Home ›
AP Statistics ›
Unit 5 ›
Visual Review
AP Statistics Unit 5 Visual Review
A topic-by-topic visual walkthrough of Unit 5: Regression Analysis — scatterplots, correlation, least-squares regression lines, and residuals.
← Back to Unit 5 hub
TOPIC 5.1
Scatterplots
explanatory (x) →
response (y)
Describe DUFS
DIRECTION: positive / negative.
UNUSUAL: outliers / influential points.
FORM: linear / curved · STRENGTH: tight?
Explanatory vs. response
The EXPLANATORY variable (x) may help
predict the RESPONSE (y).
Association is not causation — see 5.2.
Describe a scatterplot's Direction, Unusual features, Form, Strength .
The Review Hub · AP Statistics Unit 5
TOPIC 5.2
Correlation
r=−1 −0.5 0 +0.5 r=+1
What r tells you
r measures the STRENGTH & DIRECTION of
a LINEAR relationship, −1 ≤ r ≤ 1.
Near ±1 = strong linear; near 0 = weak.
r has NO units and doesn't change if you
swap x and y or rescale.
Cautions
• r only describes LINEAR patterns —
a strong curve can have r ≈ 0.
• r is NOT resistant to outliers.
• CORRELATION ≠ CAUSATION.
Always look at the scatterplot, not just r.
−1 ≤ r ≤ 1 measures LINEAR strength/direction; correlation ≠ causation.
The Review Hub · AP Statistics Unit 5
TOPIC 5.3
Linear Regression Models
ŷ = a + b·x (predicted response)
Interpret slope & intercept
SLOPE b: predicted CHANGE in y for each
1-unit increase in x (with units!).
INTERCEPT a: predicted y when x = 0
(may be meaningless out of range).
ŷ = 3 + 2x → +2 in y per +1 in x
ŷ is a PREDICTION
The hat on ŷ means "predicted," not the
actual observed value.
EXTRAPOLATION (predicting far outside the
x-range) is unreliable.
Only predict within the observed data range.
ŷ = a + b·x : slope b = predicted change in y per unit x; don't extrapolate.
The Review Hub · AP Statistics Unit 5
TOPIC 5.4
Residuals
residual
ŷ line
residual = y − ŷ
Residual plots check the fit
A residual = actual − predicted.
Positive → the point is ABOVE the line.
NO PATTERN in the residual plot → a line
is a good model.
A CURVED pattern → a line is NOT appropriate.
residual = y − ŷ ; a patternless residual plot means a line fits well.
The Review Hub · AP Statistics Unit 5
TOPIC 5.5
Least-Squares Regression
// the least-squares line minimizes Σ(residuals)²
b = r · ( sᵧ / sₓ ) a = ȳ − b·x̄
the line always passes through the point (x̄, ȳ)
The coefficient of determination r²
r² = the PROPORTION of variation in y
explained by the linear model.
r² = 0.81 → "81% of variation in y
is explained by x."
s = standard deviation of residuals
s is the typical PREDICTION ERROR — how
far actual y's fall from the line, on average.
Smaller s = a tighter, better fit.
Read a, b, r, r², s from computer output.
LSRL: b = r·(sᵧ/sₓ), a = ȳ − b·x̄ ; r² = fraction of y's variation explained.
The Review Hub · AP Statistics Unit 5
▤ Show all slides
How to use the visual review
Spend 30 seconds per slide before clicking next. Look at the graph, then ask yourself: "Could I interpret this slope, judge this residual plot, or explain r² from memory?"
Use the fullscreen button () on desktop for the best experience. Use arrow keys to navigate. Tap "Show all slides" to jump around.
This is great for review the night before the exam — fast, visual, and covers every idea you need to recognize in Unit 5.