AP Stat Chap 3 - Scatterplots, Association, & Correlation

1. Correlation is

1. Only quantitative variables, doesn’t have units, range of -1-1, indicate the strength of a linear relationship between two quantitative variables

2. correlation of zero means

probably a parabola and not linear

3. if outlier is

In the line of best fit, then it isn't an outlier but a high leverage point, as it strengthens the linear relationship

4. regression line

describes how the response variable changes with the explanatory variable

5. regression line equation and other formulas

y hat=a+bx, slope=r sy/sx,

6. residual

actual-minus-predicted value (vertical distance of a point from a line)

7. least squares regression line is

used to find the line of best fit squared, where the sum of the squared residuals is the smallest.

8. residual plot

a scatterplot of the regression residuals against the explanatory variable

9. A residual plot is linear when

scattered everywhere, and a horizontal line

10.interpret r^² (Coefficient of determination)

% of variation in the response variable is explained by the linear relationship with the explanatory variable

11. How to describe a scatterplot

DUFS, An example strong positive linear relationship between x and y variable, no unusual

DUFS

Direction, Unusual, Form, Strength

12. cause and effect from a scatterplot

changes in one variable doesn't necessarily impact the oher.

13. To find actual value

add or subtract y residual value to the predicted value

14. interpert r

“r” implies that the linear relationship between explanatory variable and response variable is strength and direction

interpret slope

For each additional 1 unit of explanatory variable our model predicts an addiontional( or decrease of)__ in the response variable

The LSRL does what?

It minmizes the sum of the squares of the residuals

interpret negative residual

our model overestimates

the standard deviation of residuals interpretation

The average error in the prediction of response varibalbe when using the regression on explanatory variable is __(value of s with units)