CHP. 8 - Bivariate Correlation Flashcards
Bivariate Correlation
Correlation between exactly 2 variables
Effect Size Strengths
r = +/- 0.10: small or weak
r = +/- 0.3: medium or moderate
r = +/- 0.5: large or strong
T-test
Statistical test to determine if there is a significant difference between 2 groups
Effect Size
Describes the strength of the association
- how well does 1 variable predict the other?
- larger effect values give more accurate predictions
P Value
Probability estimate
Type 1 vs Type 2 Errors
Type 1: false positive (false alarm)
Type 2: false negative (miss)
Outliers
Extreme scores outside of a normal distribution
- matter most with small sample
Curvilinear Association
Association that isn’t a straight line & r is close to 0
Moderator
When a relationship between 2 variables changes depending on the level of another variable
- shift in the bivariate relationship
- often a (+) to (-) trend or vice versa
- also could be signficant to N.S. or vice versa