Lesson 2: Pearson Correlation Coefficient
January 19, 2017
Thu, Jan 19
Review:
- Syllabus
- Least Squares Regression
- Sum of Squares Calculations
Presentation:
- Pearson Correlation Coefficient
- Usually abbreviated as “r”
- Range of possible values is -1 to +1
- Negative correlation means as one variable goes up the other goes down
- Positive correlation means as one variable goes up so does the other
- Coefficient closer to -1 or +1 indicates stronger correlation
- Coefficient closer to 0 indicates weaker correlation
- Use Sum of Squares calculations
- Pearson Correlation Coefficient = SSxy/(√(SSxx*SSyy))
- Demonstrate calculation
- Beer party data: {(60,10), (70,12), (80,20), (90,40)}
- Hand written example (pdf): PearsonCorrelationCoefficientCalculation_prob12-1
Activity:
- Calculate Pearson Correlation Coefficient
- Use calculations from previous class
- Data set 1: {(3,5), (5,8), (8,11), (12,10)}
- Data set 2: {(50,12), (60,15), (70,20), (80,30)}
- Start from scratch with one other data set
- Data set 3: {(2,12), (4,9), (7,6), (11,3)}
- Use calculations from previous class