nature and not manipulated
Positive Correlation: As one goes in one direction (either increases or decreases), the other variable does the same (as income increases, grades also increase)
Negative Correlation: As one variable goes in one direction, the other does the opposite (as income increases, grades decrease)
Perfect Correlation: Correlation of 1.00; each change indicates a predictable change in the other variable
Pearson Correlation: Measures degree and linear relationship of two variables
r = covariability/variability separate
Positive: Cluster around an upward slope
Negative: Cluster around a downward slop
Multiplying by a negative number does not change value, but it does change the + or - sign
Sum of Products (SP): Measures the amount of covariability between two variables
Each x can be transferred to a z score and the Pearson Correlation can be configured
Why use correlation?
Reliability
Validity
Prediction
Theory Verification
Interpreting Correlations
Value of correlation can vary based on data
Outliers can have a dramatic effect on correlation
Only describes relationship, not cause (Correlation does NOT equal causation)
A perfect correlation (1.00) does allow for predictability but anything less does not indicate a variance in "how accurate" it could be
Hypothesis Testing: The is a correlation or there is not; it is positive or it is negative (Again, correlation does NOT equal causation)
Spearman Correlation: Used with ordinal data
Point-biserial correlation: Used to measure the relationship between two variables in situations in which one variable consists of regular, numerical scores, but the second variable has only two values.
When both variables measured for each individual are dichotomous, the correlation between the two variables is called the phi-coefficient