Chapter 3 Test Review Flashcards
How to interpret Slope:
The slope is __(slope)__ which means that when the __(x)__ increases by 1, the __(y)__ increases by __(slope)__.
EX: The slope is 0.01 which means that when the the GRE score increases by 1 point, the GPA increases by 0.01.
How to interpret Y intercept:
The y-intercept is __(y-int)__ which means that when the __(x context)__ is 0, the __(context)__ is predicted to have a __(y context)__ of __(y-int)__.
EX: The y-intercept is 1.05 which means that wen the GRE score is 0, the student would have been predicted to have GPA score of 1.05.
How to interpret r² otherwise known as the coefficient of determination:
__(r² as a percent)__ of the variation in __(y)__ is explained by the linear relationship of/with __(x)__.
EX: 97.4% of the variation in distance traveled is explained by the linear relationship with the number of rubber bands.
What units do r or r² have?
Trick question they have no units of measurement.
How to interpret a scatter-plot?
Using the D.U.F.S method.
What is the LSRL (least squares regression line) equation?
ŷ = a + bx
a = the y-intercept
b = the slope
What does D stand for in D.U.F.S and how do you find it?
D→Direction
And you find that by determining if the graph is positive or negative. (Negative points down and positive points up)
What does U stand for in D.U.F.S and how do you find it?
U→Unusual/unique features
Basically asking if there are any outliers.
What does F stand for in D.U.F.S and how do you find it?
F→Form
Is the graph linear? exponential? logarithmic?
What does S stand for in D.U.F.S and how do you find it?
S→Strength
What is the correlation coefficient? Closer to 1/-1?
What happens if you have a horizontal outlier?
The slope will decrease, y-intercept increases, and the correlation decreases. A horizontal outlier will tilt the line.
What happens if you have a vertical outlier?
The slope stays the same, y-intercept will increase or decrease depending on where the point is, and the correlation will decrease. A vertical outlier shifts the point either up or down depending on its placement.
Does a horizontal or vertical outlier have the greatest impact on LSRL?
Horizontal outliers.
What is the definition of high leverage point?
High-leverage points are outliers
Influential Points
If they are removed, they substantially change the least squares regression line.
Different formulas for LSRL
ŷ = a + bx
b = r (Sy/Sx)
a = ȳ - bx̄
ȳ = a + bx̄
How to interpret a scatter-plot using D.U.F.S?
There is a __(strength)__ ,
__(pos. or neg.)__ , __(form)__ , relationship between __(context)__ with
__(no/if yes how many and what)__ unusual features.
EX: There is a strong, positive, linear relationship between the number of rubber bands and the distance traveled with no unusual features.
