Unit 3

Question 1

Response

Answer 1

UNIT 3: Measures the outcome, “dependent”

Question 2

Explanatory

Answer 2

UNIT 3: Helps explain the response variable, “independent”

Question 3

Scatterplot

Answer 3

UNIT 3: Displays relationship between two QUANTITATIVE variables, explanatory on x-axis, response on y-axis

Question 4

DFS

Answer 4

UNIT 3: Direction
- positive vs negative

Form
- shape (liner vs nonlinear)

Strength
- Use “r” if you can if not, say things like “fairly weak” “fairly strong” etc.

Question 5

Correlation Coefficient

Answer 5

UNIT 3: R -

Measures direction & strength between OF THE ASSOCIATION between two quantitative variables

-1 ≤ r ≤ 1
-1 -> perfect negative
0 -> no association at all
1 -> perfect positive

WARNINGS:
- not resistant to outliers
- doesn’t guarantee linear, use graph
- doesn’t imply causation

Question 6

Regression Line

Answer 6

UNIT 3: Y = mx + b

“least squares line” “line of best fit”

Question 7

Interpret Slope

Answer 7

UNIT 3: For every 1 (x), (y) increases by (slope) (units)

Don’t forget context!

Question 8

Interpret Y-Int

Answer 8

UNIT 3: When (x) is 0, (y) is (intercept)

Don’t forget context!

Question 9

Coefficient of Variation

Answer 9

UNIT 3: R^2

• “coefficient of determination”
• “Proportion of the variation is (y) can be explained by (x)

Question 10

Residual

Answer 10

UNIT 3: Observed y - predicted y

–> Y - Yhat

• use scatterplot to get “residual plots”
  - x-axis = x-values
  - y-axis = residuals (2nd, lst, RESID)

Residuals above the “line” are positive, below are negative

If there’s a pattern: no line
No pattern: yes line

“A linear model is appropriate because there is no pattern on the residual plot” OR “A linear model is NOT appropriate because there IS pattern on the residual plot”