AP Stats Unit 2

How is 2-Variable data usually displayed?

Two-Way table

Association

If one variable is linked to different distributions between different groups, the two variables are associated

Joint relative frequency

1 cell / entire table

Marginal relative frequency

entire row or column / entire table

Conditional relative frequency

total of 1 cell / its respective row or column

What acronym is appropriate to use when asked to describe scatterplots?

CDOFS:

C- Context

D- Direction

O- Outliers

F- Form 

S- Strength

Correlation coefficient (r)

Quantifies how close data points adhere to a line of best fit / strength of a model

What is the rule about correlation and causation?

Correlation DOES NOT equal causation.

Linear regression equation

ŷ = a+bx

Extrapolation

Using a linear model to predict an x-value outside of the given set

How do you read computer readouts?

Top left = y Top right = a

Bottom left = x Bottom right = b

Residual

How far a given point is from the model’s prediction

Residual plot

Judges the appropriateness of a model.

How do you know a residual plot is favorable vs not?

You want to see a random distribution with no apparent form.

Least Squares Regression Line (LSRL)

Linear equation minimizing the sum of squared residuals.

Coefficient of determination (r²)

Percent of variation in the response variable that can be explained by the explanatory variable.

What ways can you judge if a linear model is appropriate?

YES: r, Residual scatter plot

NO: r²

Outlier

A point that doesn’t follow the general trend, and therefore impacts the model’s strength because of its large residual

High leverage point

A point with a significantly larger x-value than any other point. CAN follow the general trend, but may impact slope or y-int