Chapter 3 Flashcards
Describing Relationtionships
Least Squares Regression Line (LSRL)
ŷ = a + bx
slope (LSRL)
b
y-intercept (LSRL)
a
CDOFS
Context
Direction (+ or -)
Outliers
Form (linear or nonlinear)
Strength (fairly, moderate, strong, weak)
Correlation coefficient (r)
describes the strength of the relationship between two variables (not necessarily linear)
falls between -1 and 1
r
Coefficient of determination (r²)
Describes the strength of the relationship between two variables
STEM r²
STEM: r²% of the variation in (response variable) can be explained by the linear regression model
What happens to r and r² when there are outliers?
affected by outliers (gets smaller/weaker)
falls between 0 and 100%
r squared
STEM: residual: y - ŷ = actual y - predicted y
STEM (positive): the actual (response variable) is greater than the predicted by (residual units)
STEM (negative): change to less than
Standard deviation of the residuals (s)
describes the average residual length
larger s
weaker correlation
STEM: s
The average error when using the line for prediction is ____(s)_ (units).